Every other diligence stream measures what the agency is; retention engineering decides how much of it the buyer keeps. The financial model assumes the revenue persists, but revenue in an agency is carried by people, and a handful of them can leave in the first week and take a meaningful slice of the book with them. Retention is the work of identifying exactly who those people are and making it expensive — financially and psychologically — for them to leave.
§ 01 · Who to keep, and the cost of losing themThe Pareto target.
| Role / risk | The number |
|---|---|
| Mid-level account manager | 50%–100% of salary to replace |
| Top producer | 150%+ of salary — plus lost commissions and client attrition |
| Top producer leaving day 3 | Up to 20% of revenue can walk if retention isn't engineered |
| Retention target | The top 20% of staff who control 80% of relationships |
Retention spending should follow the Pareto rule, not fairness. Replacing people is costly — 50%–100% of annual salary for a mid-level account manager (recruiting, training, lost institutional knowledge, team disruption), and 150%+ for a top producer once lost commissions and client attrition are factored in — and a top producer walking on day three can take up to 20% of revenue with them. So the target is the top 20% of staff who control 80% of the client relationships: the top three to five producers, the lead account managers on the largest clients, the operations manager, and the compliance or licensing lead if that's a separate role. The discipline is restraint — do not extend retention bonuses to all staff, because that creates entitlement, inflates the cost pool, and spends money on people who were never going to leave.
§ 02 · Sizing the bonusTo the risk, not the headcount.
Size a retention bonus at 10%–20% of annual salary, paid over 12–18 months — so a $120K producer earns a $12K–$24K bonus, a $65K operations manager $6.5K–$13K. Three factors set the exact number: the individual's flight risk (a flight-risk producer warrants the 20% end, a stable one the 10%), the deal's total cost and margin (a tight multiple argues for restraint), and the local market rate. Fund it from the purchase price before close, not as a reactive raise after.
The bonus is a calibrated instrument, not a flat perk. The 10%–20%-of-salary band over 12–18 months gives a working range, and the three sizing factors place each person within it: flight risk first (the producer whose comp drops under the new plan needs more than the one whose doesn't), then the deal's margin (a thin multiple can't fund generous bonuses for everyone), then the geographic market rate. The timing matters as much as the amount — funding retention from the purchase price pre-close signals stability from day one, which beats the reactive raise a buyer scrambles to offer after a producer has already started looking, because the reactive raise costs money and doesn't reliably retain. The comp gaps that create the flight risk in the first place are diagnosed in producer compensation.
§ 03 · The payout structureBuilding multiple stay-moments.
How a bonus vests shapes whether it actually retains. Cliff vesting pays 100% at a single date (say 12 months) — simple, but it creates one exit opportunity the moment it pays. Graded vesting — for example 25% at 6, 9, 12, and 18 months — builds multiple "stay moments," which is better for high-risk departures because each milestone re-anchors the producer. A two-milestone middle ground (50% at six months, 50% at twelve) splits the difference. The most structurally protective option is a forgivable loan: a lump sum paid upfront and forgiven monthly — a $24K loan forgiven at $1K per month over 24 months, so a producer who leaves at month 10 owes $14K back. The forgivable loan carries three advantages a bonus doesn't: a clawback right if the producer competes or violates a non-solicit, deferred income recognition for tax planning, and a clear signal of mutual commitment. Matching the structure to the producer's risk profile is the craft.
§ 04 · The rehire momentThe highest-leverage covenant cure.
In an asset purchase, all employees are terminated by the seller on the closing date and rehired by the buyer the next day — and that reset is the single highest-leverage moment in the lifecycle for fixing what the seller never did. The mechanics demand care: the seller provides an accrued-PTO schedule per employee, the buyer prepares new offer letters and employment agreements, and state-law compliance on final paychecks, PTO payouts, and notice requirements has to be right. But the strategic prize is that the new employment agreements at rehire are the moment to put non-compete, non-solicitation, and non-piracy covenants in place if the seller never had them — turning a routine HR step into the covenant cure. Retention also rests on getting comp right, benchmarked against three reference points: the buyer's own equivalent roles, the regional market rate, and national industry benchmarks. A producer sitting materially below market is one of the strongest 12-month turnover predictors, so the rehire is the moment to close that gap too. The covenant side of this moment is detailed in restrictive covenants.
◆
Terminology on this shelf
- Replacement cost
- 50%–100% of salary for a mid-level account manager; 150%+ for a top producer.
- Pareto retention target
- The top 20% of staff who control 80% of relationships — not the whole roster.
- Bonus sizing
- 10%–20% of salary over 12–18 months, set by flight risk, deal margin, and market rate.
- Graded vesting
- Paying in tranches (e.g., 25% at 6/9/12/18 months) to build multiple stay-moments.
- Forgivable loan
- An upfront lump sum forgiven monthly — adds a clawback right and a commitment signal.
- The rehire moment
- The asset-purchase re-papering — the highest-leverage point to install missing covenants.