Field Notes 01 · MEASUREMENT

Compared to what?

A satisfaction score with no reference point isn't a measurement. It's a number.

A regional manager opens her Monday report. One of her stores has come in at 4.2 out of 5. The cell is green. She moves on.

She has just made a decision — the decision to do nothing — on the basis of a number that told her nothing. Because 4.2 is not good. It is not bad. It is not anything at all until you say the words that almost no dashboard ever says: compared to what?

The blind spot

Most customer-experience reporting is built on absolute scores. A CSAT. An NPS. A star rating. Somebody once decided that above 4.0 is green and below 3.5 is red, and that threshold has quietly governed a decade of decisions.

The threshold is arbitrary. Worse, it is context-blind. It cannot distinguish between a store scoring 4.2 in a market where the category median is 3.6 — a genuine outperformer that deserves study and replication — and a store scoring 4.2 in a market where every competitor on the same street scores 4.8. The first store is a lesson. The second is bleeding, and it is coloured green.

Ratings are relative goods. Customers do not grade you against a rubric. They grade you against the last place they went.
The principle

The reframe

Sentiment only becomes information when it is expressed as a distance from a reference. Not a score — a divergence. How far is this store from its true peer set? From its own performance last quarter? From the category in the same city, the same week, the same weather?

This changes the shape of the chart, and changing the shape of the chart changes the conversation in the room. A gauge invites the question are we above the line? A diverging bar, drawn from a centre that represents the peers, invites the only question worth asking: who is pulling away from us, and why?

Figure 1 — The same store, two instruments
HOW IT LOOKS TODAY — ABSOLUTE 4,2 / 5 STATUS: GREEN · NO ACTION THE SAME STORE — RELATIVE TO ITS PEER SET PEER MEDIAN Peer store A+0,4 Peer store B+0,2 Peer store C−0,1 This store −0,6
Same store. Same week. Same underlying customer sentiment. Read absolutely, it is comfortably green and generates no action. Read against its peer set, it is the worst performer in its comparison group by a wide margin — and the gap is large enough to be worth someone's Monday. The absolute view didn't just under-report the problem. It actively suppressed it, because green means stop looking.

What this means for everyday decisions

Three consequences
  • Kill the absolute threshold. "Above 4.0" is not a standard. It is a superstition inherited from a spreadsheet nobody has questioned.
  • The peer set is the whole argument. Comparing a city-centre store to a motorway store is not a comparison, it's noise. If you cannot defend the peer set, you cannot defend the finding.
  • Rank the gaps, not the scores. Your attention is the scarcest resource in the business. Spend it on divergence, not on absolute position.

The takeaway

An absolute score answers a question nobody in the business actually has. No operator wants to know whether customers are broadly satisfied — that answer is nearly always yes, broadly. What they want to know is where they are losing, to whom, and by how much.

That question has a reference point built into it. A measurement system without a reference point cannot answer the only question its users are asking.

Method & scope

Position This reflects a locked design decision in Magicalia Insights: sentiment is never displayed as an absolute score. It is always shown as divergence from a defensible reference.

Figure 1 Illustrative, drawn to demonstrate the mechanism. The pattern is drawn from real client work; the values are not a specific client's data. We label illustrative figures as illustrative.

— · —

We'll show you something you don't know about your own customers — before you sign anything.

A public-signal diagnostic uses only what your customers have already said in public. No data handoff, no NDA, no procurement. You'll see what the market sees, and where the gap is.

Request a diagnostic →