The Hidden Hell of High Relevance Scores

Let’s say your Meta ad gets a 9/10 quality ranking.

Your engagement ranking is “above average.”

Your CTR is through the roof.

So why aren’t you scaling?

Why is your ROAS stuck in the mud?

Simple: You optimized for applause instead of action.

When the scoreboard becomes the strategy, you lose the game.

High relevance scores feel like a win. They make you think you’re crushing it.

But, you're just building ads that feed the algorithm’s ego…

Not your bottom line.

Why? Goodhart’s Law:

Meta says, “Here’s a score.”

You say, “Let’s make that score higher.”

And now your creative team is crafting TikTok bait — not conversion machines.

Ads that go viral and ads that sell are not the same thing.

Wanna know what high-relevance ads often have in common?

• 🧃 Scroll-stopping, but product-ambiguous

• 😂 Funny, but not felt

• 📣 Broad, but not believable

• 🔄 Clickable, but not converting

These ads aren’t failing because they're bad.

They’re failing because they’re trying to please the platform, not persuade the buyer.

Reminder: The algorithm doesn't care about your margins.

It rewards behavior — clicks, views, time-on-screen.

It doesn't know (or care) if that click turned into a $220 purchase or a bounce back to Instagram Reels.

That’s your job.

Here’s the fix

Instead of asking:

Ask:

Great ads don’t chase scores.

They trigger decisions.

TL;DR

• Relevance scores are an input, not the outcome.

• Optimize for impact, not applause.

• You can’t deposit engagement into your bank account.

Best,
Peter Delle