The claim
A table, broken down by trade, giving an average cost per click and cost per lead for the United Kingdom. Roofing so much, plumbing so much, dental so much. It appears in proposals, on agency websites, and in the round-up articles that rank for “cost per lead UK”.
What we found
There is no published, independent, industry-level cost-per-click or cost-per-lead benchmark for the United Kingdom. Not from Google, not from a regulator, not from a trade body.
The closest anyone has come to producing one was not an agency. In its 2020 market study into online platforms and digital advertising, the Competition and Markets Authority obtained real figures from Google using its statutory information-gathering powers. Those figures were then redacted from the published version of the report, which states that information has been excluded from the published version. The regulator with legal powers to compel the data got it and could not publish it.
So when a UK agency publishes an industry CPL table, it is one of three things, and it is worth asking which:
- Their own client data. Legitimate and often useful, but it is a sample of whoever happens to have hired them, not a market average. Ask how many accounts and over what period.
- United States figures relabelled. The large public benchmark datasets are American. Converting the currency does not convert the market: different competitive density, different search behaviour, different auction.
- Copied from another agency's page, which copied it from a third. This is more common than you would like.
Why it matters to you specifically
Because a cost per lead you cannot achieve is worse than no number at all. If a proposal quotes you a figure drawn from a different country and your real market runs at twice it, the campaign is judged a failure from the first invoice while actually performing normally for your area.
And the variance within a single trade in a single country is enormous. Roofing in central London and roofing in rural Lincolnshire are not the same auction. Averaging them produces a number that describes neither.
What to ask for instead
A forecast for your trade, in your postcodes, with the assumptions written down. Google's own tools will give a bid range for the actual keywords in the actual area, and that range is a real input. Where an agency has genuinely relevant client data, ask what it is and over what period rather than accepting a rounded figure in a slide.
Our own position: we publish the sources we use and their limits, and where a number comes from our own accounts we say which sectors it is drawn from. Our measured advertising results come from scaffolding, roofing, glazing and surfacing work and we do not extend them to sectors where we have research rather than results.
Questions
Are agency CPL tables deliberately misleading?
Usually not. Most are assembled in good faith from whatever was available, and the available material is American or is somebody else's client data. The failure is in presenting it as a UK market benchmark rather than saying what it is. Asking where the number came from is a reasonable question and the answer tells you a lot about who you are dealing with.
Did the CMA really redact the figures?
The published version of the Competition and Markets Authority's 2020 market study into online platforms and digital advertising states that information it considers should be excluded has been excluded from the published version. The commercially sensitive pricing detail obtained from Google is not in the public document.
So how do you forecast a campaign without a benchmark?
From the keyword-level bid ranges Google publishes for the specific terms and area, the search volume behind them, and a conversion assumption stated openly so you can argue with it. That produces a range rather than a number, which is less impressive in a pitch and considerably more honest.