When a firm tells us its marketing is not working, the diagnosis almost always turns out to be narrower than the complaint. Traffic is fine and enquiries are weak. Or enquiries are fine and nobody books. Or meetings happen and the proposal sits unanswered for three weeks. The accounting buyer's journey from first search to signed letter has distinct stages, and each one can fail independently while the symptom you notice stays the same: not enough new clients.

Our position is that you cannot decide between doing this yourself, hiring someone, or bringing in outside help until you know which stage is actually broken. Route choice follows diagnosis, never the other way round.

Below is the journey as we map it across the firms we work with, stage by stage, with what needs to exist at each point and what the common failure looks like.

The journey rarely starts with someone searching "accountant near me". It starts with a problem statement. A director types something like "do I need to register for VAT if I sell on Amazon" or "changing accountant mid year". At this point the buyer is not comparing firms. They are trying to understand their own situation.

What you need here is a page that answers that exact question better than the page currently ranking. Not a services page with a paragraph about VAT. A page whose entire job is that one question, written for the specific type of business asking it.

The common failure is a website built as a brochure. Five service pages, an about page, a blog with two posts from 2023. That structure can only compete for the handful of high-competition commercial terms every firm in the country is bidding on. Prads at Wings Online Filings went from five to seven enquiries a month to fifteen to sixteen inside the first two and a half months, and the mechanism was architecture rather than persuasion: more pages, each aimed at a real question, each with a route onwards.

Measure this stage on impressions and rankings for question-shaped queries, not total sessions.

Once the buyer understands their problem, behaviour changes. They start comparing. This is where they open three or four tabs, read your homepage, check your reviews, and look for evidence you have handled a business like theirs before.

Everyone knows reviews matter, so that is not the useful observation. The useful observation is that proof needs to be specific to the buyer's category. A wall of five star reviews from generic small businesses does very little for an ecommerce seller with inventory across two marketplaces. One paragraph describing exactly that situation and how it was handled does a great deal.

What you need at this stage:

• Review presence on the platform your buyers actually check, with enough volume to look current
• Two or three written client situations that mirror your target client's circumstances, including the messy detail
• Clear statements of who you work with and, by implication, who you do not
• Pricing model explained in words, even if you do not publish figures, so the buyer can self-select

The failure here is invisible in your analytics. Nobody fills in a form to tell you they did not believe you. Traffic looks healthy, enquiries look thin, and the instinct is to buy more traffic.

The enquiry arrives. This is the stage most firms have thought about least, and it is where the economics of the whole system are decided.

An enquiry form with name, email and message produces enquiries you cannot rank. You do not know turnover, entity type, current software, whether they have an accountant, or how urgent the problem is. So every enquiry gets the same treatment, which means either you call everyone (expensive in partner time) or you call nobody quickly enough (expensive in lost clients).

The fix is asking more, not less. Counterintuitive, but consistent in our data. A short structured form that asks entity type, approximate size, current provider status and timing filters out tyre kickers before they reach a calendar. Annabel runs an AI qualification layer on her enquiry form for exactly this reason, and the effect is that her calendar only carries opportunities worth the hour.

This is also where the AccountingWEB forum complaint about twelve leads producing one client usually resolves. Twelve unqualified enquiries and twelve qualified enquiries are different products. Comparing conversion rates between them tells you nothing.

Measure enquiry to booked meeting rate here. If it sits below half, the form is the problem before the sales conversation is.

Speed does most of the work at this stage, and almost nothing else does. An enquiry answered in five minutes converts at a different level to one answered the next working day, because the buyer has three other tabs open and one of those firms will respond first.

What needs to exist:

• An automated acknowledgement within seconds that sets expectations and offers a calendar link immediately
• A booking system that shows real availability rather than a promise to "arrange a convenient time"
• A follow up sequence that runs for weeks, not days, because a meaningful share of enquiries are researching ahead of a year end or a fallout with their current accountant
• Reminders before the meeting, since no-shows are a pipeline leak most firms never quantify

The "we will use you later" response that practice owners report so often is usually a follow up architecture problem rather than a pricing problem. The buyer meant it. Nobody followed up in February when the year end actually landed.

This is the cheapest stage to fix and the one most firms skip, because it is unglamorous plumbing rather than marketing.

The meeting went well. Now the buyer needs to make a decision, and every day of friction between decision and signature is a day a competitor can move.

A proposal that arrives as a Word attachment three days later loses to one sent the same afternoon with electronic signature, AML checks and direct debit setup built into the same flow. The work is identical. The sequencing is what changes the outcome.

Then onboarding, which is part of acquisition rather than a separate function. A buyer who signs and then waits a week for a request for documents starts doubting the decision immediately. The firms that scale cleanly have the same sequence every time: engagement letter, identity verification, authorisation codes, software access, data request, first scheduled contact. Automated, tracked, and identical for client forty as it was for client four.

This is where growth either compounds or stalls. If every new client costs you three hours of partner admin, acquisition volume becomes a workload problem rather than a revenue one, and firms quietly throttle their own marketing to protect capacity.