Ask ten practice owners about their accountancy practice growth strategies and you will hear roughly the same list: sharpen the niche, add advisory work, tidy up the website, post on LinkedIn, ask for more referrals. None of it is wrong. Most of it is also what the firm three miles away is doing, which is part of why so few firms can tell you what their growth will look like next quarter.

Our view, after building acquisition systems across dozens of practices, is simpler. The strategies fail because they are activities rather than assets. An activity produces a result once and stops when you stop. An asset keeps producing after you have moved on to something else.

The rest of this piece covers what that distinction looks like in practice, why referral dependence is a capacity ceiling rather than a marketing failure, and what each month of delay actually costs.

Here is a test worth running on your own firm. Take everything you did last quarter to bring in work, and sort it into two columns. In the first column, anything that stopped producing the moment you stopped doing it: the networking breakfast, the LinkedIn post, the one-off email to dormant contacts. In the second, anything still producing today without further input.

For most practices the second column is close to empty. That is the whole problem in one page.

A ranking page for a search term your ideal client actually types is an asset. It was written once and it produces enquiries every month afterwards. A tested paid campaign with known cost per enquiry is an asset, because once the numbers are known, spending more is an arithmetic decision rather than a leap. An automated onboarding sequence is an asset, because it absorbs the tenth new client as easily as the first.

None of these are clever ideas. They are ordinary infrastructure. The reason firms do not have them is that building infrastructure produces nothing in week one, and January always arrives before the work gets done.

We are not going to argue that referrals are bad. They convert better than anything else, they arrive pre-trusted, and any firm that has grown on them has earned that.

The issue is structural. Referral volume is a function of how many clients you already have and how often they meet someone who needs an accountant. You do not control either variable. You can influence them at the margins, but you cannot decide in March that you would like four more enquiries in April.

Two consequences follow. The first is that you take what arrives. That includes the client who negotiates on fee, the one whose records show up in a carrier bag, and the one who quietly consumes the equivalent of two better clients in partner time. The second is that planning becomes guesswork. You cannot decide whether to hire, because you cannot forecast the work that would justify the hire.

Firms that add a controllable channel alongside referrals do not lose the referrals. They gain the ability to say no, which is where margin actually comes from.

Every article on this subject tells you to niche. Few say what to niche into, which is the only part that is difficult.

Most firms pick a sector because a partner enjoys the work, or because three clients happen to sit in it. That is a reasonable starting hypothesis and a poor final answer. The better question is whether there is measurable demand you can actually reach.

Two data sets settle it. Company registry data tells you how many businesses of that type exist in your market and how quickly new ones are forming. Search data tells you how many of them are actively looking for an accountant who does what you do. A sector with plenty of businesses and no search volume needs outbound. A sector with strong search volume and few specialist firms competing for it is the cheapest position you will ever buy.

Positioning chosen this way does more than sharpen your messaging. It decides which pages you build, which keywords are worth pursuing, and which ad audiences are worth testing. It becomes an operating instruction rather than a line on the homepage.

The uncomfortable part of this subject is timing.

Search assets take time to mature. A page published this month will not carry meaningful traffic for several months. Paid campaigns need a testing period before the cost per acquired client settles. Neither of these can be accelerated by deciding to care more later.

Which means the real cost of waiting is not one quarter of missed enquiries. It is the fact that the firm which started in January is, by the following January, running on data you will not have for another year. They know their cost per enquiry. They know which niche pages convert. They can raise spend on the campaign that works and cut the one that does not. You will be at the point they were twelve months earlier, with a year of compounding lost rather than deferred.

One of our clients at Wings Online Filings went from five to seven enquiries a month to fifteen to sixteen within the first two and a half months, and took on nine new clients in that window. The mechanism was ordinary search architecture. The advantage was starting.

The failure mode we see most often is a firm that fixes acquisition and then quietly throttles it.

Enquiries increase. Nobody chases the ones that go quiet. Proposals sit for a week. Onboarding still runs on a partner sending a welcome email between meetings, chasing identity documents by hand, and setting up the client in practice software from a standing start. Within two months the enquiry flow feels like a burden, and the marketing gets paused so the team can catch up.

The fix is unglamorous. Enquiries route into a CRM with a defined follow-up sequence. Qualification happens before it reaches a calendar, so partner time is only spent on opportunities worth the hour. Onboarding runs as a workflow with digital engagement letters, automated identity checks, and structured data collection that feeds your existing systems.

Done properly, the twentieth client onboards with roughly the same partner input as the fifth. Until that is true, every improvement in acquisition just moves the bottleneck somewhere less visible.