Discussions

Hi there,

I have been working with the latest refresh of the Verisk Auto data and have a couple of questions about it.

1. When comparing the number of households in the Verisk Consumer Profiles data to external household totals from the ACS, it looks like roughly 85% of households are covered. However, There are only about 118 million unique VINs in the auto data, which is much less than the roughly 290 million registered vehicles in the US (from FHWA total registrations). Verisk coverage also seems to vary substantially by state -- for example, in California, it only seems to have about 12% of registered vehicles. This seems to imply that coverage in the Verisk Auto data is substantially worse than coverage in the Verisk Consumer Profiles data. Is this primarily because (i) many covered households have no vehicle record, (ii) households often have only a subset of their vehicles observed, or (iii) many vehicle records are non-VIN / VIN suppressed? Who exactly is covered in the Verisk auto data, and why does coverage vary so much by state?
2. When looking to newer models in the Verisk data, there are around 1.2 million unique vehicles with model year 2024. There were roughly 16 million new light-duty vehicles sold in the US in 2024. I'm wondering why totals in the Verisk Auto data for recent vehicles are so much lower than external totals. Is there a substantial lag (e.g. one or two years or more) between when a vehicle is transacted/registered and when it appears in the data? If so, what is the typical lag between a vehicle being transacted and appearing in the Verisk Auto dataset with a VIN? Based on the overall VIN counts, Verisk accounts for about 40% of vehicles in the United States, but with newer vehicles this seems to be closer to 7%, and I'm trying to understand why.

Thank you very much for your time.