What did your last price change actually do to sales?

You raised the price. Revenue looks fine. But do you actually know what happened, or does it just feel fine? This one is measurable, and the measurement has a trap in it.

The trap: small samples lie confidently

Here is the part most write-ups skip. With a few dozen sales per window, the measured number comes with a wide uncertainty band. If that band includes zero, the honest conclusion is that you cannot rule out no effect, and any projected gain built on the point estimate is a story. We enforce this in our own product: Counsel's price card computed an impressive elasticity on demo data, its own confidence interval included zero, and so the headline reads "not conclusive" instead. If a tool never tells you that, it is not measuring, it is flattering.

Run a fair test

• Change one product family at a time, so the comparison is clean.
• Avoid windows containing a holiday, a closure, or a promotion on either side.
• Judge units, not revenue, first. Revenue confounds the price with the demand response.
• Decide the judgment date before you start, then actually wait for it.

Common questions

What is a good elasticity number?
For most small businesses anything between 0 and about -1 means demand held well enough that the raise paid. Steeper than -1 means units fell faster than price rose. But the honest answer depends on the uncertainty band around your measurement, not the point number alone.

How long before I can judge a price change?
Two clean comparable windows. For most shops that is 4 to 8 weeks total. Judging in week one is how good changes get reverted and bad ones get kept.

Counsel's rule, applied to its own content: every figure above is computed, sourced from our demo businesses or your own inputs, never invented. This is cash math, not tax or legal advice. Written and maintained by the founder.