Most accounting firms that ask how to choose an SEO agency have already been burned once. A twelve month retainer, monthly reports full of ranking charts, and a client list that grew by nothing. The work may have been competent. The problem is usually that nobody agreed at the start what the engagement was supposed to produce.

Search is one of the few channels where the buyer intent is already there. Someone typing "accountant for limited company contractors" or "CIS subcontractor accountant" has a live need and a budget. The firm that appears with a relevant page, a fast site, and a clear next step wins the enquiry. The difficulty is that the market for search services is wide open. Anyone can sell it, the delivery quality varies enormously, and the feedback loop is slow enough that a weak supplier can look busy for six months before the gap becomes obvious.

This guide sets out how we assess suppliers when firms ask us to review an existing arrangement, and what we would look for if we were buying search services rather than building them. It covers scope, contract terms, reporting, compliance, realistic timelines, and the specific failure patterns we see in accounting firm engagements.

"SEO" covers at least five separate disciplines, and agencies rarely do all of them well. Before you talk to anyone, work out which of these your firm needs.

The five components

• Technical foundation. Site speed, crawlability, indexation, structured data, mobile rendering. Usually a one off fix followed by monitoring.
• Site architecture. The structure of service pages, sector pages and location pages that lets you rank for more than your firm name. This is the single biggest lever for most practices and the one most often skipped.
• Content production. Guides, answers and comparison pages that capture research stage searches and feed the pages that convert.
• Local search. Google Business Profile management, citations, review generation. Relevant if you take clients within travelling distance.
• Authority building. Digital PR, directory placements, professional body listings, partnerships. Slow, expensive, and the first thing cheap suppliers fake.

Why the distinction matters

A firm with a five page brochure site does not need a content retainer. It needs architecture first, because there is nothing for content to point at. A firm with fifty blog posts and no service page for its main sector does not need more blog posts. Agencies that sell a single monthly package regardless of the diagnosis will apply the same treatment to both.

Ask a prospective supplier to tell you which of the five your firm needs most, and why, before they quote. If they cannot answer without looking at your site, that is fine. If they answer without ever looking at your site, that is the answer.

The first conversation is where most of the useful information is available, if you ask for it. These are the questions that reliably separate operators from resellers.

1. Which of our service lines or sectors would you target first, and what makes you say that?
2. How do you decide which keywords are worth pursuing? Walk me through the data you use.
3. Who writes the content, and does anyone with accounting knowledge review it before publication?
4. What does month one look like, and what does month six look like?
5. How do you track an enquiry back to the search that produced it?
6. What happens to the website, the content and the analytics data if we stop working with you?
7. Which of your current clients are accounting firms, and can I speak to two of them?
8. How many other accounting firms do you work with in our area, and how do you handle that overlap?
9. What is the minimum term, and what is the notice period after that?
10. Who will actually do the work, and are they employed by you or subcontracted?
11. How do you handle cookie consent and analytics tracking on our site?
12. What would make you tell us that search is the wrong channel for our firm?

The last question is the most revealing. An agency that has never turned work away is either very new or very indiscriminate. A supplier confident in their process will happily tell you the conditions under which they would not take the engagement, because they have seen those conditions produce bad outcomes.

Read the agreement properly. Search engagements create assets, and the terms decide whether those assets belong to you or to the supplier.

Ownership

The website domain, hosting account, Google Analytics property, Google Search Console property, Google Business Profile and any content produced should sit in accounts owned by your firm, with the agency granted access. This sounds obvious. It is routinely the opposite. Firms discover at the end of a relationship that the analytics history belongs to the agency, the site is built on a proprietary platform they cannot export, and the Business Profile is managed under an account nobody at the practice can log into.

Term and exit

Search takes time, so an initial commitment of a few months is reasonable. A twelve month lock with no break clause is not, unless the agency is funding significant upfront build work. Look for a defined initial period followed by rolling monthly terms with sensible notice.

Deliverables

Vague scope is where retainers go to die. "Ongoing optimisation" and "content as required" mean nothing when you are trying to work out whether you got value. Ask for the scope to state volumes and types of output: how many pages, what kind, reviewed by whom.

