Skip to main content

Frequently Asked Questions

Migration Patterns​

How do your clients use this data?​

  • Our customers use migration data to validate their investments in residential home development by identifying high income regions for new living communities and excluding all regions with declining population. They need income and outflow of population information at the ZIP code level, which this dataset provides.
  • Healthcare providers use migration data to understand shifts in population and age profiles to tailor their offerings in the regions they operate.
  • CRE developers confirm their investment theses with their stakeholders by showing growing areas with relevant people of affluent backgrounds moving in.

What time frame is used to identify origin or destination?​

We use historical data from Jan 2018 to the end of 2021, excluding 2020. This is because 2020 was an unusual year, and even when we do include 2020, it doesn’t significantly change the flows .

How real time is this data? What is the time lag between a move and this data recognizing the move?​​

USPS registers moves when they are reported, and that data is updated monthly. So the only delays are based on the time it takes people to report an address change.

What demographics can you provide?​

  • We provide Age and Income.
  • Demographics are based on census data from the year 2018, which is the last one that provides population on the ZIP code level. New Census data is not yet available on this level.

How have you validated your data?​

We statistically compare with datasets that contain similar kinds of information, such as Census Population Estimates, SOI Tax Returns Data or California Credit data. Overall, our dataset is very accurate as it performs well against these datasets.

What data sources do you use to calculate migration?​

  • For estimating area inflow, outflow and netflow, we use aggregated data from USPS. The main benefit is that it has very large market share (e.g., 12 million observed moves in 2019).
  • Origin-Destination flows are based on our GPS-based dataset that uses historical data to establish the graph of typical flows, which, in combination with Area Migration data, provides origin-destination migration estimates.
  • This combination ensures high confidence in the data because it combines an almost perfect sample of observed moves (market share) and typical flow patterns trained over years of historical data.

If USPS data is available, why is this product advantageous rather than requesting the USPS data?​

USPS is available but difficult to get a hold of and can be extremely messy to work with. We have simplified the data and aggregated it to different geographical boundaries not available from USPS. We also provide origins and demographics (age & income) of the movers, which USPS does not.

Are non-standard (non-out-of-the-box) administrative boundaries available (ex. Zip+4)?​​

Not at the moment. We support State, CBSA, Counties, and ZIP code.

How should the Student Population Fraction be used?​

This metric can be used to understand if an area compromises students and thereby experiences seasonal variations. Students typically move in during a term period and move out when it ends. As an investor, you can use this information to understand if migration (high inflow or outflow) is due to student seasonality or more permanent moves.

When should I use estimated vs. normalized moves?​

We provide both estimated and normalized moves. Estimated provides moves by the calendar month, some of which contain different days (e.g. January has 31 days, February has 28 days). Normalized moves correspond to four weeks. For long-term trends, both estimated and normalized provide the same insights. We offer normalized moves for customers who need to perform strict studies comparing changes per four weeks.

What are the limitations of this dataset?​

The Migration Patterns dataset describes migration only within the US and does not describe migration of people moving into the US from other countries.