One question, asked early, filters most bad jobs

One question asked before you accept - what are you hoping to get out of this? - filters out most of the jobs that would go badly. Almost every bad job was identifiable in advance, from information the buyer would have given freely if asked.

It was not asked because asking feels like it risks the booking, and that fear costs far more than the bookings it protects.

What you are actually screening for

Not difficulty - ambiguity.

The jobs that lose money are not the hard ones; they are the ones where you and the buyer have different pictures of what is being delivered, and neither of you discovers this until the work is done.

So the question you want answers one thing: does this person know what they want, and can they say it.

The question

The general form asks for the outcome rather than the input, such as what they will do with the assessment once they have it.

A buyer who answers in a sentence with a concrete purpose is a good job. A buyer who answers with three words, or with something your format does not deliver, is the one you were going to have trouble with.

The second most useful question, when the first is already answered, is when they need it by. A vague answer to a deadline question is itself a finding.

Ask one, not three

An interrogation loses the booking, and it loses the good bookings preferentially, because a buyer who knows what they want has the least patience for a form.

Ask it in the reply where you also confirm you can do the job and restate the price, so the message reads as moving towards acceptance rather than as gatekeeping. "Yes, I can do that, £X, back within two days - what are you most hoping the assessment tells you?" screens, quotes and confirms at once.

The piece on reading a thin brief has more on the wording, including what to do when the answer comes back as thin as the brief.

Reading the answers

Three responses are worth acting on immediately.

A clear, specific answer. Accept, confirm the scope in writing, start.

An answer describing something you do not offer. Say so in one sentence and either propose the nearest thing you do offer or decline. This is a genuine win and it will not feel like one - declining quickly and politely costs one job and saves a review.

No answer at all, or a deflection. A buyer who will not spend ten seconds before paying will not spend ten seconds explaining what went wrong afterwards, and their dissatisfaction will arrive as a one-line review. Decline, or take the job only with the scope written out in unusual detail.

A very long answer is not a red flag; it is usually an anxious buyer, and anxious buyers often become good repeat buyers.

Then write the scope down

Screening tells you whether to take the job; the scope sentence protects you once you have. It is also the reference point consumer law uses: the UK's Consumer Rights Act 2015 says services should match up to what has been agreed. Doing the first does not excuse skipping the second - the two-sentence confirmation is your entire case in the one job out of fifty that goes wrong.

Buyers do not experience a single upfront question as friction, and their own account of how a commission normally opens has a clarifying exchange in it as standard - it is described here. Where the buyer's answer implies a precision you cannot actually deliver, that is the moment to say so rather than after, and what an automated score reliably does and does not claim is a useful calibration for where the honest line sits. If the request depends on material the buyer has to supply, pointing them at the standard conventions for producing it before they send anything saves the round trip.

For the platform-side version of what a Rate Cock judge can ask before accepting, and what the messaging window allows, the judges page is the source.