Digital ads don’t have to suck.

On an early Friday night about three weeks ago, still in the office, I found myself pretty dejected by the downright shitty performances of my various Facebook ad campaigns. I thought they were pretty neat — clear value props, visuals that accentuated the copy, tight targeting. But still, underperforming. That’s when I called up my buddy Alex (he’s the CEO of Audiokite and a sly little direct marketing bastard) for some pep talkage.

We talked about love, life, and the pursuit of low cost-per-clicks. That’s when he dropped some seemingly simple knowledge on my ass that I had somehow managed to overlook this entire time:

It’s all about creating a thumb-stopper.

People are flicking through Facebook like maniacs, pausing on content that catches their eye in a split second. Something out of the ordinary, unexpected, interesting.

But wait, Dave, people HATE advertising! It’s interruption marketing, it’s scammy, it’s gross. And in today’s social media powered world, nobody trusts companies blasting their logo all over the place, broheem! I wish you weren’t so dumb.

Or maybe not. After my chat with Alex, I spun into a frenzy of push-the-envelope brainstorming.

My theory: If you create ad content that is funny, unique, and unexpected, people will stop and click. Basically, play that word of mouth game I grew my early career on.

Before we get to the two ridiculously successful ad campaigns I ran, may I persuade you with a couple examples from bigger brands? Check out this video ad I ran across while perusing YouTube for ab exercises:

https://www.youtube.com/watch?v=LztFgqvFjX8

Funny stuff, right? The video above is the commercial hosted on YouTube by Geico. IT’S VIEW COUNT IS FUCKING 14.8 MILLION!! FOR AN AD. People searched for a fricken ad! Seriously. That concept is insane. And awesome. It’s awesanesome.

Here’s another one:

Guess where I found this one? On a user-generated reddit thread, calling it “brilliant”:

Ok, enough about everyone else, now onto me. Our first foray into weird-ass ads was on reddit, because where else? Here’s what the ad said:

Kind of goes against messaging convention, doesn’t it? No value prop, no benefits, no catchy tagline. Just a dare, a reason for the viewer to ask themselves, “What the hell is going on here?” It’s an attention grabber, one your brain is not used to seeing in that spot. Then came the landing page:

So, yeah, we went full bozo on this one. Swear word in the title? CHECK. Second swear word in the title for good measure? CHECK. Blatant transparency? CHECK. This ad/landing page combo was BY FAR the best reddit campaign we’ve ever run on the platform, and we’ve run a lot. Traffic to the page and conversion rates were at least double our averages on reddit.

Then we went back to Facebook, a much bigger mountain to climb. That’s when it happened. I remembered a random conversation I had with my co-founder, Mike, many months back. That conversation, which was better than any conversation you had this week — maybe even year — resulted in the purchase of one “www.woodyharrelsong.com” domain. We spun it into a microsite in which Woody Harrelson expresses his displeasure with the viewer’s music choices.

Using our reddit ad as a jumping off point, we unleashed a series of Woody-themed Facebook ads that, again, teased and enticed the viewer way more than pitched or sold them.

Guess how these ads performed compared to all my other Facebook campaigns? Yep, top of the heap. Completely overshadowed all product-first ads.

Granted, not everything with this approach is all rosy. There is a potential pitfall in attention grabbing creative that teases the audience; poor targeting that doesn’t convert. We faced this with the first round of Woody ads that resulted in a ton of traffic but crappy conversion rates. The answer? Much more art than science. We tweaked messaging, copy, visuals, and we optimized the target audience over and over and over. There’s a bit of click bait potential here, so just be careful.

So go forth and be weird. (Unless you’re in healthcare or fintech or something. Ah screw it, try it there, too.)