Mystery Solved: Social Media That Drives Walk-in Traffic

Hey everybody, Kevin Oakley from DoYouConvert.com here thanks to Kimberly Mackey and New Homes Solutions Consulting to talk to you about my session at the Builder Show that I gave at Sales Central called Mystery Solved: Social Media That Drives Walk-In Traffic. Here we go! We only have a few minutes to cover a lot. $76.69 per walk-in traffic unit. That is a number I can see and show and prove thanks to

Facebook’s attribution system, and the way that it works, is it uses what are called offline events; events that do not happen online – like walking in, making a phone call, text messaging. You can see right here on the screen where you go in your Facebook Business Manager account to upload that data from your CRM. And what it does is takes the data from walk in traffic; you upload that data – the first name, last name, phone number, email address. And it looks for Facebook accounts that match those criteria, and then when it finds them it says, “Did that individual interact with, like, share, comment, click on an ad?” This is not an impression, but actual interaction. Did they interact with that ad prior going to the sales office and registering? So this is not a cost-per-traffic walk in unit all nice wrapped up in a bow. What it is is part of the customer’s journey, and it’s proving that, in this case, for $76.69 that we were part of that journey on social media. Now, there’s more to it than that; the other parts we covered were just how smart the AI is and why we can get a number like 76 dollars in that scale with large amounts of walk in traffic.

We used to run Facebook ads like this: You had to make very small targeting, and the whole idea was to try to get the best 120,000 / 100,000 people in a specific geographic area. And we also us specific targeting information like this: length of residence, likely to move, Zillow, some people who looked at Zillow before. That’s all gone, and we don’t do that anymore thanks to, partially, this man. (Vladimir Putin.) Ever since the last election, there have been real concerns about privacy, and access to information, and it’s been going on for a while, but it’s really heightened by this last election cycle. So what Facebook has done, is they’ve taken all of these specific targeting options away from us as marketers – for the most part. A couple of them still remain, but the best ones are gone. So now what we have to do is trust in the AI system to take our hands off the wheel as marketers – which is scary – and trust that the AI system could work. It does take programming, and knowhow and set up, but here’s what can happen once you do tell the AI specifically what you want. This is a builder that we work within the Atlanta market, and this was the original campaign setup that we had at the beginning of this year/end of last year. Targeting as you would expect, the Atlanta market. And, pretty good results on its own. 32 cents per click, 48 percent bounce rate, 2.25 pages visited per visit, 1:40 of time spent on the site. This is all mobile traffic, so again, these numbers are pretty good considering mobile and cost. But we wanted to continue to test it out, so what we did is take the exact same ad creative and target the entire state instead. And you know what happened? It got better. The cost dropped, which was not too surprising because we were targeting a larger area, giving more people for the AI to choose from. But what was surprising, especially here on these numbers, was the two minute session time that they visit. This is a 30-second increase from that first campaign. That shouldn’t happen. We should get lower quality people unless the AI is really, really good. So, how do we test that it’s really, really good? We added two additional states to the targeting this time. Looking at the analytic data in this builder’s dashboard, we said, “In the past, you’ve gotten a large amount of quantity from these additional two states, let’s add them in the mix.” Again, the exact same ad copy, same images. The only thing that changed was that we gave the AI the freedom to show them to more and more people. Here’s what happened: The cost went back up a little bit to 34 cents per click, but all the other metrics continued to improve slightly, and the time on the site difference is dramatically different; especially when you consider that original campaign only targeting the Atlanta area. From that campaign to this one, we have a 54 percent increase in time on site.

The AI is getting smarter. So, there are two important takeaways from this session we gave at the Builders Show:

One is that you absolutely should be using your CRM, not just to create look-alike audiences, not just to help understand the AI who’s already been to your site, but to show value in the attribution in the customer journey by uploading and using that offline events tool.
The second part is getting the AI involved.

That’s it! A brief summary of my session at this year’s International Builders Show. Thanks again to Kimberly Mackey and New Homes Solutions Consulting. I hope to see you again soon.

Kevin Oakley has over 15 years of experience running marketing and sales operations for home builders of all shapes, sizes, and areas of expertise from entry level to multi-million dollar homes. He has worked for two different multi-billion dollar revenue builders as well as for a private family-owned builder before becoming a consultant. He has been deeply involved in the development, marketing, and sale of well over 2 billion dollars in new homes and speaks regularly at the largest events in our industry.
Kevin is the Managing Partner at Do You Convert, author of Presale Without Fail: The Secret to Launching New Communities with Maximum Results, and co-host of the Market Proof Marketing podcast.