Every time I speak at an industry event, there's a moment I've come to watch for
By Michael Erickson Facchin · August 7, 2026 · Curated by George's Blog
Every time I speak at an industry event, there's a moment I've come to watch for. It's rarely during the part of the talk I prepared the most.
It's usually a throwaway line, something like: most brands have no real idea whether their AI bidding tool is making good decisions or just making fast ones and that's the moment the room actually leans in. Not during the bid strategy slides. Not during the case study. That one line, every time.
That's the tell. That's the anxiety sitting under this industry right now that almost nobody says out loud from a podium. Automation got adopted faster than trust in it did. Manually managing bids at scale stopped working, so brands turned the algorithms on reasonably but a lot of them never built the habit of checking the algorithm's work. So now there's this quiet, widespread discomfort:
"I don't actually know why my account is doing what it's doing this week, I just know it's doing something."
I don't think that's a failure of any specific tool. It's what happens whenever a technology shift outruns the muscle of verifying it. We did the same thing with dashboards a decade ago trusted the number on the screen before we built the habit of asking where it came from.
The brands that seem most at ease, in every room I've been in, aren't the ones with the fanciest automation stack. They're the ones with a simple, boring habit: someone, on a real cadence, actually looks under the hood. Not to override every decision, just to be able to explain, in plain language, why the algorithm did what it did.
Nobody's put a clean name on this trend yet. It's not "AI adoption." It's trust calibration figuring out how much actually needs a human eye versus how much you can genuinely let go of.
That's the pattern that keeps repeating in these rooms, no matter which stage I'm on.