The four-question price survey and its limits

The four-question price survey asks each person for four prices for a described offer:

1. so cheap they'd doubt the quality;
2. a bargain;
3. getting expensive, though they'd still consider it;
4. too expensive to consider.

For any sensible respondent the four prices rise in that order. Answers that break it (a «bargain» below «too cheap») are usually removed before the analysis.

Four prices per person show where their sense of a normal price starts and ends. Nazar sent the survey about Club Plus, at so much a month, to every Club member in November; 220 of 880 answered. The next cards turn those answers into curves and a range, then look hard at what the range does and doesn't mean.

Four prices from each of 220 people become four curves. At each price, count the share of respondents for whom that price is already too cheap or a bargain (these curves fall as the price rises), and the share for whom it is already getting expensive or too expensive (these rise).

The usual reading takes two crossings:

• the range starts where «getting expensive» catches up with «too cheap»;
• it ends where «too expensive» catches up with «a bargain».

For Club Plus that gives 510 to 700 a month. Versions of the method cross slightly different curves, so the exact ends can shift; the idea stays the same. One extreme answer doesn't set the range: the curves summarise everyone who replied.

A range can hide what's happening at the price you care about, so read the curves at that price yourself.

Nazar's summary of 650 was "almost nobody says too expensive". True: only 0.45%, one person of 220, calls 650 too expensive. But the «getting expensive» curve at 650 reads 35%: 77 of 220 members had already said Club Plus is getting expensive at 650 or less. And 24.55% still call it a bargain.

So at 650, a third of the members who answered feel the price is high, even if they'd still consider it. That isn't a reason to reject 650, but it's a very different picture from "almost nobody", and it belongs in any note about the survey.

What the 510–700 range says: among the members who answered, prices from 510 to 700 neither seem suspiciously cheap nor put most of them off. That's all.

What it doesn't say:

• how many would buy at 650, or whether they'd buy at all;
• that the middle of the range (about 605) sells best;
• how much Klasna earns, which depends on purchases and on the cost of serving each member.

650 "being inside the range" means it doesn't shock the people asked. Treat it as a check that no gross mistake is being made, not as a forecast of sales or contribution.

The rising curves invite a mistake. More members call Club Plus expensive as the price goes up, which looks just like a demand curve: fewer buyers at higher prices.

But «getting expensive, but I'd still consider it» is a feeling, not a purchase. Someone can call 650 getting expensive and buy anyway; someone else can call it a bargain and never buy. A demand curve needs what people do at a real price: how many pay, at each price, when the money is actually on the table.

That's the next lesson's topic: price ladders, pre-orders and refundable deposits, which turn words into money.

Two more limits come from who answered and what they already knew.

• Only current members were asked. All 220 already pay 450 for the Club and like it enough to stay. If Club Plus is meant to bring in new people, they were never asked, and members who left this year aren't in the sample either.
• Who chose to reply. 220 of 880 answered; the replies can lean towards the most engaged members.
• The anchor of a known price. Everyone knows the Club costs 450, so "Club plus more" at 650 sounds reasonable next to it. The way the offer is described shifts the answers too.
• Stated, not revealed. Every answer is a feeling, not a purchase.

None of this makes four questions useless. It means the result describes engaged current members' feelings about a price next to 450.

Some readings of the survey pick a single "optimal" price where two curves cross, such as «too cheap» and «too expensive». It's tempting to treat that point as the price that earns the most.

It can't be. Crossing points describe how people feel about prices. Revenue depends on how many actually buy at each price; contribution also depends on the cost of serving them. For Club Plus that cost is real: two small-group sessions a month with a paid tutor. The survey measures neither purchases nor costs.

So any crossing point is, at best, a price that feels comfortable to the people asked, a useful sanity check, not a revenue or profit answer.

With all its limits, the four-question survey earns its place. It's cheap, it catches gross mistakes (a price so low it signals poor quality, or one far outside what anyone considers), and it compares groups, if you ask the right ones: current members versus lapsed members versus people who never joined.

One rule is absolute: never fill it in with an AI model pretending to be respondents. Synthetic answers repeat what the model already believes; they aren't anyone's willingness to pay.

How it fits together: four prices per person become curves and a range; read the curves at your own price; remember the range is perception among those asked, anchored on what they know, not demand. The next lesson moves from words to money: price ladders, pre-orders and deposits, tried on NMT Sprint.