Why copywriting services benchmark research needs a stated method

What to take away

What a benchmark must publish before you trust it

Start with four disclosures: sample, fieldwork dates, exact question wording, funder. A figure missing any one of them is a claim, not a benchmark.

Method matters because briefs vary so widely. One team buys a landing page rewrite; another buys a year of editorial. Hide the method and you cannot tell whether a gap between two figures comes from the work or from the sample.

Benchmark research means repeatable measurement of the same thing across many organisations, published with enough method to judge it. A survey of several hundred marketing managers, repeated in the same weeks each year, is a benchmark. One agency's average project fee is not.

Where the real UK numbers come from

For sector context, the ONS business services statistics hub publishes turnover and output data for professional services, the nearest official frame around copywriting.

For rates, the ONS employment and earnings data gives median pay by occupation and region from the Annual Survey of Hours and Earnings. Use it to anchor the cost of in-house writing before you compare agency quotes.

For older figures, the ONS inflation and price indices let you restate a 2019 rate in current money, so a stale figure stops looking like a bargain.

None of these measures copywriting directly. Treat them as the outer frame, then log every outside figure with a source and a date. The measurement and reporting guide sets out how to keep that trail clean in client reports.

Example: a model method statement

Here is the shape a stated method should take. The figures are illustrative examples of the format only.

Sample: 120 marketing managers in UK firms with 10 or more staff, drawn from a purchased B2B panel. Fieldwork: 3 to 21 March 2027. Question asked: "Thinking about the last piece of copywriting you bought from an outside supplier, how many working days passed between brief approval and first draft?" Median answer: 9 working days. Funder: none. Precision: plus or minus 8 percentage points at 95 per cent confidence.

Copy the shape, then fill it with your own sample, dates and wording. Publish it beside the figure, not in a footnote.

How to build a before and after table

Keep the unit of measurement identical on both sides. If the before figure is cost per enquiry, the after figure must be cost per enquiry too. Swapping units halfway is the commonest way a report misleads its own author.

Table: illustrative figures, shown to demonstrate format only.

A reader's own figures should come from their own records: brief approval dates, project management timestamps, invoiced costs and enquiry logs, all pulled over the same window. Where the after figure depends on which channel gets the credit, the copywriting services attribution methods piece shows what each model does to the number.

Four questions that expose a weak benchmark

Ask who was in the sample, when the fieldwork ran, what was asked and who paid. A benchmark that fails any of the four is a talking point, nothing more.

How to use a benchmark without overclaiming

Use benchmarks to frame a range, not to promise an outcome. One can show that a typical brief cycle sits inside a band. It cannot show what your team will achieve next quarter.

Write the assumption in full, give the date of the benchmark behind it, and name the owner of the number. Do it once, in one place, rather than in three sections.

Where a figure drives a budget line, keep the working. If a later comparison is skewed by a change in definition, the common measurement mistakes article explains how quickly that happens and how to correct it.

Common questions

Is a benchmark the same as a forecast?

No. A benchmark describes what was measured in a past period; a forecast states what you expect. Keep them in separate columns so readers can see which is evidence and which is assumption.

How often should benchmark figures be refreshed?

At least annually, and whenever the method changes. A figure with no publication date cannot be compared with anything, so record the date beside the number.

Can I quote a competitor's published rates as a benchmark?

Only if you know the sample, the dates and what was included. A rate card is a price list, not a benchmark, and it rarely covers the scope you are buying.

Where do I start if no UK benchmark covers my sector?

Run a small repeatable survey of your own and publish the method, even with a modest sample. A short internal study you can explain beats a number you cannot.