Hype(r) Targeting Ads

Does Not Seem Very Selly To Me

There's an odd little prank that's been making the rounds of Reddit recently. In it, a marketing pro does a concept exercise to see if he can do a Facebook campaign to an audience of one -- that one being his roommate. Using personal information of issues and interests about his target, the pro makes copy-specific creative and launches. Said roommate notices the ads, posts to his stream about how creeped out he is by this, and pulls out of Facebook before the prank can be revealed.

The short lesson from this is, of course, do not take a marketing pro for a roommate. Or, at least, not this one. The longer lesson is a little more actionable for our purposes.

Some background. I've been at three different high impact adtech companies, and while all of them were smart enough to avoid personally identifiable information (aka, PII, or the thing that ad tech companies never want to have, for reasons that Rhyme With Beagle), that didn't prevent us from making strong claims about how great our data was. And, by the power of inference, how you would have to suffer with less if you used our competitors.

Need to reach someone the instant that they were about to buy from a competitor? Three start ups ago had just the thing, with 100% deliverability and viewability in a proprietary ad format. How about doing the deed in dedicated emails where you were more polite, but had high control on frequency? That was two start ups ago, with high legal compliance and all kinds of auxiliary programs to make sure you were on the side of the angels. Want to not just reach, but deepen the lifetime value through showing additional SKUs that were certain to delight your prospect and make the cart size bigger? That was the last start up, and those recs would follow you around the Web like a bloodhound if the spend was high enough. And so on, and so on. Ad Tech Land is not exactly shy about telling you how truly wonderful their data is, and how the targeting makes a marketer's life far more lucrative and rewarding.

And yet, as you might guess from my current professional standing of not working for any of those guys, many campaigns at all of these stops would underperform, especially when you factored in ROI metrics that take into account higher spends to reach that juicy audience. How do you fix it? There are a lot of ways, from simple analysis of creative to see if the blocking and tackling (calls to action, entry points, etc.) was up to snuff. Testing the offer to see if it was truly right for the brand. Varying by daypart, platform, recency, and so on. Changing bid levels or payment methods. There's a lot of options, but sometimes targeting is not enough. Brand, offer and list trump creative, and always will.

But if you are very far apart from an acceptable ROI threshold on v1, for whatever reason, you'll never get the chance to iterate your way to tolerable. No client wants to hear that the things that can't easily be fixed -- their brand and, likely offer -- are at fault more than the targeting. And before you fold your tent and just blame the list anyway, some clients will try to complete the Hail Mary pass by making the art "stronger" with downright creepy and over-the-top targeting copy or imagery.

So, for the benefit of those about to waste good time after bad... your potential future customers do not now, and have not ever, cared about your efforts to put together a great list. Telling them about such things is pointless at best, and when you combine that sort of thing with cyber-creepy copy ("Still interested?" "We miss you!" "You forgot this!"), it not only irritates the prospect, it can also harm the brand. Oh, and it doesn't generally work in control cells against other executions, and if you do this kind of thing often enough, you can easily train a client base to wait until after a cart abandon to get a margin-crushing price. It's not easy to make retargeting a bad idea, but it can be done.

I've had any number of clients that felt compelled to tell the prospects of how special the targeting was to reach them, or how long it's been since they were on site, or why they are getting this offer now. Sometimes, they even go through with it despite my strong recommendation not to, and at that point, we're in the realm of my special Law of Dumb Clients. Which is: "If a client is going to do a dumb thing, and you've told them why it's a dumb thing, and they insist on doing it anyway... do the dumb thing *quickly*, while getting their payment ASAP, because they will be out of business soon enough from doing dumb things."

Anyway, back to the creepy text and selects. Beyond the ethics and lack of efficiency, if your list choices are tight enough, you can't scale. (And congrats, by the way, on finding an online advertising method that doesn't scale. That takes talent.) If you tell the prospect what you know about them, you are wasting everyone's time and being off-putting.

Instead, give your super-targeted list a different offer. Make the creative execution more about the SKU in question. Add a one-time coupon code and clock to redemption. Give them a price break at higher spend levels. Link to testimonials, social media plays or other reasons that would get someone over the last mile. And give them the option of a fine sausage dinner, rather than a documentary on how it's made. You'll both eat, and sleep, much better afterwards.

* * * * *

You have read this far, so feel free to connect with me onLinkedIn. I also welcome email to davidlmountain at gmail dot com, or you can hit the quote box at the top right of this page.

In addition to copywriting, direction and strategy, we also provide design, illustration, photography, coding and hosting. Tell us what you need done, and your budget, and we'll work out an RFP.