We're thrilled to announce that our Spring 2025 seasonal data is now available for the entirety of the United States. This season spans March, April, and May 2025, with data for a typical weekday (Thursday) and weekend day (Saturday) during this time period.

We've improved our route choice algorithm by identifying critical locations where key driving decisions are made (like on and off ramps or major intersections). This enhancement expands the route choice set by combining segments at these key points, providing up to 100 distinct route alternatives per trip. The improved algorithm better captures how drivers navigate around congestion, resulting in more realistic traffic distribution patterns across the network.

We're committed to bringing in the most accurate, up-to-date ground truth data available to ensure that we accurately simulate today’s built environment. We continue to update existing data sources as well as add new ones, and here are some of the updates we’ve made for this season.

We are now using station-level ridership counts as a calibration target to better align modeled transit usage with observed ground truth. Station-level ridership data has been included for the New York City Subway, BART (San Francisco), and PATH (New Jersey/New York) transit systems.

Additionally, as part of our commitment to continue releasing data quickly, we implemented a fast-follow approach for the Spring 2025 season for transit route-level ridership. We estimated ground truth for Spring 2025 by scaling Fall 2024 ridership using changes reported in the National Transit Database. Replica also introduced four new transit agency GTFS feeds: JTRAN (Jackson, MS), Coast Transit Authority (Gulfport, MS), Hub City Transit (Hattiesburg, MS), and METS (Evansville, IN). All other GTFS routes that were newly opened or changed between Fall 2024 and Spring 2025 will not be reflected in the Spring 2025 season; only routes that were running in Fall 2024 will be included. See release notes section below for a list of transit providers used for each region.

Replica’s population data is based on US Census demographic datasets which are used to create a “synthetic population” that is statistically representative of the actual population in a region.

With this season, Replica incorporated the latest releases of the Longitudinal Employer–Household Dynamics (2022) and Census Transportation Planning Products (2017-2021) datasets. These updates ensure that Places reflects the latest available data on where Americans live and travel.

See release notes section below for all Census population data sources utilized in our model. These are the same for all regions.

Auto Counts Data

Replica sources seasonal auto counts from both daily traffic count data and in-vehicle GPS data, scaled to roughly match estimated road capacity during hours of congestion. We then verify the scaled counts against federal AADT data supplied by FHWA as an additional check. The roadway network is based on the OpenStreetMap network version published on March 1, 2025. For more information on OpenStreetMap click here.

This season, Replica has expanded the number of states from which it pulls lower latency auto count data to 11. We’ve also incorporated data from New York City's Congestion Relief Zone to improve geographic coverage and accuracy of roadway volume estimates.

See release notes section for a list of all auto data sources utilized in each region.

Region-Specific Release Notes & Quality Reports

Learn more about the specific ground truth data sources utilized for each region's Spring 2025 model below.

We’ve made it easier to find what you need in Replica. We’re excited to release the following updates:

The tools you rely on are the same, but finding and launching them is now quicker and easier.

Watch the short video below to learn more:

We're thrilled to announce that Places Fall 2024 data is now available for the entirety of the United States. This season spans September, October, and November 2024, with data for a typical weekday (Thursday) and weekend day (Saturday) during this time period.

New Airport Data

We added an additional Bureau of Transportation Statistics dataset that incorporates international visitor flight counts to our airport modeling ground truth data sources.

The Fall 2024 seasonal mobility model contains population, auto, and transit ground truth data from various sources.

As part of our commitment to model travel across the United States, we added 8 additional transit agencies, and their respective GTFS feeds, to the Fall 2024 seasonal model. Here is a list of all the transit service providers and the routes modeled in the Fall 2024 model.

Replica sources seasonal auto counts from both daily traffic count data and in-vehicle GPS data, scaled to roughly match estimated road capacity during hours of congestion. We then verify the scaled counts against federal AADT data supplied by FHWA as an additional check. The roadway network is based on the OpenStreetMap network version published on September 1, 2024. For more information on OpenStreetMap click here.

We are thrilled to announce the release of our Spring 2024 seasonal data, spanning a typical weekday and weekend day from March, April, and May of 2024.

What's new with this season?

We are thrilled to announce the release of our Fall 2023 seasonal data, spanning a typical weekday and weekend day from September, October, and November of 2023.

What's new with this season?

Last week, we introduced the highly requested feature that lets you create custom geographies directly in the Replica platform enabling you to organize Replica data in ways that make the most sense for you. The list of ways to create your geographies continues to grow! Here's what you can now do directly in the Replica platform:

Buffer around a line (NEW): You can draw either a single segment line or a multi-segment line and add a buffer.

Manually draw a geography (NEW): You can draw a polygon shape and add a buffer.

In addition to drawing directly in the platform, you can still upload custom geographies in shapefile, KML, and GeoJSON formats. We're committed to enhancing our platform to meet your needs, and we look forward to bringing you more exciting features in the future. Stay tuned for more key features that will further enhance your custom geography creation.

You can now create custom geographies directly within the Replica platform. This enhancement not only simplifies the process but also grants you flexibility to shape, customize, and visualize your data. Now, in addition to uploading custom geographies as shapefiles, KML, and GeoJSON files, you can create directly in the platform using the following methods:

Stay tuned for the coming weeks as we roll out additional key features that will further enhance your custom geography creation, like the ability to add a buffer around a line, the ability to manually draw a geography on the map, and the ability to create custom geographies composed of multiple shapes.

Trip origin, trip destination, home locations and work locations are now available for download at the TAZ and zipcode level in the dataset tab. Click on the “Manage Attributes” button in the panel to add these fields to your dataset downloads.

We’re thrilled to announce that seasonal detailed network link volume data is now available for download directly from the Data Downloads page in the Replica platform.

You now have the ability to download trip volumes per network link by trip distance, mode, purpose, or start hour for a given geographic region. Data is available to download in .csv, GeoJSON, or shapefile formats.

We're thrilled to announce that we now have Places data for the state of Alaska. The new megaregion includes data for Spring 2023 and ~640,000 people.

We’ve released new features in Studies that enable you to see more detailed insights about transit travel in your region:

At Replica, we're passionate about equipping public agencies with the tools they need to thrive, especially in the ever-evolving world of public transit. We're pleased to offer Replica Transit Scores to empower public agencies in adapting to today's shifting landscape of public transit.

Transit Demand Score: Answers questions about where demand for service is highest and where it would be most utilized if made available.

Transit Equity Score: Answers questions about the existing public transit system and how it serves the portion of the community dependent on transit.

You can now access or request access to Replica Transit Scores directly from the Replica platform. Simply visit the Applications page and click on the "Transit Equity and Demand Scores".

We're thrilled to offer additional demographic insights in the summary panel for your studies. The new summary modules are available in the "People" tab and include:

These attributes are also available as filters, meaning you can filter your study to look at trip-takers who have specific primary languages, education levels, household tenures, or resident building types.

We are also introducing a new feature that allows you to manage which summaries are visible within the summary panel for your study. Simply click the “Manage Charts” button to view all available summaries with brief descriptions and toggle them on/off.

You can now find Reference Downloads (Land Use, Census Geographies, Network Links, and Transit Routes & Stops) on the Data Downloads page.

The content of the downloads has not changed. You can now see a preview of the Land Use data before downloading and can easily refer to earlier Land Use and Reference Dataset downloads on the Download History page.

Each dataset now has brief descriptions and links to relevant documentation and metadata like last updated data and frequency of updates.

The Hourly Auto Volume Profile, which is created through the same underlying model as our Annual Average Daily Traffic (AADT), allows you to see how volumes shift over the course of a week.This dataset contains information about the average auto volume per network link, provided in hourly buckets for a typical week. For Functional Road Classification (FRC) 1 and 2 roads, we also provide hourly volumes for single-unit and combination trucks. This data is available for 2021 and 2022, and is updated annually. Data is available for download in CSV, Shapefile, and Geojson format. Read our documentation for Hourly Auto Volume Profile here.

If your subscription provides access to Replica’s AADT (Annual Average Daily Traffic) data, you’ll automatically gain access to the Hourly Auto Volume profile. Otherwise, reach out through our in-app chat if you are interested in activating access to this dataset.

We’ve heard your feedback and are excited to announce that we made several updates to our Traffic datasets.

Use the Trip Intersects filter to identify all trips that started, or ended, or passed through your selected geographic area. This filter is available for all seasons.

Places Spring 2023 data is now available for the entirety of the contiguous United States and Hawaii. This season spans March, April, and May 2023, with data for a typical weekday (Thursday) and weekend day (Saturday) during this time period. For the first time, Replica is providing customers with the granular detail of Places data, with just a 3-month lag.

With the release of Spring 2023 data, Replica is for the first time including information on Vehicle Fuel Type as part of its trip data. All private auto trips are now categorized as "Battery Electric Vehicle (BEV)" or "Other Non-BEV." The Other Non-BEV category includes all non-BEV vehicles, such as internal combustion engine, plug-in hybrid, and conventional hybrid.

With this update, customers can filter and analyze trips made only by electric vehicles (or non-electric vehicles). You can study the transportation patterns of EV trip-takers by viewing Trip origins, destinations, and routes, as well as Home and School/Work locations.

You can read about our vehicle assignment methodology here. The Vehicle Fuel Type attribute only applies to private auto modes -- public transit and commercial trips are not included. We also currently do not explicitly model any differences in travel behavior between those using a BEV and non-BEV. Any differences in observed metrics are due to correlations between BEV ownership and other sociodemographic factors.

