Module 6 of 7
Testing on twenty strangers
And what to measure.
Twenty is enough to stop you generalising from your own taste and small enough to arrange in a fortnight. Strangers matter because people who know you fill in the gaps from what they already know about you, and a stranger cannot, which is precisely the condition your page has to survive.
Two kinds of test, and both are needed. The five-second test: show the page for five seconds, take it away, and ask what this is and who it is for. If they cannot answer, the top of the page has failed, and nothing further down is being read. Then the read-through: let them look properly and ask what they think it costs, what they would do next, and what would stop them buying. That last question is the most productive one in the module and almost nobody asks it.
Then measure behaviour, not enthusiasm. Did anybody enquire? Did anybody click the one action? Did anybody who said it was great actually do anything afterwards? "Looks good" is politeness, and ENT-101 already taught you what to do with politeness: log it, and weight it at nothing. Write down what people did, and where they hesitated or reread, because a hesitation is a sentence that is not working.
Five seconds for what and who; a read-through for price, next step, and what would stop them. Then count what people did, not what they said.