Can the government still audit us after we buy this company

The situation

Tigist asked the question in the first diligence call, before her team had opened a single binder: if this deal closes and the government comes back in two years to audit billing from before we owned the company, can they touch our records to do it. She led a private equity-backed buyer putting together a roll-up of clinic operators, and the target in North York was the largest single acquisition in the program so far, a chain of clinics built over a decade by Biniam. The transaction sat in the range of fifty to eighty million dollars, funded through a mix of equity from Tigist's backers and acquisition debt, and the clinics carried a portfolio of government service contracts that made up a meaningful share of their revenue.

Those contracts were the reason the business was worth buying. They were also, Tigist suspected, the reason the deal needed a harder look than a typical private acquisition. Government service contracts in healthcare-adjacent sectors routinely carry audit rights that let the funder verify billing accuracy years after services were delivered, and Biniam's agreements were no exception. What worried Tigist was not the audit right itself. It was a clause, buried in a schedule Biniam's team had not flagged as significant, that extended the funder's audit access to the successor entity's records generally, not just the records tied to the specific claims under review.

Read broadly, that clause meant a government auditor investigating billing from three years before the sale could ask to see the buyer's own post-closing financial systems, not just the historical files inherited from the seller. For a private equity-backed buyer running dozens of portfolio companies through shared back-office infrastructure, that was not an abstract worry. It was a door into systems that had nothing to do with the clinics being purchased.

Biniam, for his part, had run the clinics well and had no reason to think the audit clause was a problem; his own accountant had never been asked to interpret it in a sale context. Goran, the investment committee member overseeing the acquisition on Tigist's side, wanted the diligence finished and the deal signed before a competing bidder circled back. The clock and the clause were pulling in opposite directions.

The legal question

The question Tigist asked was really two questions stacked together, and separating them was the first job. The first was a contract interpretation question: did the audit clause, as written, actually reach the buyer's unrelated records, or did it only look that way on a fast read. The second was a structuring question: if the clause meant what it appeared to mean, what could the parties do about it before closing rather than after.

On the interpretation side, the audit clause used broad language, granting the funder access to books and records relevant to the contract and to the party performing it. The ambiguity sat in the word party. Once the sale closed, the party performing the contract would technically be the buyer, and a funder relying on the literal text could argue that gave it standing to look at whatever records the buyer chose to keep on the same servers as the clinic's billing data. Government funding agreements of this kind are typically drafted to protect the funder's ability to verify public money was spent correctly, and the language tends to be broad by design, not by accident. That made a narrow reading harder to defend on the strength of the words alone.

The second question turned out to matter more than the first. Even a defensible narrow reading would not stop a funder from asserting the broad one in an actual audit, and defending that assertion after the fact, mid-audit, with a government counterparty, was a fight nobody wanted to have with borrowed money already deployed. The real protection had to be built before closing, not argued after it.

That reframing changed how the diligence team approached the rest of the file. Instead of treating the audit clause as a legal risk to be priced into the purchase agreement and forgotten, the team treated it as an operational fact that needed an operational answer, with the legal work built around protecting that answer rather than replacing it.

There was a third layer underneath both questions, one that only became visible once we started drafting: even a strong operational fix needed a legal structure that could not be quietly undone. A segregated server environment set up on day one of a combined business tends to drift over time, as shared vendors are consolidated, as IT budgets are trimmed, as the original reason for the separation fades from institutional memory two or three years after closing, right around when an audit lookback period is most likely to reach back into the acquired clinics. If the fix was purely operational, with no contractual teeth, there was a real risk it would erode before the exposure period ended. The legal question, in the end, was not only what the clause meant, but how to make sure the answer stayed true for as long as the risk remained live.

