The 9pm test

Somebody recommended you today. Over lunch, in a meeting, in a group chat. It went well. They said you were good, and the person listening wrote your name down.

At about nine that evening, on a sofa, on a phone, they looked you up. That is the moment the recommendation is either confirmed or quietly dropped, and you were not in the room for it. Nobody is. There is no salesperson, no call, no chance to explain that the website is being redesigned.

We have started calling this the 9pm test, because once you name it you notice how much of buying now happens there.

What is actually being tested

Not your credentials. The recommendation already covered those. What gets tested at 9pm is whether the way you present yourself is consistent with what was said about you. Someone arrives half-convinced and looking for permission to be sure.

They find one of two things. A firm that looks like the description: current, specific, evidently run by people who care about detail. Or a gap between the recommendation and the evidence, which is where trust goes to die. Nobody concludes "the website is dated but the work is probably excellent". They conclude "hmm", and they look at the next name on the list.

Sixty seconds, three questions

The test is short and the questions are the same in every sector we work in.

Is this the right kind of organisation for me? Answered by imagery and tone inside about four seconds, long before any copy is read. This is why photography is not decoration: it is the fastest-loading argument you own.

Can I tell what it costs and how to start? Not a full price list, but a signal. Firms that hide the commercial reality read as expensive and awkward. Firms that say "from" read as confident.

Are these people I would want in the room? Answered by whether there are any actual people on the site. Stock photography fails this instantly. Forty inconsistent LinkedIn headshots fail it too, more slowly.

It is a phone, and it is not a good connection

Every version of this happens on mobile, often on a bad signal, usually with the lights low. That has design consequences that get overlooked when a site is signed off on a large monitor in daylight: type that holds at arm's length, contrast that survives a dimmed screen, and pages that arrive before patience runs out. A three-second load is not a technical metric at 9pm. It is the difference between being considered and not.

What to fix first, if you can only fix one thing

Photography, and it is not close. Copy can be rewritten in an afternoon and a slow page can be made fast by someone technical in a week, but imagery sets the verdict before either of those gets read or measured. A credible site with plain copy survives the test. Beautiful copy behind stock photography does not, because the visitor has already decided by the time they reach a sentence.

Second is speed, third is whether there are prices, fourth is everything else that usually dominates the redesign meeting.

The sector version of the same problem

The test is universal, but what gets checked varies, and that is what the In Frame pages are for. A parent checking a school at 9pm wants to picture their child in a corridor. Someone checking a law firm wants to know whether these are people who will return a call. An off-plan buyer wants to believe a building that does not exist. Same sixty seconds, different evidence required.

The practical move is the same everywhere, though: decide what the 9pm visitor has to conclude, then check whether your own site delivers it on your own phone, tonight, with the brightness down. Most people have never actually run the test on themselves. It is uncomfortable and it takes a minute.