You can now filter your population of interest by 6 additional attributes, allowing rich analysis about the relationship between these attributes and travel patterns. These filters are available for all seasons, including Spring 2023.

As part of our commitment to continue releasing data quickly, we implemented a fast follow approach for the Spring 2023 season. We estimated line-level Spring 2023 transit ground truth by scaling actual line level ridership from Fall 2022. Scaling factors were derived based on the change between Fall 2022 and Spring 2023 ridership values reported in the National Transit Database.

Replica introduced two new transit agency GTFS feeds for the Spring 2023 season, Greater Glens Falls Transit, NY, and Collier Area Transit, FL, and one new stop, Grand Central Madison LIRR, NY. All other GTFS routes that were newly opened or changed between Fall 2022 and Spring 2023 will not be reflected in the Spring 2023 season; only routes that were running in Fall 2022 will be included.

Additional attributes are now available for download and preview. Education, language, tenure, industry, school grade attending, and building type are now included to provide additional demographic detail about the people of interest.

When downloading the Network Link Volume map layer, you will now select your geographic area of interest. Previously, you would download the top 100k links in the megaregion.

The download workflow for map layers and the trips/people dataset has been updated. After initiating the download, you will receive a notification in the application when the download is ready, and download your file directly from the notification. Users will no longer receive Network Link Volume downloads by email.

The “Trips Passing Through” filter is now significantly faster to apply and includes transit trips that pass through the selected area. Previously, the filter only included trips that traversed major roads in the geography area and therefore did not include rail or subway transit trips. Now, all trips that pass through the area are included.

Note: Because we’ve updated the underlying logic of the “Trips Passing Through” filter in existing studies, you may notice a difference in the number of trips and trip-takers.

ID values for the fields person_id, activity_id, and household_id are no longer truncated and filled with zeroes in the Replica Application and now match the ID values in BigQuery. This allows users to identify people or trips of interest in the Replica Application and continue analyzing them in BigQuery.

You can now view the primary commute mode of people who work from home as “worked_from_home” instead of the previously null value in the People dataset preview and downloads.

Note: The “Work from Home” attribute reflects whether a person actually had a work from home activity on the modeled day. A subset of people with “worked_from_home” primary commute mode will actually have a work from home activity on the modeled day.

All out of region labels have been standardized as “Out of region.”

Previously, locations that were outside of the megaregion were listed as a combination of “Outside of region” and “Out of region” in Dataset and Summary values for trip start and trip end.

Core residents without a Work/School location now have the Work/School location values “Does not have work/school location” in dataset values, rather than null values. All Work/School location geo reference columns will also contain “Does not have work/school location” in this case.

Residents of the buffer region now have the Home location value “Buffer region” rather than “Out of region” in dataset values. All home location geo reference columns including “Trip Taker Home Location Blockgroup”, “Trip Taker Home Location State”, and more also have the value “Buffer region.”

Visitors now have the Home Location value “Visitor (no home location)” and “Visitor (no work/school location)” rather than null values. All Home Location and Work/School location geo reference columns including “Work/School Tract FIPS”, “Work/School Block Group Centroid Longitude” and more will now contain these values for visitors.

Downloads from the Origin-Destination details modal now download a file from the selected table. Previously, the download was a zip file with all three tables.

The Origin-Destination summary and Origin-Destination Details is updated to exclude the following trips:

For core residents without a Work/School location, the Work/School location summary will now show “None” as a shorthand for “Does not have work/school location”.

For residents of the buffer region, the Home Location summary will show “Buffer region” rather than out of region.

For visitors to the region, the Home Location and Work/School Location summary will show “Visitor” as a shorthand for “Visitor (no home location)” and “Visitor (no work/school location)”

We now show percentages of the total trip number, transit trip number (for transit alighting/boarding and transit routes), or number of people in the summary charts. Previously, we showed percentages of the sum of the top 20 values.

Correction of a bug in Land Use that was leading to labels of “Unknown” when we did have information about Land Use, in blocks whose Block ID changed between the 2010 and 2020 census boundaries.

Note: Because this bug has been corrected, there may be slight differences in values for Trip Origin and Destination Land Use and Building Use.

Replica has updated its data querying layer, which serves data to the Replica Application. Due to differences in loading and querying data, you may see small differences in the numbers (e.g. of trip and trip-takers) provided in the Replica Application. These differences should generally be less than 0.1%.

Our Land Use data has been updated to reflect Spring 2023 and includes approximately 500,000 more parcels that have land use information. This important update paves the way for the upcoming Spring 2023 Places season release. You can find the updated Spring 2023 Land use data in the "Land Use" tab in Trends, the Places Study “Lane Use” reference map layer, and in Reference Downloads.

View only filtered volumes on the map

On the Network Link Volume map layer, you now have the choice to either view trip volumes on all network links or to view trip volumes only on the network links you’ve filtered to.

Network link volumes downloads

Previously, you had the option to download volumes for either the filtered links, or all links in a dropdown menu. Now, the dropdown menu automatically adjusts to match the selected viewing option. This simplifies the download process and lets you easily download the volumes you need based on your selected view.

View and copy Stable edge IDs

We have made it easier for you to view and copy the stableEdgeId (link_id) in both the Network Link Volume map layer and the Network Link Filter. Now, when you hover over an edge, the stableEdgeId will be displayed, and you can simply Ctrl-C to copy it.

Confirm and refine your map filters

Now, when you use the map to select filters, you can refine and confirm your selections. After making an initial selection on the map, an overlay will appear where you can further modify your choices and then confirm them. Specifically, for the network link volume layer, you can select your desired filters on the map and choose between "Trips passing through all links" or "Trips passing through any links.” For geographies (i.e., “Trips by Origin”), you can modify your geography selection and geography breakdown before saving your filters.

We are pleased to announce that TMC data is now available for download through the Data Downloads page.

Replica's TMC table contains information about motor vehicle trip counts at most signalized intersections nationwide for each day of the week, bucketed into 1-hour intervals.

Data is available for Spring 2022 and Fall 2022 and is available for download in CSV format. TMC data will be updated annually, with data from the Fall season.

Learn more

Click here to view the methodology, quality validations, and data schema of this dataset.

Need access?

Reach out through the in-app chat (located on the bottom right corner) to upgrade your Replica subscription for access to TMC data.

We are thrilled to announce that the Fall 2019 seasons data refresh is complete for all megaregions. The refreshed data will enable better base-year to present-day comparison.

The refreshed Fall 2019 seasons include::

When re-opening old 2019 studies, you will notice that the data and results have changed, reflecting all updates listed above that impact fall 2019 runs. In addition, our unique identifiers for road segments (stableEdgeIds) have been updated in the Fall 2019 and Fall 2022 seasons to improve comparability and consistency. To ensure that existing Fall 2019 studies with network link filters are usable with the updated data, we have migrated the network link filters in these studies to use the updated stableEdgeIds.

However, ~19% of the stable edge IDs that appear in both Fall 2019 and Fall 2022 represent different geometries. While we strive to maintain comparability across seasons, when performing network-link analysis across seasons, we recommend you use the relationship table provided by Replica in BQ to map Fall 2019 stable edge IDs to Fall 2022 stable edge IDs and visually compare link IDs across years before beginning your analysis.

This migration is done on a best-effort basis. When re-opening an existing Fall 2019 study with a network link filter, make sure to inspect the network link filter to make sure it's capturing the network links that you want it to.

If you have any questions or concerns, please contact support@replicahq.com.

We’re thrilled to announce that Replica’s nationwide annual speeds data is now available for download. Visit the Data Downloads page to download the data today.

Our annual speed data includes two datasets:

Both datasets are available nationwide and include data for 2021 and 2022. We will update annually moving forward.

Learn more

Click here to view the methodology, quality validations, and data schema of these datasets.

Don’t have access?

Contact support@replicahq.com or reach out through the in-app chat to upgrade your Replica subscription for access to our annual speeds data.

We are happy to announce that Fall 2019 refreshes are underway and that North Central and Cal-Nev's new seasons are now available. The refreshed data will enable better base-year to present-day comparison. The refreshed Fall 2019 season includes:

When re-opening old 2019 studies, you will notice that the data and results have changed, reflecting all updates listed above that impact fall 2019 runs. In addition, our unique identifiers for road segments (stableEdgeIds) have been updated in the Fall 2019 and Fall 2022 seasons to improve comparability and consistency. To ensure that existing Fall 2019 studies with network link filters are usable with the updated data, we have migrated the network link filters in these studies to use the updated stableEdgeIds.

However, ~19% of the stable edge IDs that appear in both Fall 2019 and Fall 2022 represent different geometries. While we strive to maintain comparability across seasons, when performing network-link analysis across seasons, we recommend you use the relationship table provided by Replica in BQ to map Fall 2019 stable edge IDs to Fall 2022 stable edge IDs and visually compare link IDs across years before beginning your analysis.

This migration is done on a best-effort basis. When re-opening an existing Fall 2019 study with a network link filter, make sure to inspect the network link filter to make sure it's capturing the network links that you want it to.

If you have any questions or concerns, please contact support@replicahq.com.

At Replica, we are committed to providing you with the most accessible, valuable and actionable transportation data available. That's why we've updated our AADT model to include:

In addition to new data for 2022, we have also re-published 2019 and 2021 AADT data using the new modeling methodology for higher-quality longitudinal and pre-/post-covid comparisons.

