How to read these case studies

It means someone loaded the live public website on the date shown and confirmed what was and was not there. A 404 on a contact page, a footer copyright reading 2013-2020, an apex domain returning 403, two Google Business Profiles with different review counts. These are observations of public facts, stamped with the date they were made. Websites change — if a client has fixed something since, the observation was still true when it was made, and the date on the page tells you when that was.

Every one of the 55 benchmark figures we cite is published by a named third party, carries the year it was published, and links out to the source. None of them are ours. We use them because a category median is the only honest way to say whether a specific account is doing well or badly — "$149 cost per lead" means nothing until you know the category average across 816 contractors is also $149.

A target is what the programme was set up to move a number to, derived from the published category benchmarks. A result is what the client's own reporting says actually happened, reconciled against their ad platform exports, call recordings, and booking or policy system. We label these differently on every page and we never print one in the other's place. A page showing targets says so above the table.

Because reconciling a client's reporting is a real piece of work and it happens on their reporting cycle, not ours. Until a figure has been pulled from the client's own systems and matched to source, it stays labelled as a target. We would rather show you an honest target than an impressive number we cannot substantiate on request.

For their own account, every client can, and does — the proof assets listed on each case study are the exports they already receive. For someone else's account, no: that is their commercial data and it is not ours to hand out. What you can check independently is every benchmark on the page, because each one links to its published source, and every "what we verified" observation, because the site is public.

No. Nobody can, and an agency that offers to should worry you. What we commit to is the measurement standard on this page: named sources for every benchmark, dated observations for every claim about a site, and a clear line between a target and a reported result.