Everything in this cluster

Some numbers go up and prove nothing. How to tell whether any of this is actually working: real lift, not vanity metrics, in plain language and without a data team.

Measure What Matters

Vanity vs Real Metrics

Some numbers go up, feel good, and change nothing. The sorting test in depth: what makes a metric real, why vanity numbers survive, and the practice of demoting them.

Measure What Matters

What to Actually Measure

The index card: five numbers a busy practice can keep honestly — enquiries, booked share, source, days-to-paid, repeat-and-referral share — and the monthly fifteen minutes that reads them.

Measure What Matters

Attribution Is Mostly Guesswork

Why 'this channel drove 14 enquiries' is a guess wearing a report's clothes: the last-click problem, the multi-touch reality, and how to hold attribution claims without being fooled.

Measure What Matters

Would It Have Happened Anyway?

The single most useful question in marketing: incrementality in plain language — the difference between what happened and what would have happened anyway is the only thing you ever buy.

Measure What Matters

Testing Without a Data Team

The holdout method: a real experiment any practice can run with a calendar and honesty. How to pause, what to watch, how long to wait, and reading the result without fooling yourself.

Measure What Matters

Why Your Numbers Don't Match Your Sales

The dashboard says growth; the bank account shrugs. The usual causes of the drift, in the order to check them — counting differences, timing lags, funnel gaps, and phantom wins.

Measure What Matters

Did AI Send That Customer?

The newest blind spot: clients who arrive from an assistant's answer leave almost no trace. Seeing AI-referred work — imperfectly but usefully — before your dashboards can.