YouTube and video ads

How the video is built

1. Open with a hook (a question, a pattern interrupt or a joke).
2. Agitate the problem.
3. Show how your product fixes it in a way the others don't.
4. Compare it to what they already use.
5. Entertain them enough that they don't notice they watched four minutes.

We reverse-engineered that from the Harmon Brothers work at Purple, then built it in-house. The write-up is How a video ad is built. At Groove the version that kickstarted the growth was a question. "Have you seen this ring?"

My favorite test started at a conference. People would ask where I worked, I'd say Purple and get a blank stare. Then I'd say "Goldilocks smashing eggs" and they'd light up. They knew the video. They had no idea what we sold.

So we put the logo in the bottom right, where the eye goes looking for the skip button. Nothing else changed. Brand search went up 2,824% for people who saw it, compared with a hold-out group.

I was the original Director of Marketing at Purple, later Director of YouTube, Search & Display. I went back from May to September 2026 as VP Marketing & Demand Gen.

Don't match the other ads

In 2014 my kid Giovanni sorted passing cars into blue and not blue. Ads do the same thing to people. If every headline is the keyword stuffed into a sentence, you're the blue car. The vlog is Don't look like every other ad. On Meta today, your creative does a lot of the targeting for you.

You don't even have to do a big Harmon Brothers production. The wider system is What performance branding means.

Questions

Why YouTube?

You can reach people who searched the category in the last 30 days. And if they skip, you still got about five seconds of brand impression you didn't pay for.

Do we need a huge production?

No. Hook, problem, the unique fix, a comparison, and enough craft that they stay. A phone can do it when the product is real. "Have you seen this ring?" was shot in-house.

How do you know the brand work paid?

Brand search, and blended MER. The Purple logo test produced a 2,824% lift in brand search for viewers versus a hold-out.