The feedback nobody reads

Most businesses that collect customer feedback have a spreadsheet somewhere. It fills up slowly and steadily. Nobody reads it.

The form exists. Customers answer it. Responses arrive. Not much happens. This isn't unusual -- it's the standard outcome for most feedback setups running more than a few months.

The problem is rarely the form, the technology, or the volume of responses. Collecting feedback and using feedback are two different activities, and most organisations only ever build the first one.

Setting up the form feels like progress. You've done something. Data is flowing. When someone asks whether you're listening to customers, you can say yes. What you can't say with any confidence is what you actually learned in the last quarter, or what changed because of it.

Feedback arrives unevenly and without structure. Ten responses one week, forty the next, three the week after. Some are detailed. Most are a rating and nothing else. Individual responses are easy to read. The pattern across several hundred is much harder to find if you're not deliberately looking for it.

So people read the ones that catch their eye -- the very positive, the very negative -- and treat them as representative. They aren't. The one-star and five-star responses are each legible in their own way. The three-star responses, the mild frustrations that didn't motivate anyone to open Google Maps on their own, tend to be where most of the actual signal lives. And those sit unread somewhere in the middle of the spreadsheet.

Acting on feedback also requires making a call. Which pattern is a real problem? Which complaint is one person having a bad day? These judgments are slow and uncertain, and they take time that's usually being spent on something else. So responses accumulate. The calls don't get made. At some point the spreadsheet stops feeling like a resource and starts feeling like a backlog -- a pile of responses you're not sure what to do with.

There's also a structural issue: most feedback arrives without an obvious home. Reviews go to Google. Support tickets go to a helpdesk. Feedback form responses go to a folder, or a spreadsheet, or an inbox -- wherever they land, usually disconnected from the system where decisions actually get made. So the information doesn't travel. It arrives and stops.

What most feedback systems are missing is what comes after collection: a way to move from "here are 400 responses" to "here is what's actually worth looking at." Without that, you're not really listening. You're recording.

The responses that tend to matter most aren't usually the strongest opinions. They're the complaint that keeps appearing across different customers in slightly different words -- mild enough that no individual person felt strongly enough to escalate it, but consistent enough to show up thirty times in three months. That's a pattern, and patterns are what you can actually do something about.

Most unhappy customers won't say anything without being prompted, and the ones who respond to a feedback form are already a fraction of the people with something to say. Reading through them in the order they arrived, treating each as its own story, is a reliable way to miss the signal running through all of them.

Qria reads through the open text and surfaces what keeps coming up: recurring themes, questions that consistently score lower, places where the responses have shifted over time. You still decide what to do with it. But finding the pattern isn't something you have to do by hand.

There's a difference between a business that collects feedback and one that actually learns from it. Usually the gap is in what happens after the responses arrive.