Google reviews for tradies: asked for on the day the job finishes.

The end of the job, on site, while the customer can see the work. Nothing you do later comes close, because that is the point at which they are most pleased and you are still in front of them.

The alternative most trades default to is the invoice, and it is the worst possible timing. The request arrives attached to a bill, while the customer is thinking about money rather than about the work.

The gap is not knowledge, it is that the moment passes in about thirty seconds while you are packing up. What closes it is deciding in advance to say one sentence before you leave, every time.

Why on-site beats everything else

• They can see the work.

  The reason they are pleased is in front of them, which is not true an hour later.

• You are a person, not a text message.

  A request from somebody standing there is answered far more often than one from a number.

• It costs no time.

  It is one sentence at the end of a job you have already finished.

Say it in your own words while you are still there, then send the link before you drive off. The sentence does the persuading and the link removes the effort.

Sending somebody away to find your business on Google loses most of them. Google gives every business a short direct link from inside the listing, and the difference between using it and not using it is most of your reviews.

Naming the job is the one thing worth doing deliberately. A review that mentions the hot water system, the switchboard or the roof is a review that helps the next person searching for exactly that, and it only happens if you mention it.

The routine, three steps

• Say it before you pack up.

  One sentence, your own words, naming the job you just did.

• Send the link from the ute.

  Same day, while it is fresh. A saved message with the link in it takes two seconds.

• Reply when it arrives.

  Google counts replies and it takes ten seconds.

No paying, no discounts in exchange, no asking only the customers you expect to be happy, and never writing them yourself. All four risk the listing rather than just being frowned upon.

The one that catches honest trades is choosing who to ask. Asking the pleased customers and quietly skipping the difficult ones feels like common sense and is a breach of Google's policy.

Offering a discount for a review is the other common one, and it is against the rules whether or not the review is honest. The safest position is that everybody gets asked and nobody gets anything for it.

If a competitor is obviously buying reviews, they can be reported. That is a better use of ten minutes than trying to match them, because the pattern is recognisable and the usual outcome is that they lose all of them.

Google weighs how recent reviews are, so fifteen from this year usually beats sixty that all stop three years ago. A steady trickle is worth more than a burst.

This is why a newer trade can outrank an established one. The old business has the better reputation and the worse signal, and the map reads the signal.

It is also why a burst does not work. Twenty reviews in a week reads as a campaign, whereas one a week for twenty weeks reads as a business that is simply busy and asking.

For a trade competing across Newcastle and the Hunter this is often the whole difference. Everything else about the competing listings looks the same from the map's point of view, and the one with reviews from this month takes the position.