Every driver on your last Uber is terrified of a 4.

Not a 1. Not a 2. A 4. Because on a scale that runs from one to five, a 4 doesn't read as "good." It reads as "something went wrong and nobody's saying what." Drivers know it, which is why so many of them ask you, gently, to consider a 5 before you get out of the car. The scale hasn't measured driving quality in years. It measures whether you were polite enough to lie a little on the way out.

Restaurants have the same problem from the other direction. Plenty of places on Google are sitting on a 4.6 built from five 5-star reviews and one furious 1-star from someone who couldn't find parking. The whole apparatus of the star rating, the thing that was supposed to tell you whether something is actually good, has quietly stopped doing that job. It only measures one thing now: whether enough people could be bothered to leave a review at all, and they only bother when they loved it or they want blood.

Kids already fixed this. We just haven't caught up.

Which is why kids have had to invent a new scale on top of the broken one.

My lads, Oscar and George, will tell you a film is mid with the same flat certainty other generations reserved for "terrible." Ask either of them what a 3-star means and neither will say "average." They'll say it's mid. And mid, in their world, is not a polite way of saying fine. Mid means it let you down. Mid means don't bother.

On a scale that's been grade-inflated into meaninglessness by adults who only rate in extremes, the middle has become where bad things go to hide. Nothing merely competent survives the two options everyone actually uses. So a whole generation built a second, unofficial scale to say what the official one can't: this was fine, and fine isn't good enough.

Why our score looks harsh next to the app store version

I bring this up because Stacked runs a scoring system on our marketplace too, and it looks harsh next to the app store version of five stars.

Operators don't get a prompt asking them to rate their tech provider after a great demo, primed to be generous before they've actually run a busy Friday on it. We don't ask on day one, when everyone's still on their best behaviour. We ask properly, away from the vendor, once an operator has actually lived with the tool for a few months. Through the till crashes, the support tickets, the "why has this integration stopped talking to that one" moments that never make it into a sales deck. So the scores come out lower than you'd see anywhere else, because they're real.

Look across our marketplace and you'll see it: the scores cluster in the middle. Not because the tools are mediocre, but because the middle is where honest, unprompted feedback from operators who've actually used something for months always ends up. A vendor sending out a survey the day after a successful install will always score higher than we do. That's not a flaw in our system. That's the whole point of it.

What a 3 actually means on our marketplace

A 3 on our system isn't a warning sign. It's an operator telling us, unprompted and unpaid, that the tool does what it says on the tin and they'd tell a mate to use it. That's not mid in the TikTok sense. It's not a below-average score dressed up as something kinder. It's the honest version of good, on a scale that hasn't been grade-inflated by vendors sending it out at exactly the moment someone's feeling generous.

We don't show a score at all until enough real operators have actually said something, because one furious review from someone who never got their integration working is exactly the kind of noise the rest of the internet already drowns you in. What's left, once you strip that out, is closer to what a trusted mate would tell you over a pint than what a star rating usually gives you.

The incentive nobody talks about

So when you see a partner on our marketplace sitting in the middle of the scale, that's not us being stingy with the gold stars. That's what a functioning rating system looks like once you take away the incentive to lie.

Vendors can pay Stacked to be seen on the marketplace. They cannot pay to be recommended. Those are two different products, and keeping them separate is the whole reason the model works. A partner's listing tier buys visibility, a slot, a badge, a spot in the right category. It buys none of their score. The score comes from operators who have nothing to gain from being generous and nothing to lose from being honest, which is precisely the group every other rating system on the internet has quietly stopped listening to.

Same number, different measurement

Somewhere a Gen Z reviewer would look at a 3 on our marketplace and assume the product's a letdown, because they've been trained by app stores to read the middle of the scale as a red flag. An operator who's actually run the tool for six months would tell you it's one of the good ones, the kind you don't think about because it just works.

Both of them are looking at the same number. Only one of them is looking at something that was built to mean anything at all.

Want to see how a partner actually scores, and why? Browse vetted operators-first partners across POS, workforce, and inventory in the Stacked marketplace, or get in touch if you want a second opinion before you sign anything.