M&A

Losing a Top Customer During Due Diligence

Your biggest customer just told you they're leaving, and you've already signed an offer from a buyer. Now the buyer will want to pay less, and they'll often ask to pay part of the price later. That part gets paid only if your sales reach numbers the buyer picks. Sellers rarely get all of it. SRS Acquiom is a firm that handles these payments. It studied 1,174 company sales in 2026 and found sellers got about 21 cents of each dollar promised for later.

If the lower price is fair, arguing won't change it. But you can decide how the money gets paid. I'll show you how to test whether the new number is fair. Then you can make the later payment depend on the customers who stayed, not the one who left.

What customer churn during due diligence does to your price

The buyer did not change their mind about your company. They changed one number in their spreadsheet.

Start with how the offer was built. Most buyers of a SaaS company price it as a multiple of yearly recurring revenue. If you bring in $4 million a year and the buyer offers 5 times that, the offer is $20 million. That offer usually arrives as a letter of intent, which is simply a signed offer letter that sets the price and gives the buyer a few weeks of exclusive access to check your business.

That checking period is due diligence: the buyer tests every claim you made before they wire money. When a big customer gives notice during that window, the revenue the buyer priced is no longer there. The buyer reruns the same spreadsheet with the smaller number, and the price falls. When a buyer asks to lower the price after the offer letter is signed, the industry calls it a retrade.

Here is the part founders miss. The buyer is usually worried about 2 things, and only one of them is in the spreadsheet. The first is the revenue that left. The second is the fear that it was the first customer of several to leave. That second worry has no number attached, so buyers price it with a round figure. That round figure is where your room to negotiate lives.

The Same Math Test: how to tell if the new price is fair

There is one test to run on any reduced offer. I call it the Same Math Test: take the revenue that left and multiply it by the same multiple the buyer used in the offer letter. That is the fair cut. Anything above it is the buyer pricing fear.

Here is an illustration with round numbers, not a real deal. Assume the offer was $20 million, at 5 times $4 million in yearly revenue. Assume the customer who left paid $400,000 a year.

Step Amount Running total
Price in the offer letter $20,000,000 $20,000,000
Lost revenue: $400,000 times the same 5 times Less $2,000,000 $18,000,000
Extra cut for fear that more customers leave Less $2,000,000 $16,000,000
Buyer’s new offer $16,000,000

The first $2 million follows the buyer’s own math. Fighting it usually wastes the weeks you have left. The second $2 million is a bet that other customers will follow. That bet is the part you can reshape.

One honest exception. Losing a top customer can change the business the buyer is left with, not just its size. If your next largest customer now makes up a much bigger share of revenue, or your yearly growth falls, a buyer can fairly argue for a lower multiple on the revenue that remains. That argument needs numbers too. Ask the buyer to show which figure moved and by how much. A cut backed by a real change in concentration or growth is math. A round number with no figure behind it is fear, and fear is what you put into the later payment below.

The test also tells you when a cut is not about the customer at all. A buyer using the lost customer as an excuse to buy cheaper is running what people call an opportunistic retrade. These are the signs to look for:

  • The cut is far larger than the Same Math number, and the buyer will not show the math in writing.
  • The cut arrives bundled with a list of small, unrelated issues found weeks ago.
  • The buyer refuses any structure that pays you if their fear turns out to be wrong.
  • The buyer asks for more exclusive time in the same message.

The third sign matters most. A buyer who truly fears more customers will leave should be happy to pay you only if they do not. A buyer who refuses that bet is telling you the fear is not the reason.

Key takeaway

The fair cut is the lost revenue times the multiple the buyer already used. Concede that part. Negotiate the rest.

Two founders, the same lost customer

Picture 2 founders with the company from the table above. Both have a signed offer at $20 million. Both lose the $400,000 customer in week 4 of the buyer’s checks. Both get the same email asking for $16 million.

Path 1: argue the price

The first founder fights. The customer was unusual, the founder says, and nobody else is at risk. The buyer listens, and holds. Their math on the first $2 million is correct, and they know it. Meanwhile the exclusive window keeps running, the other interested buyers have moved on, and the founder’s team is exhausted. After 3 more weeks, the founder signs at $16 million, all in cash. The fear discount is now permanent, whether or not any other customer leaves.

Path 2: rebuild the deal

The second founder concedes the first $2 million the same day, in writing, with the Same Math shown. Then the founder makes an offer on the second $2 million: pay it later, and only if the customers who stayed actually stay. That later payment is what deals call an earnout, which is simply part of the price paid after closing if agreed targets are met. Our guide to how earnouts work in SaaS acquisitions covers the basics.

This founder builds the earnout in 3 specific ways, each one closing a gap the usual version leaves open.

  1. Named accounts only. The payment depends on the revenue from the 10 largest remaining customers, listed by name. It does not depend on company profit, new products or anything else the buyer controls after closing.
  2. A short clock. The test runs for 12 months, not 3 years.
  3. A clear formula. The test compares what those 10 accounts are billed in the 12 months after closing with the 12 months before. At 95% or more, the founder gets the full $2 million. At 80% or less, the founder gets nothing. In between, the payment rises in a straight line, so 90% pays about $1.33 million. Lost seats and price cuts count against the founder. Price increases count for the founder. If the accounts grow past 100%, the founder gets a bonus, up to an agreed cap.

