Method · Found in the wild

Nobody checks the windows online.

Retail has a discipline for this. Online mostly doesn't have one at all — and you can see the gap in the wild, if you know where to look.

The walkthrough retail does every morning

Physical retail has a discipline nobody has to be talked into: the store walkthrough. Check the windows as you walk in. Check the POS was actually put up the night before. Clean, straighten, look at the shop the way a customer arriving in five minutes will see it. Nobody debates whether this is worth doing. It's just what opening a shop means.

I've sat in a lot of agency meetings and performance reviews since. Numbers get discussed in real depth — ROAS, CAC, conversion rate, attribution. What almost never comes up: has anyone actually gone and looked at the thing being paid for, the way a customer would encounter it?

What that gap actually costs

The pattern shows up constantly once you start looking for it. Promotions that go live but were never actually tested end to end. Pages missing entirely. One I've seen more than once: delivery estimates wrong in Google Ads — a configuration issue showing 10+ days — while the website itself, sitting one click away, correctly says 3. Nobody caught it because nobody walked the actual path a customer takes from ad to checkout and read it as a stranger would.

None of this shows up in a dashboard. ROAS doesn't drop because your delivery estimate contradicts itself across two pages. It just quietly costs you trust, and eventually a sale, in a way no report will ever flag.

Found in the wild

This ad came up in my own feed. It's a real, live, currently-running ad — not a hypothetical.

A Pitchup.com Facebook ad showing an unfinished, empty caravan pitch site with construction materials visible, captioned 'Browse our range and find your next outdoor holiday'

A real, currently-running ad — screenshotted as encountered, not staged.

The copy says: "Browse our range and find your next outdoor holiday." The image shows a bare concrete pad, roped off with what looks like leftover fencing rope, in front of a half-finished building and a row of terraced houses. No tent. No awning. No golden hour. No sense of an outdoor holiday at all — it reads like a builder's progress photo, because that's almost certainly what it is.

This is very likely one image pulled from a large shared asset library, dropped into a template, and never looked at again by a human asking the one question that matters: does this make someone want to book a camping trip? It probably works fine for other sites in the same feed with better photography. For this specific listing, it doesn't sell the thing it's being paid to sell.

How this fits the framework

Pitchup.com — WithAire 251 listing ad

Scale

GOAS Score: 3/10

ShouldMake a stranger want to book an outdoor holiday — that's the entire job of a top-of-funnel ad.

Was actuallyShowing an unfinished building site with no camping cues at all — actively working against the booking it's paying to generate.

Red flag to checkNobody on the team had looked at this specific ad, for this specific listing, the way a stranger scrolling their feed would.

The job was never in question — a listing ad is Scale, plainly. The score is what a customer-eyes check would have caught immediately, and no ROAS dashboard ever will, because the ad might still be technically "performing" on click-through cost while quietly converting worse than it should, with nobody able to see why from the numbers alone.

The fix is genuinely simple