What we did

1. Traced the audit clause through every government contract in the portfolio. Biniam's chain held several separate government service agreements, not one, and each had been signed in a different year with slightly different boilerplate. We read all of them rather than assuming the language matched, because a fix built around one version would have missed exposure sitting in another, and a diligence memo that spoke to only the largest contract would have left the smaller ones as blind spots after closing.
2. Confirmed the practical fix before drafting anything. The twist in this file was that the real solution was not a clause at all. Working with Tigist's operations team, we confirmed that the clinics' billing systems could be kept on a segregated server environment, physically and logically separate from the buyer's shared portfolio infrastructure, at a modest ongoing cost that Tigist's backers were willing to absorb. That decision, made by the business, did more to contain the risk than any indemnity could have.
3. Drafted the legal work to protect the practical fix. Once segregation was the plan, our task was to make sure the purchase agreement and post-closing covenants locked it in rather than left it as an informal intention. We added a closing covenant requiring the segregated environment to be established before any transition of billing operations, with the requirement running for the length of the funder's typical audit lookback period.
4. Negotiated an indemnity scoped to the audit exposure specifically. Biniam's counsel resisted a broad indemnity but accepted one narrowly tied to costs arising from government audits of pre-closing conduct, including the cost of responding to any request the funder made for records outside the segregated environment. That gave Tigist a financial backstop even if the segregation plan was tested, without asking Biniam to underwrite risks the clinics' pre-closing conduct had nothing to do with.
5. Escrowed a portion of the purchase price against the audit period. Rather than relying solely on Biniam's post-closing solvency to make the indemnity meaningful, we negotiated a holdback, released to Biniam once the standard audit window on the largest contract had passed without a claim. The holdback gave Tigist a fund to draw against directly, rather than a promise she would later have to chase and collect on if a claim materialized.
6. Wrote a side letter addressed to the interpretation dispute itself. Anticipating that a funder might still assert the broad reading of the audit clause during a future review, we prepared internal guidance, reviewed with Tigist's compliance lead, setting out the narrow interpretation and the segregation facts the company could point to if an auditor ever pushed on scope. Having that position documented in advance meant no one would be improvising an answer under pressure during an actual audit.
7. Confirmed the financing lender was comfortable with the structure. Because the deal was partly debt-funded, we checked the segregation and indemnity terms against the lender's own conditions to closing, avoiding a late surprise from that side of the table. A lender uneasy about unresolved audit exposure on a major revenue contract could have slowed funding at the worst possible moment, so this step was cleared early rather than left for the final week.

The outcome

The transaction closed on schedule, in the middle of the fifty to eighty million dollar range the parties had discussed from the start. The segregated billing environment was live before the first day of combined operations, the indemnity and escrow terms were signed, and Goran's investment committee approved the deal without the audit clause becoming a last-minute reopener.

What made this a clear win was not a clever legal argument that neutralized the clause on paper. It was recognizing early that the paper argument would not have been enough on its own, and that the operational fix, done properly and locked in by the closing documents, did the actual work of protecting the buyer's broader systems. The legal spend on this file went less toward negotiating away the risk and more toward making sure the practical fix could not quietly slip after closing, once the deal team's attention moved to the next acquisition. A purely contractual answer, an indemnity alone with no segregation behind it, would have left Tigist arguing after the fact about what the audit clause was supposed to mean, with her own unrelated portfolio systems as the thing exposed while the argument played out.

Biniam accepted the escrow term as the price of a clean close, and the roughly two years of holdback gave both sides a defined end point rather than an open-ended worry. No audit had been triggered as of closing, and the segregation structure meant that if one came, it would have a clear boundary to run into. Tigist's team carried the same playbook, segregated environment first, indemnity second, into the next two acquisitions in the program, and it became a standard line item in the diligence checklist that every government-contract target was screened against before an offer went out.

The clinics themselves changed little in the transition. Staff kept their routines, billing continued on the same rhythm it always had, and the only visible sign of the fix was a technical one, a server boundary that patients and referral partners never saw. That quiet, unremarkable continuity was itself a measure of success. A structuring problem handled well rarely produces a dramatic story. It produces the absence of one.

What you can learn from this