From numbers to an actual decision

A great many businesses measure diligently and decide nothing. The reports accumulate, everybody agrees they are interesting, and the website stays exactly as it was.

Talk to us about your site

Turn numbers into decisions by asking three questions in order: are people arriving, are they acting, and are those actions worth anything. Each answer points at a different fix. Where the numbers are too small or too ambiguous to decide, say so and choose on judgement rather than pretending the data decided.

The three questions, in order

Nearly every website problem resolves into one of three, and asking them in this order stops you from fixing the wrong one.

Are people arriving? If not, the constraint is visibility. Are they acting? If not, the constraint is the page. Are those actions worth anything? If not, the constraint is who you are attracting.

Each answer points somewhere specific

The value of the sequence is that each branch has a different remedy, and applying the wrong remedy is expensive.

  • Few visitors: build the pages people search for, and be present where they ask.
  • Visitors but no action: clarity, proof, cost signal, and an easier next step.
  • Action but low value: change who you attract and what you show.
  • All three fine but revenue flat: look past the website, at follow-up and pricing.

Decide what would change your mind

Before making a change, write down what you expect to happen and by when. Not as a formality: as a way of preventing the very human habit of reinterpreting results afterwards to fit the decision you already liked.

This also settles arguments in advance. When two people disagree, agreeing on the measure and the horizon is usually easier than agreeing on the change itself.

Accept when the numbers cannot decide

Sometimes the honest answer is that the data is too thin. Four enquiries against six is not evidence, and no amount of presentation makes it so.

In those cases, decide on judgement and say that is what you are doing. A decision made openly on judgement can be revisited. A decision dressed up as data-driven becomes very difficult to reverse, because reversing it feels like admitting the data was wrong.

Prefer decisions that are cheap to reverse

When evidence is limited, weight your choices toward the ones you can undo. Rewriting a headline is reversible. Rebuilding the entire site is not.

This single habit protects small businesses from most expensive mistakes, because the irreversible decisions tend to be exactly the ones made with the least evidence.

  • Try the reversible version first.
  • Keep the old version so you can return to it.
  • Change one meaningful thing at a time.
  • Write down the date, always.

Close the loop

The step almost everybody skips is going back. Three months after a change, look at what you wrote down and check whether it happened. Then write one line recording what you learned.

Businesses that do this get better at deciding, because they accumulate evidence about their own market rather than opinions about websites in general.

How Licheo works

We report against the three questions rather than presenting a wall of figures, so the report points at a decision rather than at a discussion.

Where the numbers cannot support a conclusion, we say so and recommend on judgement, clearly labelled as judgement.

Part of a larger guide

This page is one part of Revenue per visitor. The other parts:

Questions people ask

How do I turn website data into decisions?
Ask three questions in order: are people arriving, are they acting, and are those actions worth anything. Each answer points at a different fix, and asking them in order stops you fixing the wrong one.
What if my numbers are too small to be conclusive?
Say so and decide on judgement, clearly labelled as judgement. A decision dressed up as data-driven becomes hard to reverse later, because reversing it feels like admitting the data was wrong.
How do I avoid interpreting results to suit myself?
Write down what you expect and by when, before making the change. It is a simple discipline and it prevents most after-the-fact rationalisation, as well as settling disagreements in advance.
Which decisions should I make first?
The reversible ones. Rewriting a headline can be undone; rebuilding a site cannot. Irreversible decisions tend to be made with the least evidence, which is exactly the wrong way round.
What does closing the loop mean?
Going back three months later, checking what you predicted against what happened, and writing one line about what you learned. It is how a business accumulates evidence about its own market rather than opinions.

Reports that point at a decision

We report against the three questions, and where the numbers cannot decide, we say so and recommend on judgement.

Talk to us about your site Read the full guide