Almost every partner we speak to asks a version of the same question. How do I get clients beyond referrals without gambling money on another marketing attempt that goes nowhere. The scepticism is earned, because most firms have already bought a website, an agency retainer or a round of ads and have very little to show for any of it.

Our position is that the usual failure is one of order rather than effort. Firms buy attention first, point it at a page that could describe any practice in the country, and conclude that marketing does not work for accountants.

Build it the other way round. One buyer, public evidence and one story, then a direct booking, qualification and follow up route, then controlled attention, starting with a test small enough that being wrong costs you a week.

A general practice that had never deliberately marketed itself added eleven clients and £10,430 of recurring fees in six months. The partners were sceptical for sensible reasons. They had watched peers spend money on websites and agencies and end up with a quieter phone than before.

What changed was not the size of the commitment they made at the start. It began with a seven day test built to answer a single question: does demand for this service exist beyond the people who already know us. That question is cheap to answer, and most firms never put it to the market at all.

Partners routinely describe referrals as how the firm grows. That description is worth testing, because a growth system is something you can turn up when capacity opens and turn down when the team is already stretched.

Ask three questions about last year. Did you choose the month the new work arrived. Did you choose how much of it there was. Did you choose the kind of business it came from. If the honest answer is no in all three cases, referrals are a source of work the firm receives rather than a system the firm runs. The distinction matters as soon as you want to plan anything.

A referral happens when a client, a solicitor or a bank manager decides to mention your name in conversation. That decision belongs to them. It tends to come when they happen to meet someone with a problem, which bears no relationship to whether you have a gap in March or a new manager to keep busy.

There is a second effect that firms notice later, usually when they look at their own client list with fresh eyes. Clients refer people like themselves. The demanding client with the carrier bag of receipts introduces another one, and the cycle repeats until the list looks less like a set of choices and more like something you inherited.

Then the arithmetic catches up. Every network has an edge. Once you have met the people your existing clients and contacts can reasonably reach, the next buyers in your area are people with nobody suitable to ask. They are starting up, switching after a bad experience, or looking at a deadline they cannot meet on their own.

That work does not vanish because your firm is hard to find. It goes somewhere, every month, in your town. The question worth sitting with for a moment is where it currently goes, because whoever receives it is being found at the exact moment the buyer starts looking.

Here is the principle the rest of this rests on. Buyers who have no one to ask choose whoever they can find.

That sits uncomfortably with a good firm, because it means technical quality is not the deciding factor at the point of choice. A business owner cannot assess your working papers before they hire you. They can only assess what is visible to them on a Tuesday evening when they finally deal with the thing they have been avoiding. So the job is to be findable as well as trusted. In most practices the trust is genuine and the findability has simply never been built.

The trust already exists. Your referrers feel it, which is exactly why they pass your name on. The problem is that it lives in private conversations and in nothing a stranger can reach.

Making it visible is ordinary, unglamorous work. Decide which single kind of client the firm is built for. Put the evidence in public, so that reviews, figures and named results sit somewhere a person can read them without having to ask. Then tell one story consistently across the website, the profiles and anywhere else the firm appears. A practice that says one clear thing is far easier to choose than a practice that says everything.

Only once that is in place do you point attention at it. Paid traffic aimed at a vague website is where most marketing budgets quietly die, because the clicks arrive and have nowhere sensible to land.

Behind the page there has to be a direct booking route, a simple way to qualify who is worth a meeting, and a follow up sequence for the people who enquire and then go silent for a fortnight. With those three things working, the useful part is how fast the market answers you. No year long retainer is needed to learn whether demand exists outside your network. A week will do it.

A general practice with five to fifteen staff in Stoke on Trent ran a single advert about Making Tax Digital, timed to a moment when the businesses around it were already uneasy about the change. Thirty five or more leads arrived in under a week. Month one closed with eighty one leads on £594 of spend, which works out at £7.33 per lead.

Timing did most of the work there. The service was already sitting on the shelf, priced and staffed. What changed is that it was placed in front of the right people at the point they were thinking about the problem.

Leads on their own prove very little. Plenty of firms hold a folder of enquiries that never became anything, which is why we treat lead volume as a leading indicator and nothing more than that.

The practice in Barnstaple ran the full sequence, including the qualification step and the follow up. Over six months it added eleven new clients and £10,430 of annual recurring fees. Seventy five per cent of qualified enquiries reached proposal, forty seven per cent closed, and the return came out at 826 per cent. The advert was the smallest component of that result. The handling of what it produced was the larger one.

So here is the test. Pick one service that one specific kind of business needs right now, usually driven by a deadline, a start-up or a switch away from another accountant. Publish one plain page on the firm's website that says who it is for, shows the evidence, makes clear that enquiring costs nothing, and gives a direct booking route.

Then run one Meta advert to that page at £10 a day for seven days and count enquiries. Five or more means demand exists and the firm was simply not being found. Zero or one means the words or the offer are wrong rather than the market. Two to four settles nothing, so change the wording and run it again before you conclude anything.

The alternative is another year of the current arrangement, where the timing, the volume and the character of your client list all sit in other people's hands. Nothing dramatic goes wrong in that year, which is precisely what makes it so easy to defer the decision again.

Set that against the Barnstaple figures. Eleven clients and £10,430 of recurring fees came through a channel the firm controlled, which meant it could be turned up when capacity allowed and paused when it did not. The marketing itself is the smaller difference here. The larger one is who gets to decide when the practice grows.

None of this has to be decided today. Keep it for the Monday you conclude that referrals alone are no longer enough, which tends to arrive after a quiet quarter or a client you would not take on again.

When that day comes, the whole test costs seventy pounds and a week of mild attention. At the end of it you have an answer to the question most practices avoid asking, which is whether anyone outside your network actually wants what the firm sells. Either answer is useful. A clear no tells you to change the words or the offer, and a clear yes tells you where the next year's growth is going to come from.