Data-Rich, Insight-Poor: Why Motor Vehicle OEMs Underuse the Data They Already Have

Most motor vehicle OEMs do not have a data problem. They have an insight problem. They collect registrations, warranty claims, survey responses, dealer reports, and customer records every day, and most of it never informs a decision bigger than a monthly status update.

Motor vehicle OEMs are data-rich and insight-poor when they collect large volumes of customer and dealer data but lack the capability to turn it into decisions. The constraint is rarely more data. It is the modeling, benchmarking, and interpretation that convert raw records into clear answers about where to add dealers, who to market to, and which dealers need attention.

What does “data-rich, insight-poor” mean for an OEM?

It means the raw material is sitting in the building while the finished product never gets made. The pattern is not unique to motor vehicles. Forrester has estimated that between 60 and 73 percent of all data within a typical enterprise goes unused for analytics, and Seagate’s Rethink Data report found that only 32 percent of data available to businesses is ever put to work, leaving 68 percent unleveraged. An OEM might know every unit it has sold and every warranty event, yet still make network and marketing decisions on instinct because no one has translated those records into market potential, propensity, or performance against opportunity.

Data becomes insight only when it is measured against the right benchmark. A dealer’s sales figure means nothing until you compare it to what its territory could realistically produce. A customer record means nothing until you score how likely that customer is to buy again. The records are necessary, but they are not the answer.

Why does so much OEM data go unused?

The first reason is that the data lives in disconnected systems. Sales sits in one platform, warranty in another, survey results in a third, and dealer submissions in spreadsheets. No single view exists, so no one can ask a question that spans them. This matters financially: a BCG and Google study of more than 20 brands found that companies linking all of their first-party data sources earned up to twice the incremental revenue from a single ad placement compared to those with limited data integration.

The second reason is the absence of a market potential baseline. Without an estimate of what each geography could produce, every internal number floats free of context. You can see what happened, but not whether it was good. This is the same flaw that makes most dealer performance dashboards misleading: they rank internal results against each other instead of against opportunity.

The third reason is capability. Propensity scoring, churn modeling, and spatial demand estimation require statistical methods that most OEM teams are not staffed to run. The data is willing, but the toolset is missing.

What does it look like when an OEM actually uses its data?

When the data is working, three things change. Network decisions get made against modeled demand rather than population maps, so open point and expansion calls are defensible. Marketing budget goes to scored prospects instead of the whole database, through methods like propensity modeling. And dealer conversations shift from “you are below average” to “you are capturing 62 percent of what your territory can produce, and here is where the other 38 percent is going.”

That is the difference between reporting and intelligence. Reporting tells you what happened. Intelligence tells you what to do about it.

Where should an OEM start?

Start with the decision, not the data. Pick one recurring, expensive decision the organization keeps making on instinct, whether that is where to add a dealer, who to target in a campaign, or which dealers to coach. Then assemble only the data that decision requires and benchmark it against true market potential. One decision made well on evidence builds more momentum than a year spent building a data lake nobody queries.

Sextant’s consumer and market analytics and dealer performance intelligence practices exist for exactly this gap, turning the data OEMs already own into decisions they can act on and defend.

Frequently Asked Questions

Is the problem that OEMs need to collect more data?

Usually not. Most OEMs already collect more than they analyze; Forrester estimates 60 to 73 percent of enterprise data goes unused for analytics. The binding constraint is analytical capability and a market potential baseline to measure against, not data volume. Collecting more without the ability to interpret it widens the gap.

What is the difference between reporting and analytics?

Reporting describes what happened, such as units sold last month by dealer. Analytics explains and predicts, such as how that result compares to territory potential and which customers are most likely to buy next. Reporting looks backward at facts. Analytics looks forward to decisions.

How long does it take to get value from existing data?

When the question is well defined and the data already exists, a single high-value decision can be informed in weeks rather than months. The timeline depends far more on clarity of the question and data accessibility than on building large infrastructure.

Sources: Forrester, on the share of enterprise data unused for analytics (60-73%), reported via AtScale (atscale.com); Seagate, Rethink Data Report, 2020 (seagate.com); Boston Consulting Group with Google, “Responsible Marketing with First-Party Data,” 2020 (bcg.com).