Benchmarking accounting services before you appoint: scope and sources

Benchmark research on accounting services starts with a narrow question. Pick the service, the client type and the period first, then look for figures that match. Without that, a number is easily mistaken for a going rate or a quality bar. This guide shows how to scope the comparison and which sources to trust, so an accountant's quote can be checked.

Specify the decision before browsing

Name the decision first, for example choosing between three firms to prepare year-end accounts and a self assessment return. Then define the measure, population, service, entity size, region, period and unit before searching. Without a named client group and service, a figure may be read as something it was never designed to answer.

For turnaround, set when the clock starts and stops and allow for client delays. For errors, describe how serious a mistake must be and when it counts as detected. Use decision metrics and customer groups to settle these definitions.

Give each source a narrow job

Begin with official statistical releases and records, then inspect the method before relying on research from professional bodies such as ICAEW, ACCA, AAT or ICAS. For sector surveys, check the sample, the fieldwork dates and the weighting. A sharply stated statistic proves nothing if the source did not count the people or work you care about.

Keep a firm's own case study within its stated limits. To structure a comparison between named providers, use a matched-case approach, which sets the same scope and the same questions for each firm.

Separate population counts from service measures

Enterprise totals, headcounts of practitioners and amounts of accounting activity count different units, so do not add them or read one as a substitute for another. UK Business: Activity, Size and Location reports enterprise counts by area and industry classification, not a count of practitioners.

The Annual Business Survey gives turnover and approximate gross value added, including for accounting and bookkeeping. Read the ONS business services data with its coverage notes. Companies House filing guidance and filed accounts answer company-level questions, not a total for the market.

These mismatches are the ones to catch before comparing:

Use the benchmark to choose an accountant

This article does not publish benchmark fees or results. It shows how to scope one so that a quote or a promise can be checked. A useful scope is specific: for example, a sole trader with about forty sales invoices a month, a year-end self assessment return and quarterly VAT.

For market context, ONS data on accounting firms sits under Standard Industrial Classification 2007 code 69.20, accounting, bookkeeping and auditing activities, and tax consultancy. Use that code when reading the Annual Business Survey and UK Business: Activity, Size and Location. They describe the sector's size, not any firm's quality or price.

To find named providers to compare, use the register-based directories run by the professional bodies. ICAEW, ACCA, AAT and ICAS each publish a search for member firms, and these entries are a better starting list than an advert. Check the member's status on the body's own register before shortlisting.

Then ask each shortlisted firm to quote against the same written scope, and record the fee, what the fee excludes, and the stated turnaround in one row each. The provider research and the selection checklist list the checks to run before you appoint.

Keep a provider story inside its boundary

One firm's case study describes that firm's own clients. Similar services or a favourable outcome do not show that a broader group would get the same result, so apply a case study only to the provider and circumstances it describes, and label it that way.

Treat industry research as context rather than an England rate, and look for disclosed sampling, fieldwork and weighting. A survey of larger firms says little about a sole trader's choice of bookkeeper.

Do not treat different measures as equivalent

For example, a firm's average preparation time for small-company statutory accounts cannot be set against ONS turnover and called a performance gap. One measure records time and the other money, and neither supplies a target for the other.

Use examples like this to check definitions, not to translate one measure into another. A claimed average turnaround needs its own clock and population, as the guide to averages and invisible clocks explains.

Make the work reproducible

Record the publication, edition, table and method for every figure, so a later reader can tell whether a change reflects the sector or an altered source. Keep a measurement record of the sources you considered, what you included and why.

Flag inaccessible tables, discontinued series and revised releases, and preserve calculations with their source versions. If you later claim that an accountant caused an improvement, test the claim with attribution methods.

Leave the external comparison blank when needed

When no outside measure matches your definitions, say so and leave the external comparison blank. Compare the quotes you collected against the same written scope instead, and treat differences as prompts to ask questions, not as targets.

A handful of quotes cannot establish an England market position. They can show which firm explains its scope, exclusions and turnaround most clearly, which is the question the benchmark was meant to answer.