A true two way version goes one step further. Part of the price is held back at closing in a separate account, and released to you or returned to the buyer depending on what those 10 accounts do. Buyers accept it more easily, because it protects them as well.

The founder also counts new business. Say the company signed $250,000 in new yearly contracts since the offer letter. At the same 5 times, that is $1.25 million of value. Check first that the offer letter did not already include those contracts in its revenue number. If it did not, the founder asks for that value in cash, on the buyer’s own math.

What the founder gets Path 1: argue Path 2: rebuild
Cash at closing $16,000,000 $16,000,000, plus a request for up to $1,250,000 for new contracts
Later payment $0 Up to $2,000,000 in 12 months
What the later payment depends on Nothing 10 named customers the founder already knows well
If no other customer leaves The $2,000,000 is gone The $2,000,000 is paid

Path 2 is not risk free. It turns a certain loss into a bet on customers you know better than the buyer does. That is a bet worth taking when the remaining accounts are healthy, and a bad one when they are not.

Why most earnouts disappoint, and how this one differs

A founder who has read about earnouts should be suspicious here, and the data says why. In SRS Acquiom’s 2026 study, sellers collected about 21 cents of every dollar promised in earnouts outside life sciences. As DealLawyers.com summarized the study, 52% of deals with an earnout paid at least something, but the typical deal paid just 17% of the maximum. The 2 numbers differ because one adds up every dollar across all deals, and the other looks at the deal in the middle.

66% vs 36%

Typical share of the maximum earnout paid when the test ran for 1 year or less, compared with tests longer than 3 years. SRS Acquiom 2026 M&A Earnouts and Claims Study, as reported by DealLawyers.com.

Those numbers describe the usual earnout: long, tied to profit or growth targets, and measured with the buyer’s books after the buyer takes control. Every one of those features moves the outcome out of your hands. The named account version cuts the clock short and ties the money to customers you have served for years. Earnouts are also becoming more common. SRS Acquiom found earnouts in 24% of private company deals in 2025 outside life sciences, up from 19% in 2014. So expect a buyer to raise the idea, and decide the shape before they do.

There is one more gap to close: information. After closing, the buyer runs the billing system. In September 2026, the Delaware Court of Chancery decided Winton v. The North Highland Company, an earnout dispute in which the buyer held back the data the sellers needed to claim their payment. The court wrote that earnouts “can leave sellers vulnerable after they cede operational control to the buyer.” It ordered the buyer to provide periodic reports, but not open access to its systems. The lesson for you: write into the agreement exactly which report you get, how often and by what date.

How to answer customer churn during due diligence: the protocol

When the notice arrives, work through these steps in order.

  1. Get the facts before the buyer does. Get the customer’s notice in writing, the reason they gave, and the date their payments stop. A customer who leaves after a merger of their own is a different story from one who left for a competitor.
  2. Tell the buyer within days. Hiding it is worse than losing it. The final purchase agreement will contain your written promises about your customers, and a broken promise becomes a claim against your money after closing. In the same SRS Acquiom study, 28% of deals had at least one such claim. Our guide to escrow holdbacks and indemnification explains how those claims get paid.
  3. Run the Same Math Test on the buyer’s new number, and send your math in writing.
  4. Concede the same math part in cash. Fight only the fear discount.
  5. Offer the fear discount as a named account earnout: 10 named customers, 12 months at most, paid in proportion to the revenue they keep, with a bonus if they grow.
  6. Write the information rights: a monthly billing report on the named accounts, due by a set day each month.
  7. Count new contracts signed since the offer letter, at the same multiple.
  8. Extend exclusive time only as long as the new terms need. Our guide to exclusivity extensions covers how to keep that clock short.

If the buyer’s cut is far above the Same Math number and they refuse every structure, treat it as a price grab. That is the moment to call the other buyers from your process. The terms that tend to shift in this phase are covered in our guide to what changes between the LOI and the definitive agreement.

You rarely talk a buyer out of a fair cut. You can almost always change what the cut is made of.

Frequently Asked Questions

Can a buyer change the price after the LOI?

Yes. In most deals the price in the letter of intent is not binding, so a buyer can ask for a lower price at any point before the final purchase agreement is signed. A fair request follows the buyer’s own math: the lost revenue times the multiple used in the offer.

What is an opportunistic retrade?

An opportunistic retrade is a price cut the buyer pushes for mainly because the seller has lost negotiating power, not because the business changed. The signs are a cut far above the lost revenue times the original multiple, no written math, and a refusal of any structure that pays the seller if the buyer’s fears prove wrong.

Should I walk away from a retrade?

Walk away only if the cut is far above the Same Math number and the buyer refuses every alternative structure. If a top customer actually left, a cut equal to that revenue times the original multiple is fair, and the better move is to accept it and negotiate how the rest of the gap is paid.

Should I accept an earnout when selling my business?

Accept one only if it is short and tied to results you can see and influence. In SRS Acquiom’s 2026 study, earnouts of 1 year or less paid a typical 66% of their maximum, compared with 36% for those longer than 3 years.

Do I have to tell the buyer that a customer is leaving?

Yes, and quickly. The purchase agreement contains written promises about your customers, and a broken promise becomes a claim against your proceeds after closing. SRS Acquiom found that 28% of deals in its 2026 study had at least one claim after closing.

Next Steps

If a big customer just gave notice in the middle of your sale, we will run the Same Math Test on your buyer’s new number and show you which part of the cut is fair.

Book a Free Value Assessment