With the change to our modeling methodology, the 2019 and 2021 AADT values for a given network link have changed. For 2019, links that are FRC 1-3 with AADT above 5000 have an average percent change of 2.8%. For 2021, links that are FRC 1-3 with AADT above 5000, the average percent change is 1.3%.

Need access?

Contact support@replicahq.com to upgrade your Replica subscription for access to AADT data for all available years (2019, 2021, and 2022).

Learn more about this dataset:

All Fall 2022 seasons are now available with the release of the North Atlantic (CT, DE, NJ, NY, PA) megaregion. These new seasons include:

Notes on improvements specific to the North Atlantic megaregion:

Additional resources:

Fall 2022 Places data is now available for the Hawaii (HI), Northwest (ID, MT, OR, WA, WY) and Southwest (AZ, CO, NM, OK, TX, UT) megaregions. These new seasons include:

Additional resources:

See below a list of available seasons:

Fall 2022 Places data is now available for the Great Lakes (IL, IN, KY, MI, OH, WI) and Mid-Atlantic (DC, MD, NC, VA, WV) megaregions. These new seasons include:

Additional resources:

See below a list of available seasons:

Fall 2022 Places data is now available for the South Atlantic (FL, GA, SC) megaregion. This new season includes:

Fall 2022 Places data is now available for the North Central (IA, KS, MN, MO, ND, NE, SD) megaregion. This new season includes:

North Central megaregion specific note related to the Quality Report:

Additional resources:

Fall 2022 Places data is now available for the Cal-Nev (CA,NV) megaregion. This new season includes:

Cal-Nev megaregion specific note related to the Saturday Quality Report:

Additional resources:

Fall 2022 Places data is now available for the Northeast (MA, ME, NH, RI, VT) and South Central (AL, AR, LA, MS, TN) megaregions. These new seasons include:

Additional resources:

We are thrilled to unveil two more exciting enhancements that improve your experience viewing network link data in Studies. With these enhancements, you will have greater control and flexibility to access the data you need. You can now:

When moving around the Network Link Volume map layer, you now control when the map layer updates. Previously, the layer would reload when you moved to a new area of interest on the Network Link Volume map layer after panning or zooming. With this change, you can retain the context of the loaded Network Link Volumes while choosing the area you want to see and simply click the "Update Network Link Volumes" button when you're ready to refresh the link volumes.

We have increased our network link volume download available in the layer panel from the top 20,000 network links to the top 100,000 network links. With this enhancement, your download will be e-mailed to you when ready.

We are thrilled to announce two major enhancements that improve your experience using network links in Studies. These enhancements add new functionality to our platform, enabling you to do highly requested analyses that weren’t possible before. Based on your valuable feedback, you now can:

Previously, the network link filter enabled you to filter trips that passed through any of the selected network links. Now, you can also filter trips that pass through all selected network links. For example, this could be useful to see which trips passed through the Hudson Tunnel from New Jersey to Manhattan and then the Midtown Tunnel from Manhattan to Queens. You can select this option by selecting "All selected links (and)" in the Trips Passing Through dropdown when creating a Network Links filter.

When downloading trip volumes in the network link volume layer, you now have an additional download option that will only show network links volumes for the network links you have filtered. Previously, when downloading the network link volumes, the top 20,000 links by volume in the megaregion were downloaded. Downloads are available in CSV, GeoJSON, and Shapefile formats.

As of today, February 2nd, we have integrated a new data provider into our Trends mobility pipeline, effectively ending our maintenance period.

As we have shared over the past month, one of our data providers fell short of our standards, so we have integrated a new data provider to address this. While we are committed to publishing high-quality data that our customers can trust and use, our data resiliency fell short in the past few weeks.

We have republished mobility data that was previously released in January 2023, including data for the weeks of:

Mobility data prior to January 2023 remains the same.

With our January 2023 backfill, we have observed a shift in absolute numbers in mobility metrics. In particular, we have observed a downward shift in private auto mode share at the nationwide-level.

In order to ensure data quality moving forward, we are updating our release schedule for mobility data only moving forward.

We will release Trends mobility data from two weeks ago instead of from the previous week. Next Thursday, we will release Trends mobility data for the week of January 21. Spend data will remain on the same schedule.

Please note that with this change, our weekly mobility data release will reflect data that is a week behind our spend data release.

After additional testing and analysis, we will not resume publishing the following mobility metrics: trip purpose, working-from-home, and intra-geo trips.

We’re excited to announce a more seamless Trends download experience! We upgraded our Trends download page for easier access to Trends mobility and spend data. Below are the notable changes:

We will continue to improve the download experience for our Places product and additional datasets in future releases expected sometime later this year.

Please note that you may see small differences in estimates with this new download experience. We’re in the process of improving our backend data infrastructure and moving to a store that is more precise. This will result in a change in precision with regards to trip and spend estimates. These differences are more likely to occur at higher geographic granularity.

We’re also bringing you higher-quality spend data with a new spend data pipeline. We ingested disaggregate spend data into our modeling pipeline to produce higher-confidence spend estimates across all spend categories. With this data refresh, you’re less likely to see unexplainable variations in the data and have more accurate spend estimates in smaller, less-populated geographies. We were also able to more closely calibrate to nationwide e-commerce Census data for our at-home spend estimates.

This new pipeline will allow us to develop more granular spend data estimates in the future, broken down by origins and destination and more granular categories of spend. Please note that with this update, we are also updating the timeframe of our spend data to match our mobility data releases. Each week of data will reflect spend estimates from Saturday to Friday of the prior week.

Replica is excited to release Scenario, a scenario planning tool that can be used to model the impact of population and employment changes on travel demand and the transportation network nationwide. Scenario is available in two ways: you can access our standard growth scenarios which contain simple data forecasts based on historical growth rates, or you can request a custom scenario.

Scenario harnesses the Replica Places modeling pipeline (for more information on Places data and methodology, click here) to produce baseline conditions for your selected geography or a ‘Base Year’ run. This ensures that any observed changes are due to the Scenario input data by specifying the base season, day of week, and precise geographies used for the Scenario runs.

For custom scenarios, you can then set growth projections for a future year. Users can input the following projections:

The model is then run for your scenario and you can expect the following data outputs forecasted for your desired future year:

The data outputs will be summarized and available to users via Hex. Please note that Scenarios will be run for a random 25% sample of the population (scaled up to 100% in the data outputs) to reduce computation time and complexity. This is a large enough sample for analysis at any geographic level. For an extended version of our Scenario methodology, click here.

How can I access Scenario?

To access Scenario, click on the “Scenario” tab on the Replica homepage. From there, you can access Standard Growth Scenarios (when available) or request a new Custom Scenario.

In the standard growth scenarios, what data was used to determine the growth rates?

Standard Growth Scenarios are built off of a [2022] Fall Thursday base year, with population and employment projections based off of the 2012-2022 average growth rate for each county, applied forward to 2022-2030. We source this data from:

Standard Growth Scenarios include the following additional assumptions:

How can I model a custom scenario?

On the Scenario page, click “Request a Custom Scenario” and include a brief description of the type of Scenario you would like to run. Alternatively, you can email [ ]. We will reach out to connect you with someone from our Solutions team to help create your custom scenario.

What are the data requirements for a custom scenario?

The following data is required for a Custom Scenario:

When requesting a Custom Scenario, we will share more details around data requirements, schemas, and example files.

What is the turnaround time for a custom scenario?

Turnaround time for a Custom Scenario may vary depending on the complexity and number of Scenarios run, as well as overall computational/staff availability for Replica. We currently expect turnaround times to take roughly ~1 week from the time input data is finalized. Scenario is a new product and we will update this estimate as we run more of them.

Can I model network changes such as new transit service or a road closure in Scenario?

Currently, you cannot model network changes in Scenario. We plan to release our first network change feature, road closures, in the first half of 2022. Additional network changes are on our future roadmap, and we welcome any feedback as to which would be valuable for you.

What format is the output data available in?

Standard Growth Scenarios are available in Studio, with all of the same download options as standard megaregions including .csv, .geojson, and .shp downloads.

Custom Scenarios will be delivered in a custom report that includes interactive map modules and CSV downloads for all data tables (e.g. population changes, ODs, trip mode, road/transit line volumes). It will be provided as a managed service, so we can also provide additional data outputs/queries to meet customer needs.

Can I access the full trip and population tables for my Scenario run?

For Standard Growth Scenarios, customers can access the full trip and population tables.

For Custom Scenarios, by default we will simulate a random 25% of the population in order to speed up processing and reduce computational demand. This is a large enough sample so that any output data (e.g. ODs, network link volumes, transit volumes) will be accurate at any spatial level. However, it means that the full trip and population tables will be only ¼ of the full size. If your analysis requires the full trip and population tables, we can discuss running 100% of the population [for an additional fee].

This month, Trends mobility data is undergoing maintenance which is resulting in the removal of three metrics as we test their quality and reliability: Trip Purpose, Working-from-Home, and Intra-Geo Trips. As of today, these metrics have been removed from Trends Dashboards and Downloads and will remain unavailable until we have completed further testing.

Replica is committed to publishing high-quality data that our customers can trust and use to inform decision-making. We will continue to inform you about important changes to Trends mobility data during this maintenance period. Please reply to this email if you have any questions.

Note: This maintenance does not impact Trends spend data.

We’re excited to release nationwide travel, demographic, and land use data for the Fall 2021 season! This new season of disaggregate data is available for all megaregions in our Places product. With this release, we made several notable upgrades to our modeling pipeline. Below is a detailed list of the data improvements we made with this new season:

What ground truth data sources were used to model the Fall 2021 season?

