Somewhere in your client list is an account that quietly loses you money. Not the one that pays late, that one is annoying but fixable. The one that loses money is the one where the hours spent, the scope creep absorbed, and the emotional tax of dealing with them exceed what they pay you, month after month, and you have not run the math because you are too busy doing the work.
The math nobody runs
Most operators track revenue by client. Far fewer track cost by client, which means a $4,000 a month retainer that consumes 30 hours of senior time looks identical on the P&L to a $4,000 retainer that takes 8. One is healthy. The other is a loss you are financing with your other accounts.
The same pattern shows up outside agencies. A nonprofit program that technically breaks even on its grant but eats two staff members’ bandwidth all quarter. A ministry partner relationship that generates goodwill but zero budget support after three years of promises. A sponsor that pays for a table at your event but demands white-glove service worth triple the sponsorship. The dollar amount is not the problem. The ratio is.
The fix is not better willpower about saying no. It is a repeatable way to measure an account, decide what to do about it, and exit cleanly if that is the call, without torching the relationship or the quarter’s cash flow.
Three numbers that tell you if an account is bleeding
You do not need a cost-accounting system to see this. You need three numbers per client or account, checked quarterly, and an honest log of hours.
- Effective hourly rate. Take what the account paid last quarter and divide by actual hours logged against it, including the owner’s time in calls, revisions and damage control. If your blended target rate is $120 an hour and an account nets $60, you found it.
- Scope drift. Compare what the original agreement or grant covered to what you are actually delivering now. A retainer that started as “monthly content” and now includes ad-hoc design requests, rush turnarounds and free strategy calls has drifted without a corresponding price change.
- Collection friction. Count how many invoices in the last six months needed a reminder, a resend, or a renegotiated due date. Chronic friction is a signal independent of the dollar amount, and it usually gets worse, not better, as the relationship ages.
Run these three on every account once a quarter and the bleeders surface on their own. You rarely need a fourth metric. The ones that fail all three at once are the ones to act on first.
Why operators keep bad accounts longer than they should
Three reasons show up constantly, and none of them are really about the client.
- Sunk cost on the relationship. You have history, you like them personally, or you brought them in yourself and feel responsible for how it turned out.
- Revenue anxiety. Losing a line on the P&L feels worse in the moment than the slow bleed that caused by keeping it, even when the math says otherwise.
- No plan for the gap. It is easier to tolerate a bad account than to sit with an empty slot, even a slot that costs you nothing to leave empty.
The second and third reasons are solvable with process, which is the point of this article. The first one you solve by deciding in advance what the exit looks like, so when the moment comes you are following a plan instead of negotiating with yourself.
The decision, before the exit
Not every underperforming account should be fired. Some should be repriced, some rescoped, and only a minority actually need to end. Work through the options in this order before you reach for the exit.
- Reprice first If the work is good and the relationship is sound but the rate has not kept up with the scope, the first move is a renegotiation, not a goodbye. Bring the effective hourly number to the conversation. Most reasonable clients respond better to “here is what this actually costs us to deliver” than to a vague request for more money.
- Rescope second If the client will not move on price, move on scope. Cut the ad-hoc requests, the free revisions, the after-hours availability, back to what was originally agreed. This alone fixes a surprising share of bleeding accounts, because the drift was the real cost driver, not the base rate.
- Set a trial window Give the repriced or rescoped arrangement one full billing cycle to prove out. Put a date on the calendar to review it rather than leaving it open-ended, so the decision does not quietly evaporate.
- Exit if neither works If repricing and rescoping both fail, or if the account is toxic in ways no price fixes (the sponsor who berates your staff, the client who disputes every invoice), move to the exit plan below. Do not run a third attempt hoping it sticks.
Exiting without wrecking cash flow
The actual termination is the easy part. The planning before it is what protects you.
- Give notice on your terms, not theirs. Thirty to sixty days is standard for a retainer or ongoing service. It gives you a real runway to backfill the slot instead of absorbing an abrupt gap.
- Collect before you cut. Settle outstanding invoices, final deliverables, and any deposit reconciliation before the relationship ends, not after. Collection leverage drops to near zero once the work stops.
- Protect the pipeline. Tell your two or three warmest leads or your referral sources that you have capacity opening up. A bad account rarely needs to be replaced dollar for dollar if the hours it freed go toward higher-margin work.
- Document the reason, briefly. One paragraph in your records on why the account ended protects you if the client disputes the decision later, and it saves the next person who evaluates a similar account from relearning the lesson from scratch.
We kept a sponsor for three years because the logo looked good on the program. The year we finally dropped them, we replaced the revenue with two smaller sponsors who paid on time and asked for nothing we hadn’t agreed to in writing. Net income went up and my blood pressure went down.
Event director, regional conference series
Building the check into your normal rhythm
This only works if it happens on a schedule, not as a reaction to a bad month. Pair the quarterly three-number check with whatever account review or weekly pipeline meeting you already run. If your invoicing and project tracking live in one system, the effective hourly rate and collection friction numbers are usually a five-minute pull rather than a spreadsheet reconstruction project, which is the difference between actually doing this quarterly and promising yourself you will get to it eventually.
The same discipline applies whether you call it a client, a grant, a sponsor or a partner church. Anywhere money and hours cross a relationship boundary, the same three numbers apply, and the same decision tree, reprice, rescope, trial, exit, holds.
Key takeaways
- Track cost per account, not just revenue per account. Effective hourly rate, scope drift and collection friction surface bleeders that look fine on a revenue-only P&L.
- Reprice and rescope before you exit. Most underperforming accounts are fixable without ending the relationship.
- Plan the exit like a project, not a reaction. Notice period, collection before cutoff, and a pipeline plan for the freed capacity all have to happen before you send the goodbye email.
- Run the check quarterly. A scheduled review catches a bleeding account in one quarter instead of four.
Questions to ask your team
What if the bad account is also my biggest account by revenue?
Size makes this harder but not different. Run the same effective hourly rate math. A large account at a thin margin is often more dangerous than a small one, because losing it hurts more and the temptation to keep tolerating bad terms is stronger. Reprice conversations with your biggest account should happen earlier, not later.
How do I know if it is the client or my own delivery that is the problem?
Check scope drift first. If the original agreement was reasonable and the work has quietly grown without a price change, that is on your side to fix with a clearer scope document next time, not a reason to blame the client. If the client disputes clearly delivered work or renegotiates terms after the fact, that is a client problem.
Is it ever worth keeping a money-losing account for strategic reasons?
Yes, sometimes. A flagship logo, a case study you need, or a relationship that opens other doors can justify a loss leader, but only if you name it as one on purpose and set a time limit. The failure mode is treating every bleeding account as if it might be strategic so you never have to make the call.
What is a reasonable notice period if there is no contract specifying one?
Thirty days is the floor for anything billed monthly. Sixty is safer if the client depends on you for something hard to replace quickly, like ongoing giving processing or event logistics close to a date. Put a notice clause in every new agreement going forward so this is never a guess again.
