"Is Facebook listening to my conversations?" and other thoughts about "creepy" digital ads

Is Facebook listening to my conversations?

We just bought a new car for my wife, a small GM SUV. But what happened the week before was one of those “creepy” moments that lead to questions like “Is Facebook listening to my conversations?” (Which, incidentally, is the first item that came up in my Google search result autofill, so I’m obviously not the only one who has wondered this.)

The backstory is this: one Sunday afternoon, I suggested to my wife that we go to look at some SUVs on the dealer lots, since I knew there were no salespeople to accost us on a Sunday. We left the house and drove to the local “motor mile,” discussing the available options. Neither of us did any online searching, or opened up any dealer websites.

We arrived at the GMC dealer, browsed the inventory, and found a vehicle we liked. Next, we walked a short ways to a Buick dealer, agreed that we weren’t happy with their options, and then got in our car and drove a half-mile to a Chevrolet dealer. After reviewing what they had for sale, we agreed we liked the GMC option the best, and that I would call the dealer on Monday to discuss a test drive. Again, no apps were opened or searches commenced.

When we got home is where things got “weird.” We had quit talking about the cars, and my wife had left the room. I opened up my home computer’s browser with the intention of opening a new tab to find out details about the GMC dealer. But I got distracted and instead refreshed a page I already had open.

And it happened. The three ad blocks on that particular page repopulated with . . . wait for it . . . a GMC ad, a Buick ad and a Chevrolet ad. The GMC and Chevrolet ads were actually for the exact vehicles we looked at, and the Buick’s ad was for a close cousin of those two.

How could this have happened? The tin-foil-hat-wearing folk among us might conclude that my wife’s Facebook app heard us discussing these three options, and then when we returned home and the phone attached itself to our wifi network, it identified our network as a target for ads. Since I had the day before used Facebook on that desktop computer, then the ads were served thanks to some sinister and Orwellian plot.

However, Facebook has insisted through Forbes,  Wired, The Telegraph and others that it does not use your phone’s microphone for adverting purposes. According to Rob Goldman, Facebook’s VP of Advertising, “We don’t — and have never — used your microphone for ads. Just not true.”

So, if Facebook isn’t listening, what’s the explanation? It’s far too specific to call three ads for the very three dealers I visited — and the specific models we looked at in two of the cases — a “coincidence.” So there definitely is some sort of tracking going on here. What is it? And, if it’s unavoidable, can we use it to our advantage as marketers?

If you believe Facebook’s denials, then most likely explanation is this: Google has guessed that we owned an older small Chevrolet SUV. I’ve done plenty of searches for small parts for repairs, or tires, or for insurance quotes, for this model for them NOT to identify that piece of information with my IP address.

The Google Maps app on my phone also knew that we went to GMC, Buick and Chevrolet car dealers. Although we didn’t use Google Maps to get to the dealers, my location services settings do not restrict Google Maps from knowing where I was. The ads on that page on my home browser — most likely served by a service that uses Google data — put two and two together and said, “Hey, this guy is looking at GM vehicles to replace the small SUV he has right now; let’s show him the ad we have in inventory for the most likely equivalent.”

Another possible explanation? That these dealers had geotargeted me — they knew I visited their dealerships (or even their competitors), and they had ads served to me when I got home. (And, just to add fuel to you conspiracy theorists out there, our media and sourcing director, Pamela Wilcoxson, says that she HAS had some conversations with media reps that have told her they ARE beginning to allow targeting based on device listening — just not Facebook, presumably.)

So how can this work in the B2B space? A few suggestions:

1. Use geo-targeted advertising to reach specific demographics in a specific geographic location — we did this for engineers at a trade show. Our client had a new product they were launching at a big show in a major city, so we tried to reach those engineers through ads on their phones while they were within a specific distance of the show location.

2. Use device retargeting to serve ads at home for products your audiences have searched for while at work. Media companies know which devices a particular user of Facebook, for instance, logs into, so builds a catalog of devices they are likely to use. You can use that to reach your targets while they are less likely to expect an ad from you, and possibly be more inclined to remember or act on it.

3. Use search retargeting, where you target those who have searched for one of your competitors. It can be even more effective if your audiences see an ad for your product several hours after they’ve searched for your competitors, again, when they aren’t expecting to see it.

4. Use delayed targeting. Again, we helped one of our clients identify prospects attending a large annual event. Rather than be part of the advertising “noise” at that event, we waited until those prospects returned home to their jobs, and THEN we targeted them with advertising for our client.

It all sounds a bit creepy, but the goal is to serve audiences information about products or services they will actually find useful. Frankly, I’d much rather see an ad for a vehicle I’m considering buying, or a marketing service that might be useful, than, say, baby formula or needlepoint supplies. Creepy? Or beneficial? It’s actually probably some of both.

To learn more about how digital advertising can work for you, check out our programmatic advertising infographic.

Add Comment

Sign-up for Special Content

Gain unlimited access to case studies, white papers, and more.

This feature requires that cookies are enabled in your browser
Real Time Web Analytics