The sales figures that flatter you, and the one that cannot

A note first. This article discusses publishing your own trading figures. It is not legal, data protection or advertising compliance advice. A published performance figure is a claim, so it needs to be one you can evidence and keep accurate.

Our piece of 10 November 2026 said that a vendor choosing between two agents wants to know whether you can sell their house for something like the price you quote, and that the figure answering it is sitting in your own records rather than on your website. We looked at eleven agency sites and found one publishing anything of the sort.

The obvious next question is which figures. Not all of them are equal, and we recommended one in that piece that has a flaw worth owning.

Percentage of asking price has a hole in it

It is the figure we would most want to see, and an agent will spot the problem with it straight away.

Percentage of asking price goes up when the asking price comes down.

An agent who prices cautiously will report a better percentage than one who priced ambitiously. So will an agent who reduces the price before agreeing a sale. Both of them can beat a competitor who put a brave figure on a house, held it, and in the end got the vendor more actual money. Taken on its own, the number rewards exactly the behaviour a vendor should least want from an agent, which is talking a price down to protect a statistic.

That is not a reason to drop the figure. It is a reason to publish a version of it that cannot be worked.

Measure against the first asking price. The figure the property launched at, not the last one before a sale was agreed. Then say that is what you are doing, in those words, because the saying is half the value. An agent publishing ninety four per cent of our original asking price is making a claim that a reduction cannot improve, and one that a competitor quoting the softer version cannot match without changing how they measure.

The same logic points at a second figure most agents will not want to publish: how many of your sales needed a price reduction, or the proportion that went through without one. It is uncomfortable. It is also unanswerable by a competitor who has not published it.

A sale agreed is not a sale

Everything above counts sales agreed, because that is what agents measure, what their systems report, and what the one firm in our sample published.

But a vendor is not asking how long until a board outside says Sold Subject to Contract. They are asking how long until they move. And an agreed sale can fall through, after which the house goes back on and the clock starts again.

Which means the figure a vendor would value most is the one nobody in our sample published at all. Of the sales you agree, what proportion complete.

It is the hardest figure to flatter, because there is no pricing trick that improves it. It rewards qualifying buyers properly, chasing a chain, and keeping a sale together through the dull middle months, which is most of the actual work of being an estate agent and none of the part that gets marketed. If your completion rate is good, you are holding a number none of your competitors are showing, and it is worth more than any award on the wall.

If you have never calculated it, that is worth doing whether or not you ever publish it.

Time to sale agreed, and its honest companion

Average days to a sale being agreed is a reasonable figure and easy to pull. It also has an obvious partner that makes it meaningful, which is average days from sale agreed to completion.

Published together, they tell a vendor the actual shape of what they are signing up for: this long to find a buyer, this long to get them over the line. Published alone, the first one is the flattering half of the story.

When not to publish

Here is the part that would be dishonest to leave out.

The one firm in our sample publishing trading figures presumably publishes them because they are good. That is a selection effect, and it would be glib to turn one firm's confidence into advice for everybody.

For a firm averaging well below asking price, or taking a long time to agree, publishing converts a quiet, neutral position into a documented disadvantage. It hands every competitor a benchmark to quote in a listing presentation. And it cannot really be withdrawn, because a vendor who saw it will remember, and the internet keeps copies.

For a small independent selling twenty houses a year there is a second problem, which is variance. Twenty sales is a small enough number that one difficult chain or two reduced prices can move a percentage several points. A quiet quarter can look like incompetence when it was mostly luck, and the honest way to handle that is a longer period and a stated volume, so a reader can see what the average is made of.

So if your numbers are not good, this is not advice to publish them. It is advice to find out what they are, which plenty of firms genuinely have not calculated, and then work on them. If they are around the middle, publish the ones you are strongest on, state the period and the volume, and leave the rest. Publishing everything is not the point. Publishing something true, specific and current is.

The order we would publish them in

If you want a running order, this is ours, best first.

• Completion rate, over a stated period. Hardest to flatter, most relevant to what a vendor actually wants, and nobody else is showing it.
• Percentage of the first asking price, over a stated period, with the volume. The strongest price claim that cannot be improved by reducing a price.
• Days to sale agreed and days from agreed to completion, published as a pair. The real timeline rather than the flattering half.
• Sales agreed over a stated period. Useful context, and the figure that makes the percentages mean something, because ninety seven per cent of eleven sales is a different claim from ninety seven per cent of fifteen hundred.
• Proportion of sales agreed with no price reduction. The brave one. Few will publish it, which is exactly why it works for the ones who can.

And on all five, the period is not a footnote. Our earlier piece found the best example in that sample carrying figures labelled October 2023 to September 2024 while the copy around them said in the last year. A figure without a current period on it stops being evidence and becomes something a vendor has to take on trust, which is where we started.

The measure

Pick the one figure above you would least like a competitor to publish about you.

That is the one worth working on. If it is already good, that is the one to put on the website.