Accountants sit on something genuinely rare: comparative financial data across dozens of businesses in the same market. Every firm eventually has the idea of publishing it, because it would be the most useful content in the local market by a distance.
It is also the content most likely to go wrong, and the constraints are stricter than general client confidentiality. Tax practitioners in particular sit under a separate regime governing return information, in which marketing counts as a use.
General professional confidentiality is the rule most firms have in mind. In the United States there is an additional statutory regime specifically covering information furnished for the preparation of a return, under which disclosure or use generally requires specific, written, advance consent. Other jurisdictions have their own equivalents.
The important implication is that consent obtained for one purpose does not extend to marketing, and a general engagement letter clause is unlikely to be the specific consent contemplated. Treat anything sourced from a return as unusable until someone competent confirms otherwise for your jurisdiction.
The instinct is to strip the name and publish the story. In a local market this rarely works. "A restaurant client who came to us three years behind on filings" identifies a specific business to anyone who knows the town, and the people most likely to work it out are exactly the audience you are publishing for.
The test is not whether you named them. It is whether anyone could reasonably recognise them. Sector, size, location and timeframe together are usually enough, and removing only the name removes almost nothing.
Genuine benchmark content is possible, but it needs to be built to a standard rather than assembled from memory. That means a sample large enough that no single business is identifiable within it, figures that cannot be reverse-engineered to one client, and clarity about what the sample is and is not.
There is also a commercial question worth asking before the compliance one. Publishing that your local competitors' clients run at a given margin is useful to prospects and occasionally uncomfortable for the clients who supplied it. If a client would be unhappy to learn their numbers were in the pool, consent was not really informed.
Almost everything a firm wants to say through client examples can be said through the pattern instead. "Businesses in this sector are often caught by this rule in their second year" carries the same expertise as a story about a specific client, and needs no consent from anyone.
Publicly available data is also underused. Official statistics, published sector figures and the regulator's own guidance can be interpreted and made useful, which is the actual value you add, and none of it touches a client file.
Publish the pattern, never the client: return information sits under its own consent regime, and in a local market anonymising almost never anonymises.
Noian writes, schedules and publishes this kind of content for accounting firms, with every post held for your approval first.
See how it works for accountants