Science rules in digital marketing.

Social media is a place of vanity. Those of us who use social media often get to see flattering images of just about everything - people, places, food, cars, whatever. Have you ever seen a picture of a friend posted as their new profile picture and thought, "Wow, that's a good picture of them."

One might believe that the same holds true for automotive ads. On websites, it has been widely accepted that real pictures of inventory work better than stock photos, but on social media we have access to the gorgeous pictures that are supplied by the OEMs. Will pretty advertising pictures outperform pictures of live inventory on ads that are sending traffic to the vehicle details pages and search results pages?

We have done a ton of A/B testing over the past few months and we have pretty compelling data, but I want to get the opinion of the community here before posting those results. What do you think?

Here are some of the criteria for a test we ran for a Hyundai client:

• All ad copy had the same titles, status text, and link description
• The ads linked to the search results page for new Hyundai vehicles
• They were targeted at intenders - people within driving distance to the dealership who had indicated they intend to buy a new Hyundai in the next 180 days
• The only difference was the image

We ran two concurrent campaigns for 1 month. One had beautiful images pulled from the OEM. The other had live inventory images. Here are samples with the branding omitted:

Which type do you think got more clicks to the dealer's website?