Engagement Acceptance: What to Check Before the First Billable Hour
Acceptance is the last moment a firm can decline work at no cost. Every check that happens after the first billable hour is a negotiation instead of a decision.
Every firm has a story about the engagement it should not have taken. The client who could not pay. The conflict nobody spotted until the other side complained. The scope that assumed a system access nobody had confirmed. In almost every case the information that would have stopped it was available before the work started, and the reason it did not stop anything is that nobody had a moment where stopping was the expected outcome.
Acceptance is that moment. It is not paperwork and it is not a formality to be completed in parallel with mobilisation. It is the gate where a firm decides, deliberately and on the record, that it is willing to be associated with this client and this piece of work. The cost of declining at acceptance is the proposal effort already spent. The cost of declining three months in is the fee, the team, the relationship, and sometimes the professional exposure.
The five checks that decide whether work can start
Different professions regulate this differently, but the substance converges on five questions. A firm that can answer all five in writing before mobilisation is in a defensible position. A firm that answers them afterwards is explaining itself.
- Independence: is there any financial, personal or business relationship that would compromise the firm's objectivity, or appear to? Appearance matters as much as fact, because the test applied afterwards is what a reasonable observer would conclude.
- Conflicts: is the firm already acting for a party whose interests oppose this client's? This includes matters closed recently, because knowledge does not expire when an engagement does.
- Credit and commercial standing: can the client pay, and on what terms? A credit check is not an insult; it is the difference between a receivable and a write-off.
- Sanctions, politically exposed persons and adverse media: is there anything that would make this client one the firm cannot lawfully or reasonably serve?
- Capacity and competence: does the firm actually have the people, the skills and the availability to deliver what has been described, on the dates promised?
Risk tiering, so the checks are proportionate
Running every check at full depth on every engagement is how acceptance becomes a rubber stamp: the process is too slow, so people route around it. The answer is to tier the work by risk and vary the depth of the checks rather than whether they happen at all.
A useful tiering considers the client's regulatory status, whether the output will be relied on by a third party, the fee at risk, the jurisdiction, and whether the engagement touches personal or price-sensitive information. A low-tier engagement might need a name screen and a partner sign-off. A high-tier engagement might need a second-partner review, a documented independence declaration from every team member, and a written conclusion on why the work can proceed.
What to do when a check fails
Most acceptance frameworks describe how to pass. The valuable part is what happens when something fails, because that is when judgment is required and when firms are most tempted to improvise. Three outcomes should be available and each should be recorded differently.
The first is decline, which needs no justification beyond the failed check and should be easy to choose. The second is accept with safeguards, where the risk is real but can be managed: an information barrier, a different partner, a scope limitation, a payment on account. The safeguard has to be specific and someone has to own it, because a safeguard nobody implements is a decision to accept the risk without saying so. The third is escalate, where the decision belongs above the engagement partner, and the escalation should name who decides rather than leaving it to circulate.
Recording the decision so it survives a challenge
An acceptance decision is only as good as the record of it. A year later, when a regulator, an insurer or an opposing party asks why the firm took the work on, the answer has to be a contemporaneous document rather than a recollection. That document should show what was checked, what was found, who decided, and what safeguards were imposed.
The most common failure here is a record that shows only the conclusion. "Accepted" with a signature proves that someone signed, not that anyone looked. A record that lists the five checks, their outcomes and the date each was performed is dramatically more defensible and takes very little longer to produce, particularly if the checks are performed in a system that keeps the evidence attached to the engagement rather than in a mailbox.
Re-acceptance: the check most firms skip
Acceptance is usually treated as a one-time event, which is a mistake for any engagement that runs longer than a few months or recurs annually. Clients are acquired, ownership changes, people move between firms, and a conflict that did not exist at the start can appear halfway through. Continuance is a real decision and deserves a lighter version of the same gate.
A practical rule is to re-run the conflicts and sanctions screens at each anniversary and whenever a material change is known: a change of control, a new adverse media hit, a significant scope extension, or the arrival of a new party to the matter. The cost is small and the alternative is discovering a conflict from the other side.