Which of last year's instructions came from your website

Ask the principal of a lettings or management firm how much of last year's new business came through the website, and you usually get a pause, then a guess. Sometimes a confident guess. It is almost never a number anyone can point at.

That gap does more damage than it looks like it should. Every decision about the site rests on the missing number. Whether a rebuild is worth it. Whether the monthly upkeep fee is money well spent or a direct debit nobody dares cancel. With no figure, both come down to instinct, and instinct never has to admit it was wrong.

What is usually there instead

Two situations, and we see about as much of one as the other.

Nothing at all. Analytics were never fitted, because nobody asked for them and the developer did not offer. There is no record that any visitor has ever arrived. Change the homepage and you will have no way of knowing whether it helped.

Something that stopped working in 2023. Google retired Universal Analytics that year, and firms who never moved across to the replacement were left with a login that still works and a dashboard that still loads. The chart on it is flat, because nothing has been collected since. This is the worse of the two, because it looks like measurement. People glance at it, see a screen full of graph, and assume somebody is watching.

There is a third version, less obvious. The site has working analytics, the numbers go up and down, and the report says a few thousand visits last month. Nobody in the office can do anything with that. It does not say whether the phone rang.

The number worth counting

Enquiries, and where each one came from. That is the whole of it.

Split them by what the person wanted, because they are not worth the same. A landlord asking about full management is a different prospect from an applicant asking about a flat that went last week, and a valuation request is different again. If you have read what an empty unit really costs, you already know that one enquiry arriving three weeks earlier can be worth more than the entire annual cost of the website.

Then record where each one came from. Not in fine detail, just the honest categories: a search for something like "commercial property to let in Preston", a search for your firm by name, your Google Business Profile, a direct visit from someone typing the address off a board, a click through from a portal listing or another site.

The mechanics are simple enough that any competent build should include them.

Every form on the site carries hidden fields that record the page the person was on and how they arrived, so the enquiry lands in your inbox already labelled. Then a phone number that appears only on the website and nowhere else, so calls that begin on the site can be counted apart from calls off a board or a letterhead. And every submission stored somewhere you control. Not fired off as an email that someone later deletes.

Attribution in property will never be clean

This is where most agencies stop being honest, so here is the limit.

Say a landlord drives past one of your boards on the Tuesday, looks up the firm that evening, reads four pages on his phone, decides you look like the right size of outfit, and rings the office on Thursday morning. Your call log will record an inbound call. Nothing anywhere will record that the website did the convincing. A website-only number catches some of these and misses plenty, because people ring whichever number they saw first.

So the goal is to narrow the gap. Closing it is not on offer, and neither is a clean audited figure. What you can have is a rough split, and a set of numbers that move when you change something. That beats nothing by a distance, and it is enough to decide where the next few hundred pounds goes.

One more limit, since it comes up. Capturing the source of an enquiry and sending an alert within seconds of it arriving is a build problem, and we can solve it. Whether anyone rings that landlord back inside the hour is a habit inside your firm, and no website has ever fixed that.

What a one-page report should contain

One side of A4, on the first working day of the month. Any longer and it will not be read.

Enquiries this month, by type, with the same month last year printed next to it. Seasonality in property is real, and month-on-month comparisons will mislead you.

Where they came from, in the plain categories above.

The pages people arrived on most, and whether enquiries came from them. A page with heavy traffic and no enquiries is telling you something specific.

What changed on the site last month. A new page, an amended fee table, listings added or taken down, a form rebuilt. Without this line you cannot connect a jump in enquiries to anything you did. It also exposes a common problem in a hurry: if the answer is "nothing" for six months running, the real issue may be that nobody at the firm can change the site at all.

Anything broken. A form that errors on submit, a page returning a 404 where a listing used to be, a phone link that does not dial on a mobile, an alert address still pointing at a member of staff who left.

Then someone has to read it. Ten minutes at the start of a Monday meeting, once a month, with one person named as the owner. A report that arrives and gets filed unopened is the same as no report, only it costs more and lets everyone feel they are on top of it.

If you would like a site that reports for itself, tell us about the business and we will quote for it.