Stop Asking "Are You Satisfied?" Start Asking "Are You Making Progress?"

Every quarter, the customer success team runs the surveys. The Customer Satisfaction Score (CSAT) comes back strong at 4.2 out of 5. Net Promoter Score (NPS) is healthy. The promoter count is climbing. The slides get made. The executive team nods. "Customers love us."

Then renewal season arrives and three accounts that scored high on NPS don't renew. The reason given on the exit survey: "Going in a different direction." Which tells you precisely nothing.

So what happened? The surveys weren't wrong. They measured what they asked. CSAT captured how individual interactions felt, and those interactions were fine. NPS captured whether they'd recommend the product, and they would have, at the time they were asked.

But neither survey asked the question that would have predicted the non-renewal: is this product making your job easier than it was before you started using it?

That's a different question. CSAT measures the touchpoint. NPS gets closer to the outcome, as someone who'd recommend a product is probably getting real value from it.

But NPS gives you a score without telling you why. A 9 could mean "this product saves me two hours a week" or "the interface is pretty." You know they'd recommend it. You don't know what progress the product is delivering, which dimension of the job it's helping with, or where it's falling short.

A progress survey helps fill that gap.

The Scope Gap

Calm is a good example. After any given meditation session, a user might rate the experience highly — the interface is lovely, the narration was soothing, the ten minutes felt worthwhile. That's real. The session was good. If you asked a CSAT question right then, you'd get a high score.

But the question that predicts whether they renew is a different framing: "Am I meditating more consistently than I was before I had this app?" That's a progress question — and it's the one CSAT wasn't designed to ask.

A progress survey is built on a before/after comparison. It doesn't ask the customer to rate you. It asks them to judge change in their own ability to get the job done.

The core question: Compared to before you started using this product, are you getting the job done faster, more easily, or more confidently?

That shift does something neither CSAT nor NPS can do on their own — it connects the measurement to the customer's world. Their work. Their pressure. Their definition of "done." Not your features, not your interface, not your support team. The outcome they hired the product to help with. NPS tells you they'd recommend it. A progress survey tells you why, and where the product is falling short.

And progress isn't just one thing. For some jobs, progress is speed — getting it done faster. For others, it's confidence — feeling more certain the output is right. For others, it's quality — fewer errors, less rework, fewer "wait, are we sure?" moments.

JTBDUX evaluates experiences through a set of questions that connect to these dimensions directly: does the experience match how the job naturally unfolds (speed)? Does it reduce anxiety at the moments anxiety peaks (confidence)? Does it show the user clearly enough what's happening that they can trust the output (quality)?

Pick the one or two dimensions that represent your product's promise — the reason someone would switch to you — and measure those. Not a general satisfaction score. The specific kind of progress your product was hired to deliver.,n

The Scariest Answer Is "No Change"

Progress surveys produce a result that looks different from CSAT and requires different interpretation.

A high progress score means: we're helping them get the job done better than they could before. A low progress score means: the product isn't making the job easier, faster, or more reliable.

And "no change" is not neutral.

"No change" means the product hasn't made a measurable difference to the outcome. The user might not be dissatisfied — nothing is broken, the sessions are fine. But "fine" isn't what earns renewal. Making the job easier is. And "no change" means the product hasn't done that yet. That's a subscription one budget review away from being questioned.

This is where progress surveys become an early-warning system. Touchpoint-level CSAT might not move — each individual session still works, nothing visibly breaks. But the progress score shows the trend: the product is making less of a difference than it used to.

Why? There could be many reasons. The job's context shifted. The team's needs evolved. Competitors caught up. The product still works session by session. It just isn't moving the needle on the outcome anymore.

Touchpoint-level CSAT isn't designed to catch that — it's measuring a different thing at a different scope. Progress surveys catch it because they're asking the outcome question directly: is the product making a difference?

And the open-ended follow-up — "what's the main reason for your answer?" — is where the real signal lives. Customers will tell you what progress means to them in their context. They'll tell you what's blocking them. They'll tell you what they're comparing you against. That's job-friction data — and job-friction data tells you what to fix next.

When to Ask and What to Ask

The timing matters more than the question count. Ask after the moment the user attempts to get the job done — after completion, handoff, publish, resolve, send, close. Not after login. After the moment where an outcome of value should have been produced, because that's when the user has fresh evidence about whether the product helped.

The survey itself should be short enough that it doesn't feel like work:

One progress rating: "Compared to before, are you able to [do the job] faster / about the same / slower?" Or on a five-point scale from "much less progress" to "much more progress."

One diagnostic question: "What's the main reason for your answer?"

And segment who you sample — because progress means different things at different stages. New users tell you whether the product helps them start getting the job done. Established users tell you whether it's still making a difference. Power users tell you where the job breaks at scale.

Mix those together and your averages will look calm while your customers split into "this is essential" and "this didn't change anything."

CSAT is useful for what it measures — sentiment about the interaction. NPS gets closer to the outcome — someone who'd recommend a product is probably getting value. But neither tells you the specific answer to some very important questions: is the product making the job easier, in what dimension, and where is it falling short?

A progress survey does. It's the instrument that connects measurement to the job — not "how do you feel about us?" but "are you getting this done better than before?" That's the question that tells you what to fix, what to double down on, and whether the product is earning its place in their workflow.