Latest Study on Dealership Website platforms finds out...

[yagoparamo]

@Alexander Lau

I agree with you that on paper that sounds good, but if you read DI testing parameters "For Optimizely, in general you need 12,000 website visitors per test variation". That is a lot of traffic on each variation in order to yield the proper data, so it means that the time line is long (several months to get enough traffic on that particular change). The problem is that in the car business as our inventory changes, the OEM offer changes, model changes, availability changes, weather changes, etc customer input starts to vary with that so the intent of month 1 in our sample is different of the intent on month 2 however our results will have those together. I believe that a multi dealer A/B test that allows for a reduced time sampling yields better results to apply changes.

We added heat mapping 5-6 years ago to all our sites and almost to the last one, regardless of the color, design, banner, etc they all had the exact same patterns. We even tried, within the norms of compliance, to change consumer behavior as some dealers accepted the challenge to send customers thru a funnel of our choosing. That proved difficult. So I came to believe, an idea that I can't fully develop right now on this paragraph, that consumers follow a certain set of rules and patterns when working with a website and that these didn't change very much so we started focusing on following these patterns.

Lots of data here https://www.nngroup.com/articles/

 

[Chris Cachor]

My 2 cents: Since most dealer sites sit on a shared codebase, you can A/B test among a larger pool of sites.

However, I think A/B testing for these types of sites is overkill. All you want to do is find out better ways to convert. Strategies around pricing, different lead forms, etc. You don't need to do anything fancy for this, just be willing to experiment. I think if something is working, it's going to be pretty obvious. "Get e-Price" is a perfect example. People responded to that quickly.

I think the problem today is we're still seeing the same generic lead forms. People are skipping that whole entire process and landing on the lot. Huge opportunity there IMO.

 

[Jason@nabthat]

My 2 cents: I think if something is working, it's going to be pretty obvious. "Get e-Price" is a perfect example. People responded to that quickly.

I think the problem today is we're still seeing the same generic lead forms. People are skipping that whole entire process and landing on the lot. Huge opportunity there IMO.

I would like to add that A/B testing for any site (as long as you get more than 3,000 uniques per month) should grant some good insight as long as you keep the testing to simple modifications, i.e. changing 1 button (name or color), etc. Additionally, Optimizely and others give you the ability to be very targeted in the A/B testing you want to do; therefore you can do experiments only against iPhone users or all Desktop or repeating visitors.

Chris - Regarding the "Get e-price" button or any button on a dealerships site, they are not very intuitive and do not convert as well as they should; normal conversion we see from these buttons is approx .50%-1% of unique visitors. This is not good enough (I believe it is terrible), you have people that are coming to your site for a purpose; schedule service, inventory (new/used) and the only thing your site should be "great" at is converting!

Our lead button can be installed in any dealership site (any vendor), completely customizeable, can be pushed into OEM's and converts! Normal increase is at minimum 35% (low traffic dealers), but most expect an increase of 50% to 2x. See below:

 

[Chris Cachor]

Jason@nabthat I like the idea. I think this is really a much better "Get e-Price". One thing that stands out to me (and I think shoppers) is the branding makes it feel gimmicky. I feel like if there was no "Powered by" branding that it would feel more authentic and less lead-gen. I don't know if authentic is really the right description but hopefully you get what I mean.

 

[Alexander Lau]

interesting software
http://perq.com/auto-dealer-software

 

[Devorah Wolf]

I'm also wondering about the effect of multiple CTAs in one space: unlock, get pre-qualified, and all the options on the left (request information, schedule a test drive, etc)- makes me wonder if too many options lead people to do nothing. Or maybe they are more likely to convert on the eye-catching red buttons?

 

[Alexander Lau]

I'm also wondering about the effect of multiple CTAs in one space: unlock, get pre-qualified, and all the options on the left (request information, schedule a test drive, etc)- makes me wonder if too many options lead people to do nothing. Or maybe they are more likely to convert on the eye-catching red buttons?

Surely, there very well could be too many. Clean and simple usually work over complex and too many options.