Decide when to fire a client after checking what the account contributes, what makes it expensive, and whether the relationship can fit your service. I'd repair avoidable work, price a supportable scope, and plan the replacement before choosing an exit. A weak account deserves a calculation before it gets a goodbye letter.
The MSP model sets a $2,000 minimum monthly client fee and a 50% fully loaded managed-service gross-margin target. Neither is a break-even test. An account can miss the target and still leave gross profit to pay costs you'll retain if it leaves.
Keep the Fee, Margin, and Fit Tests Separate
Review clients quarterly on margin and fit, as the model recommends. Keep the fee test separate. I'd cut a combined red score that conceals whether the actual problem is price, unnecessary work, or a responsibility the team can't support.
| Test | What to Check | What the Result Means |
|---|---|---|
| Minimum fee | Compare managed monthly revenue with the model's $2,000 client minimum. | Below the minimum calls for an account review; it does not establish a loss. |
| Delivery margin | Deduct burdened delivery labor and tools from the managed-service revenue. | Below the 50% target calls for cost and scope review; positive gross profit still contributes. |
| Fit | Record standard-stack exceptions, agreed scope, and unresolved working requirements. | Identify what must change and whether both sides will accept it. |
Use matching revenue and cost boundaries. Keep resold product apart from managed delivery, and include the owner's technical work at fair-market pay. The model's 10% EBITDA danger line applies to the whole business after operating costs, not to a client's service margin.
Carry a hypothetical account through the decision. Give the model's 20-seat client the $150 floor price: 20 multiplied by $150 is $3,000 a month. It clears the minimum, which tells you nothing yet about whether its workload fits the fee.
Work Out What Would Actually Change If the Client Left
For a starting cost comparison, use the US model's 700-seat staffing case at 1.0 tickets per seat per month, within its typical range of 1.0 to 1.4. That case has $109.99 of delivery labor per seat per month; tools add $22.50. For the hypothetical 20-seat account, calculate allocated cost as 20 × ($109.99 + $22.50) = $2,649.80. Calculate gross profit as $3,000 − $2,649.80 = $350.20, then divide by $3,000 for about 11.7% gross margin. These results come from the assumed allocation.
That remains positive contribution in this illustration. Before ending the account, separate supplier charges and other expenses you can remove from payroll and overhead you will keep. Add the transition work, then compare the result with continuing on the existing terms.
Released technician time can help deliver another signed agreement or complete work already sold. Write down the proposed use and check it afterward. Spare hours alone do not increase EBITDA.
Fix the Work That Makes the Account Expensive
Look at the tickets and time behind that cost. Separate client-environment issues from missing documentation, weak handoffs, or incomplete records inside your own team. The model favors a refresh that removes unnecessary hours before relying on a price increase.
For the hypothetical account, suppose a proposed refresh addresses recurring work, and the post-project record confirms a change from 1.0 to the model's well-run 0.5 tickets per seat per month. That's 20 tickets becoming 10. To calculate ticket hours at the model's weighted 48.30 labor minutes per ticket, use 20 × 48.30 ÷ 60 = 16.1 hours before and 10 × 48.30 ÷ 60 = 8.05 hours after; alert work is separate.
That workload change is an assumed result to verify, not a forecast that a refresh halves tickets. Check whether the promised work is still happening and whether requests moved elsewhere or were deferred. Use the client's actual records to establish the result.
Continue the hypothetical allocation using the well-run model's $103.10 per-seat cost. Calculate monthly cost as 20 × $103.10 = $2,062 and gross profit as $3,000 − $2,062 = $938. Dividing that by $3,000 gives about 31.3% margin. These are allocations from the model's different staffing cases, not a claim that this single refresh removes the difference from payroll.
Reprice the Scope You Can Deliver
The illustrated account has improved, but its allocated margin remains below the 50% target. For this hypothetical calculation, divide the allocated cost by the revenue share left after the margin: (20 × $103.10) ÷ (1 − 50%) = $4,124 monthly. Divide that result by 20 seats for $206.20 per seat, the price also shown in the model's pricing section for exactly 50% margin.
The model's $225 goal would instead quote $4,500 for those seats. For this hypothetical allocation, calculate gross profit as $4,500 − (20 × $103.10) = $2,438. Divide by $4,500 for about 54.2% managed-service margin. Check the actual scope and cost before presenting either figure as your client's price.
Start with the weakest accounts and explain the revised scope and fee. Don't claim the client loses money unless the calculation shows it. Keep the recurring rate intact in negotiation; the model puts concessions on a bonus item instead of cutting MRR.
A yes on price doesn't settle fit. Check the workload after the new terms start. Record the standard-stack changes and service boundaries both sides accepted, then inspect any condition that remains unresolved.
Make the Replacement Plan Earn Its Place
The model favors replacing weak revenue with better revenue so the total doesn't shrink. Match the sales plan to delivery and cash. An open opportunity can't pay a salary, and a signature doesn't prove the new scope earns the margin you expect.
The US model's $25,000 acquisition cost is blended sales and marketing spending divided by wins. It isn't an additional bill every time a client leaves. Use actual acquisition spending, expected starts, and onboarding commitments when evaluating the replacement.
Track the new agreement's managed revenue, delivery costs, and start-date confidence together. More signed MRR does not establish better contribution. In the hypothetical account, even the repriced fee needs a delivered-cost check before it can be called a successful repair.
Settle the Exit Decision and Handoff
My view is that an exit becomes the stronger choice when required changes cannot be agreed, the account cannot meet a supportable price and scope, or a documented fit problem remains unresolved. Compare that choice with the cost and risk of continuing. Do not require every situation to pass the same invented score or waiting period.
When you decide to leave, check the agreement's notice and transition obligations before promising dates. Assign an owner for the client communication and another for the operational handoff if they are different people. Give the team a plan they can execute.
Agree how documentation, authorized access, open work, and tool removal will transfer to the client or incoming provider. Confirm who owns each action and when support responsibility changes. Keep a record of the completed handoff and any remaining obligations.
Bring the result back to the account review: did costs actually fall, did the replacement start, and did the released capacity go where you planned? I'd use those answers to improve the next decision. The useful outcome is a better business after the departure, with the numbers to explain why.
mspoutpost.com/articles/when-to-fire-a-client-after-checking-margin-and-fit
Written by Owen Gallagher
MSP Outpost publishes information, not legal, financial, insurance, tax or compliance advice, and reading it creates no professional relationship. For decisions about your business, talk to a qualified professional.