For several years I’ve been frustrated trying to identify which action items are needed and which ones are working based on a single site conversion metric. Diving into the issue, one begins to realize the real problem: overall conversion rates are not good for much except website developer bragging rights. And they don’t even do a good job of that, because providers are free to define conversion any way they want to. The point of collecting performance metrics is to know what actions you need to take to improve the situation. There are four things one needs to control for when trying to identify the right action needed to drive more leads to the store:
- The denominator needs to be known, defined consistently, and measured consistently in all cases.
- What type of shopper objective the lead was based on (i.e. vehicle sales, finance, service, or parts)
- What kind of device the shopper used (i.e. desktop, tablet, or mobile)
- What type of lead the shopper sent (i.e. phone, online form, chat)
Most experts seem to agree, use unique visitors as the denominator. The source to get it from is Google Analytics. Google Analytics is the gold standard for data in automotive retail. It’s what the 20 group providers use for their internet composites and the source of virtually all meaningful benchmarks. Benchmarks and best practices need to be based on a robust sample size, and Google Analytics is far and away the most common measuring instrument. An added bonus is trust between the dealer and the website provider. Google’s metrics have a set definitions and the measurement method is both consistent and independent.
The manager or analyst needs to know the shopper objective, because there may be some types of inquiries the store has capacity for and some it does not. Capacity challenges can relate to handling of the leads or providing the service. The cost of increasing traffic is not the same across all shopper objectives, and neither is the benefit. The action items to improve conversions are completely different for fixed ops, finance, or sales. A breakdown of conversion by shopper objective is essential.
If the conversion ratio is low or declining, it’s important to know where the site has the greatest room for improvement. More than ever, that segmentation is based on the type of device the shopper is using. When new traffic is driven to the sight from mobile users, it’s important to be able to isolate the impact on the conversion ratios for mobile users exclusively. This can be one of the big challenges with a Responsive Web Design, a challenge that does not exist with an Adaptive Web Design. Although generally not done, it is possible to split out the conversion rate by the consumers’ shopping devices. Of course, you cannot evaluate and improve the various website responses as effectively without this individualized conversion data.
If a store sets appointments at twice the rate for phone and chat leads vs. form leads, then it makes since to increase conversion of phone and chat leads, even if there is some cannibalization of form leads in the process. But you can’t know any of that is happening if you don’t have the metrics broken out. If you look at five actions to improve conversion, they will probably improve conversion across the three lead types in the different proportion.
When I joined Dealer eProcess I had no idea they were rolling out a complete solution to the problem I was just beginning to understand. The new Lead Conversion report shows all 36 conversion metrics as well as the raw number of leads. That raw number is especially important when capacity issues come into play. The 36 for conversion metrics come from 3 devices X 3 lead forms X 4 customer shopping objectives = 36.
Not only is this approach more diagnostic, it’s more accurate. As I wrote in my first book, Sales Integration, the customer transitions from the online touch-points to the store in one of four ways: phone, email, chat, or walking in. None of us can accurately measure walk-in traffic very well, but we can darn sure know the other three. These three variables do not always go up or down in proportion to one another, so if you only have one or two of the three, it’s a real stretch to think you can estimate the true conversion rate with this limited data. We are already missing a big source of traffic from the site, walk-ins. There is no room to leave out any of the other three and still pretend the amount of data is sufficient to truly understand what is going on.
The risk is deciding to do or not do something based on inaccurate data. From a research standpoint, conversion is a fantastic dependent variable with which to measure the impact of site changes. It’s the right metric to use. This is what Google and virtually every other search engine provider preaches. Clearly, any and every conversion metric must be as accurate as possible.
Every dealer deserves transparency in their data, but they also deserve accuracy and completeness. Without that, you can’t very well help yourself. And no trainer, consultant, ad agency, or vendor can make up for that lack of core data from the website. The transparency, accuracy, and completeness must come from the website provider. The website is not only the hub of your digital marketing effort; it is the hub of your analytical ability and continued improvement.