Guide

SaaS funnel diagnosis

“Conversion is low” is not a diagnosis. A SaaS funnel diagnosis names the step that is losing people, and refuses to recommend a fix until that step is obvious.

The steps worth separating

For a product that is already launched, five counts are enough to start:

• Visitors — people who saw the product.
• Signups — people who started an account or a trial.
• Activated users — people who reached the moment of value, defined by you in one sentence.
• Paying customers — new payments in the same window.
• Retention — who stayed, if you have enough customers to count churn without kidding yourself.

Each rate is the later count divided by the earlier one. Visitor → paid hides which division is the bad one. That is why a diagnosis looks at the steps, not the blend.

What “activated” has to mean

If activation means “logged in twice,” you will diagnose the wrong thing. Activation is the action after which a reasonable buyer understands the product. “Created their first invoice.” “Imported a repo and saw a result.” Write that sentence down before you compute the rate. Otherwise signup → activation is just a vibe.

Sample size is part of the diagnosis

A step with a tiny base is not a broken step. Five visitors and zero signups is an acquisition problem, or no problem yet. You do not get to declare the landing page a failure. A diagnosis says “not enough data” when that is the honest answer.

Diagnosis is not the experiment

Finding that activation → paid is the thin step does not tell you to cut the price. It tells you the argument is about value, packaging or trust, and that rewriting the top of the page is a distraction. The next move is one cheap test aimed at that step, with a result that would prove you wrong.