General

Why small business data quietly goes bad, the five ways it drifts from reality, and the habits that let a one-person business actually trust its own numbers.

I ended a free shipping banner test on day six with fourteen orders behind it and called it a nine percent lift. Checking a running test is itself a decision, and the rule ‘stop when it looks good’ can only ever finish one way.

A plain-English look at marketing attribution: what it is, why the common models disagree, and how a small business can use them without fooling itself.

A plain-English guide to RFM analysis: how to score customers on recency, frequency, and monetary value using nothing but a spreadsheet and your own order history.

Analytics said 41 sales, the processor said 38, the ad platform claimed 52. All three were correct, and the Saturday afternoon I spent deciding which one to believe cost more than the gap was worth.

A plain-English walkthrough of customer acquisition cost and payback period, and how the two numbers together tell you whether your growth spending is actually safe.

A plain-English guide to finding your north star metric: the one number that shows whether customers are getting real value, how to find it in your own data, and the mistakes that turn it into a number you game instead of grow.

A join against a table I assumed had one row per port left my total data figure about 40% too high for eleven days. Two row counts would have caught it in a second.

A plain-English breakdown of statistical significance, p-values, and sample size, explained with a coin flip instead of a formula, for anyone testing changes in a small business.

A plain-English guide to running an honest A/B test on small traffic, and the specific ways solo founders trick themselves into calling a coin flip a win.