Data protection

The agency will handle personal data through forms, call tracking and analytics. That makes them a processor. You need a data processing agreement, clarity on sub-processors, and confidence that tracking technologies meet ICO expectations under PECR. As the data controller, the responsibility sits with your firm regardless of who installed the script.

Reporting is where the relationship either stays honest or drifts. Most agency reports lead with the metrics that look best rather than the ones that matter to a practice.

Metrics that tell you nothing useful

• Impressions. A measure of how often you appeared, not whether anyone cared.
• Keyword rankings in isolation. Position three for a term nobody searches is worthless. Rankings also vary by device, location and search history.
• Total sessions. Traffic from people looking for free tax advice does not become fee income.
• Domain authority scores. Third party estimates invented by tool vendors, not signals used by Google.

Metrics worth reporting on

• Enquiries by source, split between organic search, paid, referral and direct.
• Enquiry to consultation rate, which tells you whether the traffic is the right traffic.
• Consultation to client rate, which tells you whether the positioning matches what you actually sell.
• Average annual fee of clients won through search, compared with your other channels.
• Cost per acquired client, tracked over a rolling period rather than month by month.

Set the baseline first

Before any work starts, record where you are: monthly organic sessions, monthly enquiries, current conversion rates, and your average client value. Without a baseline, every future report is unfalsifiable. We have reviewed engagements where the firm could not say whether enquiries had gone up, because nobody had counted them before.

Agree the reporting format in advance and insist that the first slide is enquiries, not rankings.

Search does not produce results on a monthly cycle. The work compounds, and the compounding starts slowly.

A reasonable pattern for a firm starting from a weak position looks something like this. The first month is audit, architecture and technical fixes, with no visible movement. Months two and three bring the first pages live and early ranking movement on low competition terms. Months four to six usually produce the first meaningful enquiry volume. Beyond that, the curve steepens as the site accumulates pages and links.

Two of our own client engagements illustrate the shape. Prads at Wings Online Filings moved from five to seven enquiries a month to fifteen to sixteen, with nine new clients, across the first two and a half months, on the back of SEO architecture and a content system. Niall at OD Accountants saw monthly visitors up four times over and ten to fifteen enquiries in the first month after launch, because the website was rebuilt around conversion rather than content volume. Those are faster than typical, and both firms had specific conditions in their favour. Treat them as what is possible, not what is standard.

The economics of cheap search work

We are not going to print figures, but the logic holds regardless of the number. Search work is labour. A low monthly fee buys a small number of hours, and those hours have to cover strategy, production, technical work and reporting. What usually happens is that reporting and templated content consume the budget, and nothing structural changes. The firm pays for twelve months, sees no movement, and concludes that search does not work for accountants. The channel was fine. The allocation was not.

Firms often ask whether they should hire a specialist accounting marketing agency or a strong generalist. The honest answer depends on what you need the specialism for.

Where sector knowledge genuinely helps

An agency that has worked with accounting practices already knows the search language of the market: that people search for "limited company accountant", "CIS accountant", "e-commerce accountant" and "MTD for landlords" rather than for "financial statement preparation". It knows that the buying cycle involves a consultation rather than a checkout. It knows that professional bodies impose standards on how you present qualifications and services. It knows that a page about R&D claims needs to be careful about how it describes eligibility.

Where it is oversold

Technical SEO is technical SEO. Site speed, indexation and internal linking work the same way for a practice as for a plumbing company. A generalist with strong technical depth will often outperform a sector specialist who only produces content.

The overlap question

Ask directly how many other accounting firms the agency works with, and where. If they serve three practices in the same city, chasing the same local terms, someone loses. Some agencies operate territory exclusivity. Others do not, and are open about it. Either is workable. Silence on the question is not.

What actually differentiates

The suppliers worth hiring understand what happens after the enquiry arrives. A form submission that sits in an inbox for two days is a lost client. If the agency has no view on response times, qualification, or how the enquiry reaches your practice management system, they are optimising for half the problem.

Run this as a structured process over three to four weeks rather than as a series of unrelated sales calls. It costs a few hours and prevents a year of wasted retainer.

Five patterns account for most of the disappointing outcomes we see when firms ask us to review an existing search arrangement.