Ground truth data sources, including population, transit, and auto ground truth sources, for each season are included in the megaregion-specific release notes. You can find our megaregion release notes in our Help Center here or in your Places study by clicking on the selected megaregion on the filter bar on the top of page and the “View Data Source Details” option (see below).

What year are the Census boundaries used in the Fall 2021 season?

The Census boundaries for all Replica Places seasons, including the Fall 2021 season, are sourced from the 2010 decennial census.

Is there data for previous seasons for the Hawaii megaregion?

Currently we do not plan to backfill data for previous seasons for the Hawaii megaregion.

Replica is excited to debut a new look! We’ve upgraded our platform to allow for quicker and easier access to our data. With this redesign, we're introducing new features including:

And lastly, we refreshed our Reference Downloads where you can find datasets that are complementary to our trip and population estimates. Reference Downloads include Geographic Data (Census Geographies, Network Links, Transit Routes) and aggregate Land Use Data.

Watch below for a quick overview of the updated platform or log in to studio.replicahq.com to try out the new experience.

Starting the week of Monday May 9th, 2022, we plan to publish a full refresh of our historical Trends mobility data. This refresh incorporates our new mobility data pipeline and you can now perform more accurate Trends analyses, with higher certainty in trip counts, mode split (including the addition of walking and biking modes), and trip purpose.

In May, we published new mobility data beginning from January 2021 through the first week of May 2022. As of June 9th, we published 2019 and 2020 with the new data pipeline.

Our Trends data is now generated by a full activity-based model, run for the entire country for a typical weekday and weekend day each week (roughly 1 billion daily trips). These weekly models are calibrated to existing Places models, and validated with data from the National Household Travel Survey (NHTS), Federal Highway Administration (FHWA), and National Transit Database (NTD). This is a significant change from our previous modeling pipeline, which used aggregate data instead of modeling individual trips. This change will substantially improve the quality and resilience of our mobility metrics, and make Trends data more comparable to Places models of the same region.

Users can also capture Trends mobility estimates with higher accuracy for their custom geographies. Previously, Trends estimates were produced at a tract level, and data was aggregated to custom geographies based on tract centroids. Now, mobility data is aggregated to custom geographies based on the specific latitude/longitude of each trip start and endpoint (or resident home location). Similar to Places, each polygon has to contain either 100 residents or 100 employees (500 residents/employees for best results). Spend data is still aggregated to custom geographies based on the tract (for spend by home location) or block group (for spend by merchant location) centroids.

Finally, our new pipeline integrates new mobile location data from Cuebiq, to supplement additional location data (e.g. from vehicles and POIs). This increases overall coverage and provides additional signal on non-private auto trips, such as walking and biking.

With this data refresh we also introduced new metrics:

Please note that Trends mobility estimates do not currently include freight trips or trips made by visitors. We expect to add these to our mobility estimates in the future.

How do you calibrate the new pipeline?

Since we aren’t able to collect ground-truth data from public agencies at the speed we release Trends metrics (weekly), we calibrate our models internally with data from different sources – for example, comparing mobile location data to in-vehicle GPS data. We also compare our Trends data to previously-released, calibrated Places models and to data from the National Household Travel Survey (NHTS), Federal Highway Administration (FHWA), and National Transit Database (NTD).

When will 2019 and 2020 mobility data be available?

As of June 9th, this data is now available.

Does this change affect non-mobility Trends data?

Our new modeling pipeline currently only applies to our mobility estimates. This data refresh does not affect our Trends Spend data. Spend data for all years from 2019 to 2022 will continue to remain available on the platform.

How does this new mobility modeling pipeline differ from the previous one?

Our Trends mobility estimates are now generated by running a full activity-based model for the entire country for a typical weekday and weekend day each week (roughly 1 billion daily trips). This is a significant change from our previous modeling pipeline, which aggregated data and assigned trips at the tract level based on common origin-destination pairs, instead of modeling individual trips.

Can I compare new Trends data to data that I downloaded previously?

Feel free to look at the old and new Trends data to see how the mobility estimates changed in your area. However, because the datasets are generated through entirely different modeling pipelines using different data inputs, we do not suggest using them together or in direct comparisons on projects that you are working on.

How do you model Work-From Home?

We model the number of people that work from home based on two data types: mobile location data, and surveys conducted by the Bureau of Labor Statistics (BLS). The BLS survey provides a lagging monthly estimate of the nationwide share of employed people working from home by industry of employment (by NAICS code). We use mobile location data – specifically, the number of hours observed devices spend at home and work locations – to:

Using the above estimate, we randomly assign each person in our synthetic population with the appropriate home location and industry of employment to a “Work-From-Home” persona. These people will spend a period of time working from home, and will not make a commuting trip to their place of employment. They may still make trips for other purposes throughout the day.

We also model baseline/non-COVID-19 specific Work-From-Home behavior based on the Census American Community Survey (ACS). This estimate is static over time, but varies by geography (PUMA), income, employment, and household role. These synthetic individuals (approx. 5% of the employed population nationwide) are the only ones that worked from home prior to March 2020, and continue working from home throughout the pandemic.

How are you capturing workers who are working from a location that is not their home or their work location?

We do not explicitly label employees that work from a third location outside of their home or typical place of employment. We use the number of hours a person is at home and work-like locations (i.e. places they regularly spend daytime hours) combined with BLS data to estimate the proportion of all employed people working from home in each geography and industry. From there, we randomly assign employees to be “WFH” until that proportion is met.

You now have access to improved Fall 2019 seasons for the following megaregions in Places:

Updates include:

You now have access to new Fall 2019 and Spring 2021 seasons for the North Atlantic megaregion (CT, DE, NJ, NY, PA) in Places. Both seasons include data for a typical weekday (Thursday) and weekend day (Saturday) and ~46M people.

North Atlantic Resources:

As a reminder, all new Fall 2019 and Spring 2021 seasons include:

You now have access to new Fall 2019 and Spring 2021 seasons for Northeast and South Central megaregions in Places. Both seasons and regions include data for a typical weekday (Thursday) and weekend day (Saturday). Learn more below.

Northeast (MA, ME, NH, RI, VT) Overview

Northeast includes coverage of MA, ME, NH, RI, VT, and ~11M people.

Northeast Resources:

South Central (AL, AR, LA, MS, TN) Overview

South Central includes AL, AR, LA, MS, TN, and ~23M people.

South Central Resources:

Starting the week of Monday May 9th, 2022, we plan to publish a full refresh of our historical Trends mobility data. This refresh incorporates our new mobility data pipeline and you can now perform more accurate Trends analyses, with higher certainty in trip counts, mode split (including the addition of walking and biking modes), and trip purpose.

We will first publish new mobility data beginning from January 2021 through the first week of May 2022. We will remove existing 2020 mobility data from Trends as we work to backfill 2020 and 2019 data with our new data pipeline.

Our Trends data is now generated by a full activity-based model, run for the entire country for a typical weekday and weekend day each week (roughly 1 billion daily trips). These weekly models are calibrated to existing Places models, and validated with data from the National Household Travel Survey (NHTS), Federal Highway Administration (FHWA), and National Transit Database (NTD). This is a significant change from our previous modeling pipeline, which used aggregate data instead of modeling individual trips. This change will substantially improve the quality and resilience of our mobility metrics, and make Trends data more comparable to Places models of the same region.

Users can also capture Trends mobility estimates with higher accuracy for their custom geographies. Previously, Trends estimates were produced at a tract level, and data was aggregated to custom geographies based on tract centroids. Now, mobility data is aggregated to custom geographies based on the specific latitude/longitude of each trip start and endpoint (or resident home location). Similar to Places, each polygon has to contain either 100 residents or 100 employees (500 residents/employees for best results). Spend data is still aggregated to custom geographies based on the tract (for spend by home location) or block group (for spend by merchant location) centroids.

Finally, our new pipeline integrates new mobile location data from Cuebiq, to supplement additional location data (e.g. from vehicles and POIs). This increases overall coverage and provides additional signal on non-private auto trips, such as walking and biking.

With this data refresh we also introduced new metrics:

Please note that Trends mobility estimates do not currently include freight trips or trips made by visitors. We expect to add these to our mobility estimates in the future.

How do you calibrate the new pipeline?

Since we aren’t able to collect ground-truth data from public agencies at the speed we release Trends metrics (weekly), we calibrate our models internally with data from different sources – for example, comparing mobile location data to in-vehicle GPS data. We also compare our Trends data to previously-released, calibrated Places models and to data from the National Household Travel Survey (NHTS), Federal Highway Administration (FHWA), and National Transit Database (NTD).

When will 2019 and 2020 mobility data be available?

We will backfill 2019 and 2020 mobility data, including our new metric over the next few weeks and will share more details as we firm up release dates.

Does this change affect non-mobility Trends data?

Our new modeling pipeline currently only applies to our mobility estimates. This data refresh does not affect our Trends Spend data. Spend data for all years from 2019 to 2022 will continue to remain available on the platform.

How does this new mobility modeling pipeline differ from the previous one?

Our Trends mobility estimates are now generated by running a full activity-based model for the entire country for a typical weekday and weekend day each week (roughly 1 billion daily trips). This is a significant change from our previous modeling pipeline, which aggregated data and assigned trips at the tract level based on common origin-destination pairs, instead of modeling individual trips.

Can I compare new Trends data to data that I downloaded previously?

Feel free to look at the old and new Trends data to see how the mobility estimates changed in your area. However, because the datasets are generated through entirely different modeling pipelines using different data inputs, we do not suggest using them together or in direct comparisons on projects that you are working on.

