Why Your Customer Feedback Score Is Lying to You. Let's check your customer feedback questions.

This was originally shared on LinkedIn. Expanded here with a practical guide to asking better questions.

In 2005, Bain & Company surveyed 362 companies and found that 80% believed they delivered a superior customer experience. When they asked customers the same question, only 8% agreed.

That gap — 80% versus 8% — has a name. Researchers call it the delivery gap. Most organisations call it nothing, because they never see it.

What's interesting is that the same researcher behind this finding, Frederick Reichheld, also invented NPS. The most widely used feedback tool in the world. Which is, at its core, a rating question.

We've known about the delivery gap for twenty years. And we kept building rating scales to measure it. That's not a coincidence. That's the whole problem.

The validation trap

Most businesses are not bad at collecting feedback. They're bad at asking the right questions.

"How would you rate our service?" is not a feedback question. It's a validation question. It's structured to nudge positivity. It assumes the customer had a satisfactory experience and just needs to quantify it.

You get a 4.2. You report it upward. You feel good about it.

But 4.2 doesn't tell you what almost made them not come back.

Businesses have been asking questions designed to confirm they're doing well, not questions designed to find out where they're failing. The question you ask shapes the answer you're allowed to receive.

The fix is not just asking more open questions

Open-ended questions sound like the solution. "Tell us about your experience." But they create a different problem — you collect thousands of unstructured responses that are hard to act on, and customers abandon the form halfway through because it feels like work.

The real fix is asking questions inside your circle of control.

Before you add any question to a feedback form, ask yourself: if I get an honest answer to this, can I actually do something about it?

If the answer is no — don't ask. You're collecting noise and calling it data.

What a better question looks like

Instead of: "How would you rate your overall experience?" Ask: "Was there a moment during your visit where we could have done better?"

Instead of: "Would you recommend us to a friend?" Ask: "What would need to change for you to visit us more often?"

Instead of: "How satisfied were you with our staff?" Ask: "Did our team make it easy for you to get what you came for?"

The difference is direction. Rating questions look backward and measure sentiment. The right questions look forward and surface action.

The deeper issue

Most feedback systems are designed by people who want to feel good about their scores, not people who want to find out what's broken. That's a culture problem, not a technology problem.

The tool you use to collect feedback matters less than the intention behind the questions you ask.

Collect to improve, not to report.

Q: What is the delivery gap in customer experience? The delivery gap is the difference between how organisations perceive their own customer experience and how customers actually experience it. Bain & Company research from 2005 found 80% of companies believed they delivered superior CX — only 8% of their customers agreed. The gap persists today.

Q: Why do NPS scores not tell the full story? NPS measures likelihood to recommend but doesn't tell you why a customer feels that way or what specifically needs to change. A score of 7 out of 10 gives you a number but no direction. Paired with specific, actionable questions, NPS becomes more useful — on its own it's a sentiment gauge, not an improvement tool.

Q: What makes a good customer feedback question? A good feedback question sits inside your circle of control — meaning if you get an honest answer, you can actually act on it. It's specific enough to surface a real insight, forward-looking rather than backward-rating, and short enough that a customer will actually answer it.

Q: How many questions should a customer feedback survey have? As few as possible. Research consistently shows that response rates drop sharply after three to five questions. One well-designed question answered honestly is worth more than ten rating scales completed on autopilot.

Q: How does QRate approach feedback question design? QRate uses rating scales and single-select questions as the primary format because they're fast, mobile-friendly, and produce data that's easy to act on. The platform is designed around the principle that the right question at the right moment captures more signal than a long survey sent after the fact.