There is a great deal written about what the AML reforms require and almost nothing about how to run them profitably. Vendors won't write it — they sell the automation. The professional bodies can't — fee-setting isn't their remit. So here it is.
Updated 2 August 2026
Most firms budget for the electronic verification check, which is the small and visible number. The expensive part is labour: collecting documents, tracing beneficial ownership through a trust with a corporate trustee and four beneficiary classes, and — above all — chasing clients who don't respond.
Chase time is the single most underestimated input. It is also the one that scales worst, because it lands on whoever is least able to refuse it, usually an administrator or the partner who picks up the phone.
The second thing firms get wrong is recurrence. Verification is not once per client — it attaches to engagements and to structures, and it re-triggers when circumstances change. A model that works for twelve clients a year fails at eighty.
No separate charge. The cost sits in overhead and is recovered, in theory, through general fee levels.
Suits: Firms with very few designated services a year, where the admin of billing it exceeds the fee.
Where it breaks: Works until it does not. Firms consistently underestimate recurrence — verification is per-engagement and per-structure, not once per client. This is the default that quietly becomes a five-figure annual write-off.
A set amount per new client or per entity verified, charged at engagement.
Suits: Most small and mid-size practices. Simple to explain, simple to systemise.
Where it breaks: A single fee across individuals and complex trust structures either overcharges the simple cases or bleeds on the complex ones. Most firms end up needing at least two tiers.
Individual, company, trust and multi-layer priced separately, reflecting the actual work in tracing beneficial ownership.
Suits: Firms with meaningful trust and structuring work — where the variance between easiest and hardest is large.
Where it breaks: Requires you to classify the client before quoting, which means your intake process has to be good enough to spot a three-layer structure before you have committed to a price.
No separate line. The company or trust establishment fee rises to cover the verification work it now triggers.
Suits: Firms where designated services are a defined product with an existing price.
Where it breaks: Cleanest client conversation — the price of that service went up. But it hides the cost, so nobody in the firm sees whether the work is actually profitable.
Pass through the electronic verification cost at cost, absorb the labour.
Suits: Firms wanting a defensible, transparent charge without a margin conversation.
Where it breaks: The disbursement is the small number. Chasing a client for trust deeds and beneficiary details is the expensive part, and this model recovers none of it.
Lead with the obligation, not with the fee.
The change is that the law now requires verification before you can act. That is the news; the fee is a consequence of it. Firms that open with the price sound like they are monetising a rule change, which is the impression you least want.
Tell existing clients before they hit it, not when they do.
A short note to the client base explaining what changed on 1 July and what they will notice costs you nothing and converts a nasty surprise into a piece of proactive advice. The firms handling this best treated it as a client communication, not a billing change.
Say what happens if they do not comply.
Not as a threat — as clarity. If verification cannot be completed you cannot provide the service, and it is kinder to say that upfront than to discover it three weeks into a job the client believes is underway.
Do not apologise for it.
The single most common failure. Firms introduce the fee with so much hedging that clients infer it is negotiable, and it becomes negotiable. It is a cost of doing regulated work, priced like any other.
Decide who is allowed to waive it.
If any partner can drop the fee to keep a client happy, the fee does not exist. Name who can waive it and require a reason — that alone preserves most of the revenue.
AML is the current example, but the pattern is general: every new obligation lands on practices as unpriced work first, and becomes priced work only if someone decides it should. Tranche 2 is unusual only in how many firms it hit at once.
The firms that will be fine in three years are not the ones with the best compliance software. They are the ones that treated a regulatory change as a service with a price attached, on the day it arrived, rather than as an overhead to be quietly absorbed until someone notices the write-offs.
We cover the commercial side of regulatory change — the part nobody else writes. One email a week, free.
Subscribe free →