How do you model Work-From Home?

We model the number of people that work from home based on two data types: mobile location data, and surveys conducted by the Bureau of Labor Statistics (BLS). The BLS survey provides a lagging monthly estimate of the nationwide share of employed people working from home by industry of employment (by NAICS code). We use mobile location data – specifically, the number of hours observed devices spend at home and work locations – to:

Using the above estimate, we randomly assign each person in our synthetic population with the appropriate home location and industry of employment to a “Work-From-Home” persona. These people will spend a period of time working from home, and will not make a commuting trip to their place of employment. They may still make trips for other purposes throughout the day.

We also model baseline/non-COVID-19 specific Work-From-Home behavior based on the Census American Community Survey (ACS). This estimate is static over time, but varies by geography (PUMA), income, employment, and household role. These synthetic individuals (approx. 5% of the employed population nationwide) are the only ones that worked from home prior to March 2020, and continue working from home throughout the pandemic.

How are you capturing workers who are working from a location that is not their home or their work location?

We do not explicitly label employees that work from a third location outside of their home or typical place of employment. We use the number of hours a person is at home and work-like locations (i.e. places they regularly spend daytime hours) combined with BLS data to estimate the proportion of all employed people working from home in each geography and industry. From there, we randomly assign employees to be “WFH” until that proportion is met.

You now have access to new Fall 2019 and Spring 2021 seasons for Great Lakes and Mid-Atlantic megaregions in Places. Both seasons and regions include data for a typical weekday (Thursday) and weekend day (Saturday). Learn more below.

Great Lakes combines the Great Lakes East and Great Lakes West megaregions to include coverage of IL, IN, KY, MI, OH, and WI and ~52M people.

Great Lakes Resources:

Mid-Atlantic now includes coverage of DC, MD, NC, VA, WV, and ~28M people.

Mid-Atlantic Resources:

Large Downloads Now Available in Trends and Places

You can now easily download larger datasets in Replica! Now, when you have applied necessary filters for your project in Places Explorer and need to download data for over 25 million trip or population records, or you need to download large origin-destination datasets, you’ll have the option to receive your download via email. On your Trends Dashboards, you’ll also now be able to download trip origin-destination information for a range of dates using the large download feature.

What constitutes a large download?

In Places, a large download contains more than 25,000,000 trip or population counts. In Trends, any origin-destination data download for a range of dates is considered a large download.

How will I receive large downloads that cannot be processed in my browser?

You will receive an email from support@replicahq.com with a link to the download once it has finished processing. Please allow up to 30 minutes to receive the email.

The download will contain a zipped file of the dataset in .csv format. Please note that due to the large file size, you may not be able to open the dataset in applications such as Microsoft Excel.

Are there download limitations even with large downloads?

In Places, the 20,000 row limit for network link volume downloads remains with this update.

In Trends, you can download origin-destination data for a date range of up to 2 months, regardless of how many records are contained in the download. If you’re interested in origin-destination data for a time period longer than 2 months, you will have to download the data multiple times for each 2 month time frame within the larger time range.

I never received an email with a link to my download, what can I do?

Please contact Replica through our chat option or email support@replicahq.com if you do not receive your large download via email after an hour

You now have access to new Fall 2019 and Spring 2021 seasons for the North Central megaregion in Places. North Central's geographic coverage remains the same, with data available for Iowa, Kansas, Minnesota, Missouri, Nebraska, North Dakota, South Dakota, and ~22M people. Both seasons include data for a typical weekday (Thursday) and weekend day (Saturday).

You now have access to new Fall 2019 and Spring 2021 seasons for South Atlantic and Southwest megaregions in Places. Both seasons and regions include data for a typical weekday (Thursday) and weekend day (Saturday). Learn more below.

South Atlantic (FL, GA, SC) Overview

South Atlantic now includes coverage of FL, GA, and SC and ~37M people.

Note: NC is no longer part of South Atlantic and instead, will be a part of an updated Mid-Atlantic megaregion, coming soon. We will not have coverage for NC in our platform for a brief period of time, though we can provide data to you directly in the interim as needed.

South Atlantic Resources:

Southwest (AZ, CO, NM, OK, TX, UT) Overview

Southwest combines New Texoma and West megaregions to include coverage of AZ, CO, NM, OK, TX, UT and ~51M people.

Southwest Resources:

We’re excited to introduce a new metric to our weekly Trends spend data releases: You now have access to a breakdown of online and in-person (offline) consumer spend by home location for three major categories: ​​ Retail, Grocery Stores, and Restaurants and Bars. With this update, we've also added the ability to see spend by both home location (this includes both online and offline spend) and by merchant location (this includes in-person transactions, only). This data is available from January 2019 to the most recent week.

You can use this data to analyze the share of online and offline spending for residents in a jurisdiction, track the change over time, and identify tracts in your study area with higher or lower levels of either online or offline spend. You can also use spend data in conjunction with our land use data to compare spend totals and land use distribution down to the Census tract level.

Spend estimates are modeled from multiple sources including cardholder and merchant transaction data which is then calibrated against ground truth data from the Census and Bureau of Economic Analysis. Previously, we showed estimated spend totals that included both in-person and online spend. With this new modeling pipeline, we are able to differentiate spend transactions by home location into online and offline transactions for the categories mentioned above.

How did you model and calibrate this new spend data?

Our updated spend pipeline incorporates cardholder and merchant spend transaction data as well as ground truth data from the Census and Bureau of Economic Analysis (BEA). Total spend was calculated at the tract level from aggregations of cardholder and merchant transaction data. Merchant spend transaction data also incorporated POI data to accurately map spend transactions to merchant locations. This data was then calibrated against nationwide monthly spend totals, including online spend estimates, published by the Census. State-wide spend totals from the BEA were also used to calculate the proportion of spend in each state.

Why are online and offline breakdowns not available for certain spend categories?

The share of online and offline spend was not distinguished for certain spend categories due to the high variability of online and offline spend in these categories. For example, in our “Gas Stations, Parking, Taxis, and Tolls” spend category, gas stations essentially have no online transactions whereas taxis and TNCs have a significant number of online transactions. In the future, we hope to provide spend data further broken down into subcategories of spend.

How can I use this data to calculate taxable sales income for my city or municipality?

To calculate taxable income for a select geography, you can combine the online spend estimates by home location and the in-person spend estimates by merchant location to capture a reasonable estimate of taxable sales income in an area. The in-person spend estimates by merchant location will capture spend transactions that are made by both residents and visitors to the area. For example, spend estimates by merchant location in New York City will also include purchases made by visitors from the neighboring areas such as New Jersey and Connecticut.

Why is spend by merchant location not broken down by online and offline transactions?

Currently, spend data by merchant location captures in-person or offline transaction records only. This is primarily because online transactions are difficult to pinpoint to a singular merchant location. For example, online purchases made at major online retailers, such as Amazon, cannot be mapped to a singular merchant location.

How would you categorize online food delivery services, like Door Dash, Uber Eats, etc.?

These transactions would be classified as online spend for restaurants.

How can you use spend data in conjunction with land use data to compare spend totals with land use?

You can analyze spend data in conjunction with land use data by first capturing the spend totals for your study area by either home location or merchant location at the city or tract level. You can then download the land use data for the same area and the same geography level and your desired time period. You can then join the two datasets by the “geo_id” field to compare spend totals and land use composition in your study area.

You now have access to the Spring 2021 season of Cal-Nev's megaregion in Places. The season includes data for a typical weekday (Thursday) and weekend day (Saturday) during March - May of 2021, and is the first time that detailed, disaggregate, activity-based model data exists for the Covid era.

We are excited to release Spring 2021 seasons for all megaregions. Cal-Nev is our first megaregion to have Spring 2021 data and we will continue releasing Spring 2021 seasons between now and the end of June.

What this means for you:

Resources for Cal-Nev's Spring 2021 season

You can now access the Cal-Nev megaregion in Places. With data available for Fall 2019, this megaregion includes all of California and Nevada, and combines what were previously two smaller regions (NorCal and SoCal) into a single model covering more than 40 million people.

Please note that we identified lower trip volumes for the week of December 6th. Specifically, average weekday trip volumes are ~3-4% lower than what we would expect. We will let you know if this issue is resolved in the future.

Improved Spend Estimates Now Available in Trends

On Monday, November 22nd, we released higher quality data on nationwide spend estimates in Trends. We made significant data quality enhancements by integrating cardholder transaction data and merchant data into our pipeline and using Census ground truth data to bring our spend estimates in closer range to Census totals in each category, as shown on the charts below.

With these improved data estimates, you can track spending trends over time with increased accuracy and more closely examine differences in spending trends between nearby geographies.

The release includes a full refresh of historical spend estimates dating back to January of 2019.

FAQS

What are the significant changes with this new data refresh?

We implemented numerous changes in this new spend pipeline, including:

You should expect to see the same metrics that we currently have in Trends, but with the following quality improvements.

Are unbanked people represented in this new data?

Yes, unbanked people who used cash or non-credit forms of payment during merchant transactions at restaurants or grocery stores are represented in our spend estimates.

Why did the total estimated spend in a category change so substantially in absolute dollar amounts as compared to previous totals?

The previous version of our spend model calibrated totals to census retail estimates. With the new pipeline we are calibrating each category (e.g. restaurants/bars, retail, grocery stores) independently, which in some cases changed the absolute total spend estimates. The newest numbers most accurately reflect category-level spending.

