Why Bad Data Kills Car Searches

automotive search

By George Nenni, Vice President of Operations 

I’ve recently started searching for a replacement family vehicle. Because I work in the automotive industry, I used this as an opportunity to immerse myself in the car shopping experience. While continuing to learn throughout this process, I have realized how broken the inventory management and search sectors are of the automotive industry. I’ve worked in the inventory management and vehicle data business since 1993, making problems in this area even more painful. Here’s why:

I began a search for a specific make, model and trim level using a major automotive search portal. In this case, I was searching for a used 4-door Jeep Wrangler Unlimited. When selecting my trim preference, I noticed they had “Unlimited” as a trim option. This is incorrect, since the choices should have been Rubicon, Sahara, etc. When the website did not find any matches for such a popular vehicle, I was very surprised. I reluctantly decided to remove the trim level and search for Wranglers (all trims). I know this means potentially wasting time sifting through vehicles that are not exactly what I am looking for, but didn’t see another alternative. This search engine and/or vehicle data was not compiled in a way that helps shoppers find the exact model they are searching for. In my search, this website returned 102 matching vehicles. It was now up to me to either waste time sifting through all of the results, or try adding some keyword-searches into the mix. What seems so simple a concept- having good data, is nonexistent on this website.  

In many cases, portals receive data from thousands of different sources.  They may receive a large feed from Dealer Specialties or another aggregator, or receive individual data sets from dealerships. The individual vehicle records may be described cleanly, with trim levels and options standardized for good searchability, or the data may be inconsistent, with trim levels and options missing or misspelled. The best way to ensure that data is clean, standardized and normalized is to make sure it is collected using robust inventory management software on a smartphone, tablet or laptop. Ideally, this software is used while the individual is actually touching the vehicle, verifying the trim, transmission, options and packages. The individual can then put the data standardized format with style IDs, OEM option codes and OEM package codes. Otherwise, even the best search engines will have a hard time properly searching the data the way the consumer wants to shop.

In addition, the front-end search interface on the consumer portal or dealer website must be properly designed. Each search parameter must be correctly mapped to trim levels, transmissions, options and option packages. The vehicle inventory data must be described in the way that the inventory search engine will locate it. Ideally, dealers should choose vendors that have strong vertical integrations and offer a more complete platform solution. This mean, they collect, describe, and VIN decode inventory in the same way that websites or automotive portals search for it. For many providers, this integration is too complicated to warrant the effort. However, it is incredibly necessary if the car shoppers’ needs are a high priority. 

In an ideal world, both the front-end search interface and the back-end inventory descriptions would be built using a quality VIN decoding and configuration data set.  In addition, both the dealer’s inventory management tool and website interface would be built with that same data. In that case, it might look something like this:

data search

By consistently describing the data, and searching for it using the same approach, dealers give car shoppers a large number of choices and drill-downs.  Dealers also have confidence that when the consumer searches for specific trim level, options and packages, only those vehicles are displayed for them.

Share