Why is there a spike in gas spending in July 2019?

We use many methods to correct for changes in the samples our vendors provide - using multiple sources of spend data as well census ground truth and mobility data. However, when there is a very large and abrupt change in sample, it can cause a temporary artifact in the output data (typically until we have the next month of data). This happened in July 2019 in the gas category. We recommend not comparing to that month of data, but all data before and after that month can be used for analysis.

Now, you can visualize and download land use data for all census geographies (down to the tract-level in Trends and block group-level in Places) and custom geographies. With this new dataset, easily access and quantify total square feet of land and building use for any study area. Included is data on all major categories, such as single- and multi-family residential, retail, office, industrial, and more.

Where can I learn more about this new dataset?

Read through our Land Use Data Overview and FAQ Help Article.

Where can I access this data?

Data is available for visualization and download in both Places Explorer and Trends dashboards.

How can I use this data?

Some examples we’ve heard from customers similar to you:

The ability to filter trips by the land use of their origin or destination will be available with the Spring 2021 season release (note that this functionality is not available for the Fall 2019 seasons).

How detailed is the data?

With this release, you have access to this data aggregated for any census geography (down to tracts in Trends and block groups in Places) or custom geography.

We will use the parcel-level data to assign land uses to each trip, and may make parcel-level data available directly to you in the future.

You can now filter trips that pass through a selected geography or multiple geographies with our new pass-through filter in Explorer.

When you select a study area using the geography selection tool, apply the "Trip Passes Through" filter. Once you click "Apply Filters", Explorer will filter down to all trips passing through the selected geography on a highway, primary, or secondary road. Note: This will not include trips that start or end in the selected area or trips that do not travel on network links, such as subway or rail trips.

The Great Lakes West megaregion covers the entirety of the states of Illinois, Indiana, and Wisconsin, and ~26m people. Data is available for a typical weekday (Thursday) and weekend day (Saturday) for the Fall 2019 season.

Below are resources for the megaregion:

If your organization currently has access to all available megaregions, then you now have access to this region. Click here to explore the new region.

The Northwest megaregion covers the entirety of the states of Idaho, Montana, Oregon, Washington, and Wyoming, and ~15m people. Data is available for a typical weekday (Thursday) and weekend day (Saturday) for the Fall 2019 season.

Below are resources for the Northeast megaregion:

If your organization currently has access to all available megaregions, then you now have access to this region. Click here to explore the new region.

The Northeast megaregion covers the entirety of the states of Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont, and ~11.5m people. Data is available for a typical weekday (Thursday) and weekend day (Saturday) for the Fall 2019 season.

Below are resources for the Northeast megaregion:

If your organization currently has access to all available megaregions, then you now have access to this region. Click here to explore the new region.

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the Northern California Fall 2019 megaregion and as of today published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Notable improvements

Re-calibrated transit ridership against ground truth data provided:

Expect the following differences compared to the previous version:

Below are resources for the Northern California megaregion:

The South Atlantic megaregion covers the entirety of the states of Georgia, North Carolina, and South Carolina, and ~26m people. Data is available for a typical weekday (Thursday) and weekend day (Saturday) for the Fall 2019 season.

Below are resources for the South Atlantic megaregion:

If your organization currently has access to all available megaregions, then you now have access to this region. Click here to explore the new region.

​

The South Central megaregion covers the entirety of the states of Alabama, Arkansas, Louisiana, Mississippi, and Tennessee, and ~23m people.

Below are resources for the West megaregion:

If your organization currently has access to all available megaregions, then you now have access to this region. Click here to explore the new region.

The West megaregion covers the entirety of the states of Arizona, Colorado, and Utah, and ~16m people. Data is available for a typical weekday (Thursday) and a typical weekend day (Saturday).

Below are resources for the West megaregion:

If your organization currently has access to all available megaregions, then you now have access to this region. Click here to explore the new region.

The North Atlantic megaregion covers the entirety of the states of Connecticut, Delaware, New Jersey, New York, and Pennsylvania, ~47m people, and includes data for a typical weekday (Thursday) and weekend day (Saturday) for the Fall 2019 season. As a reminder, megaregions have the same fidelity as existing metropolitan Places models, but with greater geographic coverage.

If your organization had access to the New York City Places model, then you now have access to the North Atlantic megaregion. Click here to explore the new region.

Read the North Atlantic megaregion profile to learn more about its geographic coverage, population size, and the transit routes included. Profiles for all megaregions can be found in the Help Center > Places Overview > Megaregion Profiles section.

​

​

​

If your organization does not have access to North Atlantic but you would like to learn more about it, please reach out to your Customer Success Manager to schedule a demonstration.

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the Great Lakes East Fall 2019 megaregion and as of today published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Updated synthetic population with new demographic attributes:

Access the new synthetic population in Explorer, Report Templates, or download the full export on the Places Data Downloads page under "Population Data".

​Improved modeling of transfers between agencies: We fixed a bug that under-estimated the number of transfers happening on transit trips.

Improved modeling of long-distance driving (private auto) and auto passenger (carpool) trips: The previous version of the model did not include long-distance driving and auto passenger trips above 325km. The newest version includes driving and carpool trips between 325-500km.

Expect the following differences in trip and population data compared to the previous version:

Below are resources for the Great Lakes East megaregion:

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the North Central Fall 2019 megaregion and as of today published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Re-calibrated auto volumes with the addition of KC Scout auto ground truth counts.

Updated synthetic population with new demographic attributes:

Access the new synthetic population in Explorer, Report Templates, or download the full export on the Places Data Downloads page under "Population Data".

​Improved modeling of transfers between agencies: We fixed a bug that under-estimated the number of transfers happening on transit trips.

Improved modeling of long-distance driving (private auto) and auto passenger (carpool) trips: The previous version of the model did not include long-distance driving and auto passenger trips above 325km. The newest version includes driving and carpool trips between 325-500km.

Expect the following differences in trip and population data compared to the previous version:

Below are resources for the North Central megaregion:

It is important that we produce and maintain high-quality data in Replica. We recently identified areas of improvement in the Mid-Atlantic Winter 2019 megaregion and as of today published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

West Virginia is now included in the region.

Updated synthetic population with new demographic attributes:

Access the new synthetic population in Explorer, Report Templates, or download the full export on the Places Data Downloads page under "Population Data".

​Improved modeling of transfers between agencies: We fixed a bug that under-estimated the number of transfers happening on transit trips.

Improved modeling of long-distance driving (private auto) and auto passenger (carpool) trips: The previous version of the model did not include long-distance driving and auto passenger trips above 325km. The newest version includes driving and carpool trips between 325-500km.

Added two transit agencies: Virginia Railway Express, West Virginia University Personal Rapid Transit (PRT)

Changes to be aware of

Expect the following differences in trip and population data compared to the previous version. As a reminder, the region now includes mobility and demographic data for West Virginia:

Below are resources for the Mid-Atlantic megaregion:

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the New Texoma Fall 2019 megaregion and as of today published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Notable improvements

Updated synthetic population with new demographic attributes:

Access the new synthetic population in Explorer, Report Templates, or download the full export on the Places Data Downloads page under "Population Data".

​Improved modeling of transfers between agencies: We fixed a bug that under-estimated the number of transfers happening on transit trips.

Improved modeling of long-distance driving (private auto) and auto passenger (carpool) trips: The previous version of the model did not include long-distance driving and auto passenger trips above 325km. The newest version includes driving and carpool trips between 325-500km.

Changes to be aware of

Expect the following differences in trip and population data compared to the previous version:

Below are resources for the New Texoma megaregion:

Please let us know if you have any questions. We're here to help.

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the Northern California Fall 2019 megaregion and as of today published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Additional transit ridership data was included for calibration and quality comparisons. The list of agencies this relates to includes:

Additional auto count data was included for calibration and quality comparisons. The type of auto count data sourced was:

Expect the following differences in trip and population data compared to the previous version:

Below are resources for the Northern California megaregion:

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the Northern California Fall 2019 megaregion and as this afternoon, published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Updated synthetic population with new demographic attributes:

Access the new synthetic population in Explorer, Report Templates, or download the full export on the Places Data Downloads page under "Population Data".

Additional transit ridership data was included for calibration and quality comparisons. The list of agencies this relates to includes:

Improved Taxi/TNC trips: We improved estimates of Taxi/TNC ridership.
​
​Improved modeling of transfers between agencies: We fixed a bug that under-estimated the number of transfers happening on transit trips.

Improved modeling of long-distance driving (private auto) and auto passenger (carpool) trips: The previous version of the model did not include long-distance driving and auto passenger trips above 325km. The newest version includes driving and carpool trips between 325-500km.

Expect the following differences in trip and population data compared to the previous version:

Below are resources for the Northern California megaregion:

Please let us know if you have any questions. We're here to help.

Now, you can view and download trip and traveler data for trips that start in a particular geography and geography type (e.g., a specific county) and end in a different census-based geography or uploaded custom geography and geography type (e.g., the central business districts in your region). This feature is now available in Places Explorer and the Places Origin/Destination Report Template.

As you select an origin point using the geography selection tool and click "Apply filters", you can update your geography type using the geography dropdown at the top of the page. Set your destination geography type and then click on the map to re-launch the geography selection tool.

​

Click here to explore this new feature.

It is important that we produce and maintain high-quality data in Replica. We recently identified an area of improvement in the Florida Fall 2019 megaregion and as this afternoon, published a new version. We will continue to alert you any time a new season or new version of an existing season of a Places model that you have access to is published.

Updated synthetic population with totals more closely aligned to American Community Survey (ACS) data. The updated Population Quality Report shows how Replica's population compares to ACS data and other census ground truth data.

The updated population also includes new attributes:

Access the new synthetic population in Explorer, Report Templates, or download the full export on the Places Data Downloads page under "Population Data".

Improved modeling of transfers between agencies: We fixed a bug that under-estimated the number of transfers happening on transit trips.

Expect the following differences in trip and population data compared to the previous version:

Below are resources for the Florida megaregion:

Please let us know if you have any questions. We're here to help.

Last week, we resumed publishing mobility data in Trends. As part of this update, we republished mobility data from January 6, 2020 - May 31, 2020. See below a detailed summary about the resulting changes to Trends mobility data.

Data for the week of June 7th available in Trends

As of this morning, we've published data for the week of June 7th in Trends. On Thursday, we will resume our weekly release schedule and publish data for the week of June 14th.

As a reminder, we are hosting a series of office hours this week to answer your questions about the changes to Trends mobility data. Register for a session below.

Trends Mobility Update Office Hours

Tuesday, June 22nd at 1 PM PT/ 3 PM CT / 4 PM ET

Register here

Thursday, June 24th at 12 PM PT/ 2 PM CT / 3 PM ET

Register here

In the meantime, please reach out through our in-app chat with immediate questions.

This morning, Replica resumed publishing mobility data in Trends. As part of this update, we have republished mobility data from January 6, 2020, through May 31, 2021.

Background

Starting in mid-April, we noticed abnormally low trip volumes in our Trends product — first in New York and California, and then nationwide. We stopped publishing mobility metrics until we identified the problem and implemented a resolution.

The problem

The root cause of the issue was the release of Apple’s iOS 14.5 with App Tracking Transparency (ATT). We deeply value privacy and are supportive of tighter privacy controls, but they had a substantial impact on the location-based services (LBS) data ecosystem. Our mobility pipelines are robust to variations in location data and are typically able to correct for changes in device quantity and quality, but in advance of the iOS release, numerous data providers unexpectedly exited the industry entirely, degrading data availability so substantially that our algorithms could not correct for it. These changes have affected the entire location and mobility data industry — all products that use location-based service data will have been affected.

The solution

Fortunately, Replica builds its models using multiple types and providers of data. To account for decreased quality in mobile location data, we corrected mobile location-based Trends estimates with additional location data obtained from in-vehicle GPS providers, which were not affected by iOS 14.5. We find these data sources to be very reliable and of high quality. Overall, we are confident that our Trends mobility estimates are now more accurate than they were in the previous paradigm.

About the new mobility data

Here is what you need to know about this update:

Next steps

On Tuesday, June 22nd, we will release mobility and spend data for the week of June 7th. On Thursday, June 24th, we will resume our weekly release schedule and publish mobility and spend data for the week of June 14th.

We continue to make improvements to Trends data over time, and we welcome your feedback and observations of the new mobility data. We will host a series of office hours over the next two weeks to answer questions about this update. Register to attend one of the sessions below.

Trends Mobility Update Office Hours: Option 1

Friday, June 18th at 9 AM PT/ 11 AM CT / 12 PM ET

Register here

Trends Mobility Update Office Hours: Option 2

Tuesday, June 22nd at 1 PM PT/ 3 PM CT / 4 PM ET

Register here

Trends Mobility Update Office Hours: Option 3

Thursday, June 24th at 12 PM PT/ 2 PM CT / 3 PM ET

Register here

Please reach out through our in-app chat if you have immediate questions.

In mid-April, we identified an issue with abnormally low trip volumes in Trends that affects all mobility metrics. Since then, we’ve identified the root cause of the issue, are nearing the end of our testing phase, and finalizing the implementation plan and date.

Publishing high-quality data that our customers can use to inform decision-making is imperative to us, so we will not publish new or backfill mobility data from the week of April 12th until we've implemented this solution.

In the meantime, we continue to publish new spend data weekly. Spend data for the week of May 31st is now available.

We’re eager to continue delivering mobility metrics in Trends and will inform you of the implementation plan soon.

In mid-April, we identified an issue with abnormally low trip volumes in Trends that affects all mobility metrics. Since then, we’ve identified the root cause of the issue and are nearing the end of our testing phase to implement a resolution in the next few weeks.

Publishing high-quality data that our customers can use to inform decision-making is imperative to us, so we will not publish new or backfill mobility data from the week of April 12th until we've implemented this solution.

In the meantime, we continue to publish new spend data weekly. Spend data for the week of May 24th is now available.

We’re eager to continue delivering mobility metrics in Trends and will inform you once we finalize the implementation plan.

Spend data for the week of May 17th is now available in Trends. Publishing high-quality data that our customers can use to inform decision-making is imperative to us, so we will not publish new or backfill mobility data from the week of April 12th until we've implemented a solution that resolves abnormally low trip counts.

We're in the testing phase of implementing the solution and expect to have the issues resolved within the next few weeks. We will inform you once we finalize the implementation plan.

Increased network link selection capacity in Explorer
Select up to 20,000 network links using the network link selection tool in Explorer to understand network utilization or pass-through travel corridor studies.

​

​Batch-select geographies in just a few clicks in Explorer
Use the new polygonal lasso tool to batch-select census-based geographies or portions of uploaded custom geographies. See this new feature in action in the image below, where we've selected the block groups in downtown Sacramento, CA to set as the home location of our study area.

​

We are in the testing phase of our resolution to fix abnormally low trip volumes in Trends and backfill most or all of the affected data. We continue to make this our highest priority and expect to implement a resolution within the next few weeks. We will inform you once we've finalized our implementation plan.

As of today, we've refreshed spend data for the week of May 10th. Publishing high-quality data that our customers can use to inform decision-making is imperative to us, so while we continue to work through the low trip counts issue, we will not publish new or backfill recent mobility data.

We are continuing to work to resolve the issue around abnormally low trip counts in Trends. We have identified the root causes in our data pipeline, and are working to resolve the problem and backfill most or all of the affected data. We do not yet have a definitive timeline as it involves working with our data partners but hope that it will be resolved soon.

By Thursday, May 13th, we will refresh spend data for the week of May 3rd and backfill spend for the weeks of April 19th and 26th. We will not yet refresh mobility data for the week of May 3rd or backfill mobility data for the weeks of April 19th and 26th. Publishing high-quality data that our customers can use to inform decision-making is imperative to us.

We are continuing to work to resolve the issue around abnormally low trip counts in Trends. We have identified the root causes in our data pipeline, and are working to resolve the problem and backfill most or all of the affected data. We do not yet have a definitive timeline as it involves working with our data partners, but hope that it will be resolved soon.

Publishing high-quality data that our customers can use to inform decision-making is imperative to us, so while we investigate the causes of this issue we have chosen to not publish Trends data for the week of April 26th.

We are continuing to work to resolve the issue around abnormally low trip counts in Trends. We have identified the root causes in our data pipeline, and are working to resolve the problem and backfill most or all of the affected data. We do not yet have a definitive timeline as it involves working with our data partners, but expect that it will be resolved this week. We will provide daily updates with the latest information, and please feel free to reach out with additional questions.

The Florida megaregion covers the entirety of the state of Florida and ~11m people. Megaregions have the same fidelity as existing metropolitan Places models, but with greater geographic coverage.

Read the Florida megaregion profile to learn more. Profiles for all megaregions can be found in the Help Center > Places Overview > Megaregion Profiles section.

Reply directly to this email with any questions.

We have been working hard to resolve the issue with low trip volumes that caused us to pause our Trends dashboard update for this week. We believe we have identified the root causes, and are working closely with our data team and external vendors to find a fix. It seems that the issue affecting New York and California was driven by the same problem affecting the remainder of the US this week, so we expect that everything will be resolved concurrently. We also are hopeful that the resolution will include a backfill of the affected data, but are still working to confirm this. We do not yet have a time to resolution but expect to have more information this weekend. We apologize again for the inconvenience and please don't hesitate to reach out to us if you have other questions.

This week, we identified abnormally low trip volumes nationwide. Publishing high-quality data that our customers can use to inform decision-making is imperative to us, so while we investigate the causes of this issue we have chosen to not publish Trends data for the week of April 19th.

We previously identified an issue with low trip volumes affecting New York and California during the weeks of April 5th and April 12th. Please note that while we initially said the April 5th issue was affecting mobility data for the entire United States, we've since realized that it was just affecting New York and California. We are investigating whether this was due to the same root cause and if so expect to resolve it jointly with the previous issue.

It is our highest priority to resolve these issues and we sincerely apologize for any inconvenience it might cause. We are working on a resolution plan as fast as we can, and will share an update by the end of the week.

Custom geographies are designed to make it easier to view data for the study areas most relevant to you, like city neighborhoods or state districts. Visit the new “My Data” page to upload custom geographies as zipped shapefiles or GeoJSON layers.

Once custom geographies are successfully uploaded and are processed by Replica, you can view them in Trends dashboards, Places Explorer, or Report Templates, just as you previously have with census-defined geographies.

We've identified an issue with low trip volumes for the weeks of April 5th and April 12th in Trends. For the week of April 5th, we've identified between an 8-10% decrease in total trips compared to the prior week nationwide. For the week of April 12th, we've identified abnormally low counts in New York and California.

We are still investigating this issue and how it can be resolved for future data refreshes. We will continue to post status updates in our platform and in our weekly Mobility Roundup newsletter. If you have any questions in the meantime, please let us know.

As of this morning, Trends has been refreshed with data for the week of April 12th. We've identified an issue with abnormally low trip counts in New York and California for the week of April 12th. It is our highest priority to investigate this issue and confirm how it can be resolved for future data refreshes.

The new megaregion covers the entirety of the states of Kansas, Missouri, Nebraska, Iowa, South Dakota, North Dakota, and Minnesota, and ~21M people. Megaregions have the same fidelity as existing metropolitan Places models, but with greater geographic coverage.

Note: While the Kansas City region within North Central has significantly improved since our last seasonal release, the KC Streetcar ridership counts are currently below observed ridership. Replica is actively investigating and once resolved, will communicate immediately.

Read the North Central megaregion profile to learn more about the megaregion. Profiles for all available megaregions can be found in the Help Center > Places Overview > Megaregion Profile section.

As of this morning, Trends has been refreshed with data for the week of April 5th. We've identified between an 8-10% decrease in total trips compared to the week of March 28th. We're working with our data vendor to investigate this issue and confirm if it can be resolved for future data refreshes. Data prior to the week of April 5th remains unaffected.

The New Texoma megaregion covers the entirety of the states of Texas, New Mexico, and Oklahoma, and ~35M people. Megaregions have the same fidelity as existing metropolitan Places models, but with greater geographic coverage.

Read the New Texoma megaregion profile to learn more about this Places model. Profiles for all available megaregions can be found in the Help Center > Places Overview > Megaregion Profiles section.

If you have full access to our Places product, you can explore the new region today by visiting the Places Overview page and opening up Explorer, Reports, or Data Downloads. If you’re only subscribed to Trends or to a specific Places model, reply to this email if you're interested in getting a demonstration of the new region.

In mid-February, we implemented quality improvements for all mobility metrics in Trends. Since the release of the February improvements, we’ve partnered with some of our customers to compare Replica’s trip volumes against customer-provided ground truth. During this analysis, we identified more areas for improvement to our mobility data. As of Friday, April 2nd, you will see new mobility data for the entire United States, impacting all metrics and time periods. While the impact will vary based on your geographic location, the new mobility data shows a more significant and sustained impact of COVID-19. Since these quality improvements affect historical data, we recommend re-downloading mobility Trends data for any current or ongoing analyses that you’re working on. Click here to see the updated Trends data for your geography.

We've made it easier to select different date ranges in Trends to view in the dashboard or download data for, like 6 months ago, 1 year ago, 2 years ago. You can also select different months too.

We've updated Explorer to remove the auto-save function when creating a new Workbook. Instead, you can click the "Save Workbook" button located at the top of Explorer any time you open up a new workbook.

When you save a workbook, you can access it again by visiting the "My Workbooks" page in the left-hand menu.

Explorer now supports bulk network link selection through our new network link lasso feature. Watch the video below to learn more about this capability.

Northern California/Nevada and Southern California megaregions are the third and fourth of fourteen total megaregions that will be available in Places in 2021. Together, these two models cover the entirety of the state of California, as well as Nevada, and ~42 million people. Megaregions have the same fidelity as existing metropolitan Places models, but with greater geographic coverage. Here is a snapshot of the region in Places Explorer.

Learn more about both megaregions by reading the profiles below. You can find these in our Help Center > Places Overview > Megaregion Profiles section.

Performance and infrastructure improvements have enabled us to offer a number of new downloads features:

Mobility Data Upgrade

We've implemented a major update to the way we process mobility data in Trends, both incorporating additional data sources and improving our modeling algorithms. The result is increased confidence in our mobility metrics, resilience to upstream data issues, and more accurate comparisons over long periods of time, like year over year. As part of these improvements, we correct for times where input data sources previously underestimated trips, so overall trip counts will increase by 15-25%, depending on the geography. You’ll see these improvements today, applied to all mobility data, past and future.

Trends Events

Use Events to tag interventions or occurrences to understand their impact on Trends in your geography.

Learn more about this new feature below.

The Great Lakes East megaregion is the second of 14 total megaregions that will be available in Places in 2021. The region includes the states of Kentucky, Michigan, and Ohio, and has a combined population of ~27M people. Megaregions have the same fidelity as existing metropolitan Places models, but with greater geographic coverage. They come with their own Quality Reports.

We’ve rolled out additional geography breakdown options when downloading O/D data in Trends. This means that when you select O/D data as the metric category on the Trends Downloads page or Download Data model in your geography’s dashboard, you can download O/D by county for your state, for example.

This week we launched new choropleth maps in Trends on our Overview, Mobility and Spend tabs:

Replica continues to invest in new data sources and advancements in modeling techniques to improve our algorithms in Trends. Our latest improvements focus on our consumer spend metrics.

We’ve recently introduced a new data source. There are three main parties in consumer transactions: issuers/banks, merchants, and credit card companies. Replica now has data from two of these parties - issuers and merchants, and we’re evaluating adding the third.

With these improvements, we’ve made a number of changes to the interface, including:

Please note that you’ll see changes in the past and future consumer spend numbers. As with all changes, we welcome feedback, so please reach out to us with any questions.

Watch this short video to learn more about our spend updates:

A month ago, we launched confidence indicators in Trends to be more open and transparent about our methodology. This week, we promoted select metrics to “Quantified Trends Estimate.” This indicator means that these metrics were based on measurements against historical ground truth data and we can report a margin of error for the metric and geography.

Below are the metrics labeled with the “Quantified Trends Estimate” indicator.

Note: To see the margin of error, hover over each data point on the module. Learn more about this update in the video below.

We’ve prioritized releasing 2019 data for a set of mobility metrics at specific spatial fidelity levels so you can see year-over-year comparisons. Below is what is included in this initial release.

All metrics that show 2019 data have a confidence indicator of “Early Access.” The recency of the release means that these metrics may see an occasional bug or data glitch. Please reach out through our chat icon if you have any feedback.

Learn more about this update in the video below.

Our new Trends Data Downloads page allows for a faster and easier way to download the data you need on a regular basis. Using this new page, you can download data for any location, metric, and date range of your choice, in just a few clicks. You can also download data broken down by sub-geographies, like Mode Split by county for your state.

Readme Files have been added to each download on the Trends dashboard to provide more detail on each metric and associated fields/categories available. The readme files provide:

With a goal to be open and transparent about our methodology, we’ve added confidence indicators to each metric on the Trends dashboard. The indicators include “Early Access” and “Trends Estimate” to provide more information about the data.

Note: The data has not changed, we’re just surfacing explanations we’ve shared with customers directly in the interface to accommodate our growing user base. As we continue to add additional data sources and improve algorithms, metrics will move up categories over time.

We've made significant improvements to our data pipelines and algorithms. These technical enhancements have improved the quality of all metrics. These quality improvements are most notable in total trips and person trip rates but also flow through to all other modules in the dashboard.

The Residential VMT module shows the total number of vehicle miles traveled by all residents of the selected geography on an average weekday in the relevant week. This metric represents the number of miles traveled in automobiles by people who live in the selected geography. Trips count toward this total regardless of whether or not the trips themselves took place within the selected geography.

This module is available for each geography in the United States, meaning it’s available at the Nationwide level, down to the census tract level.

The Mode Split module shows the mode split for trips taken on an average weekday in each week. It’s available for each geography in the United States, from the Nationwide level, down to the census tract level. This metric can be presented in three ways including, total volume, the proportion of trips taken for each purpose or mode, and the relative change against a baseline date of your choice.

Each mode option is defined below:

The Trip Purpose module shows the purpose for all trips starting in a geography on an average weekday in each week. It’s available for each geography in the United States, from the Nationwide level, down to the census tract level. This metric can be presented in three ways including, total volume, the proportion of trips taken for each purpose or mode, and the relative change against a baseline date of your choice.

Each category of purpose is defined below:

The Sheltering-in-Place module shows the proportion of all residents estimated to have sheltered-in-place during an average weekday in each week. A person who is sheltering-in-place is defined as someone who is estimated to not have taken a trip outside of his or her home during that day.

Sometimes very short trips, such as visiting a convenience store on the same block as one’s residence, or trips that don’t have a unique staypoint, such as a run with no stops that starts and ends at home, may not register when calculating this metric. If a resident’s only trips for a given day fall into either of these categories, they will be counted as a person sheltering-in-place.

The Travel to Work and School module shows the proportion of all residents estimated to have traveled either to school or work during an average weekday in each week. All residents, regardless of age or employment status, are included in this calculation.

The Consumer Spend module shows the relative change in consumer spend for the selected geography. Categories of spend are derived using the Merchant Category Code (MCC) system and are listed below.

Relative spend is calculated for a given geography, not a set of residents. For example, the relative spend displayed for a given county corresponds to all spending activity that takes place within that geography, regardless of where the consumer lives.

"Proportion of Intra-Geo Trips" modules shows the percentage of trips that both start and end in the selected geography on an average day in each week. This percentage is calculated by dividing the number of trips that both start and end in the selected geography by the number of total trip origins for that geography.

Demographic data is now available for every geography in the U.S. In addition to the total population of a geography, demographic characteristics include:

Please note, the demographic data will be updated annually based on the U.S. Census data.

The ability to select multiple geographies is available on the nationwide data dashboard. You can combine up to 10 defined geographies to view aggregate metrics. These custom geographies will be saved under "Favorite Places" allowing for quick reference on the side bar navigation.

We have launched COVID-19 data in Replica by displaying the following metrics:

These metrics are shown to illustrate correlations between movement and case counts and available at the state and county level.