The earnout does not break your covenant. It takes ninety percent of your cash, in the year you do well.

Nothing is inherited yet. A target has come across the desk and the terms are being argued. The earnout threshold—the target's gross profit, $2,800,000.00—is one of the most negotiated elements of a GovCon acquisition and one of the least rigorously modeled.

It goes under-modeled because the obligation is not determinable. It is performance-based and it is multi-variable: what the target does, what your own business does in the same year, and when in that year the performance lands. The target can clear the threshold on six good months, fall off in the next six, and the payment is still owed. The target can perform while you do not, or you can perform while the target does not—different obligations, against a different balance sheet, in different years.

None of that is arithmetic on a term sheet. It is a sensitivity problem—which is why it gets settled on instinct and discovered later. This page runs it as one.

The usual approach starts from a model somebody builds by hand, for this deal, from the ground up. The baseline this analysis runs on is last night's close—the P&L that synced, the indirect rates computed this morning, the contract waterfall reflecting actual backlog.

Which means the acquirer's side of this deal needs no new inputs at all. EBITDA comes out of the trial balance by adding back the two lines already sitting on it.

The baseline at stop 02 came entirely from the ledger. Everything the analysis still needs is a term somebody is negotiating. The target does $8,400,000.00 of revenue on $2,700,000.00 of gross profit and $1,050,000.00 of EBITDA; the price is five times EBITDA, or $5,250,000.00; seventy percent is financed with senior debt. Those five are the only numbers on this page that did not already exist, and sources and uses follow from them without another assumption.

The price is struck on EBITDA and the earnout is measured on gross profit, and that is deliberate. Gross profit is the line the seller's team still controls after close. EBITDA below it is a function of your indirect structure—your overhead, your G&A, your allocation base—which the seller does not run and should not be paid on. Arcvue also carries the definition, because in GovCon it is the whole argument: gross profit as booked, net of subcontractor pass-through, or net of ODCs as well. A target can clear a gross-profit threshold on pass-through volume that earns nothing, and the definition is what stops it.

With the terms from stop 03 in place, the threshold can finally be priced. The earnout runs as a linear ramp—nothing at the threshold, the full $2,000,000.00 cap four hundred thousand dollars above it, which is five times each dollar of gross profit in between. At the negotiated threshold of $2,800,000.00, in the upside case where the target delivers $3,000,000.00, it pays $1,000,000.00.

In the base case it pays nothing, because base case is the threshold. In the downside it pays nothing. That is usually where the conversation stops—one number, three cases, and a decision.

Stop 04 priced one cell. But the threshold is the thing under negotiation, so the honest question is not what does it pay at $2,800,000.00—it is what does it pay at every threshold the seller might get you to, under every performance level that is plausible.

The matrix runs the negotiated threshold plus or minus four steps—nine rows—against downside, base, and upside performance. Twenty-seven outcomes. And the extremes are not symmetric: at the lowest threshold, in the upside case, the cap binds and the earnout pays its maximum of $2,000,000.00—twice what the negotiated cell pays.

The twenty-seven outcomes from stop 05 each imply a post-close debt position, and each is tested against the actual covenant thresholds in the actual credit agreement—not a generic benchmark. This is the test everybody runs, and on this deal it is uneventful.

Including the combination nobody models: the target performs and you do not. Leverage is tested on the combined entity, so your own year is an input to it, not a constant. Run the acquirer fifteen percent below plan against the corner cell—the earnout still pays in full, on the target's performance, into a smaller business—and leverage reaches 1.92×. It still passes.

Leverage never gets near the covenant, so a model that stops at covenant compliance reports a clean deal. The obligation is still real, and it is settled in cash.

The company holds $2,221,300.00 across its operating and payroll accounts. The corner cell pays $2,000,000.00. That is 90.0% of the cash the business has, leaving $221,300.00—and it competes directly with everything else that money was for: bid and proposal investment, hiring ahead of an award, working capital, paying down the line.

And it is rarely owed in the year that earned it. Measurement can be annual, cumulative or quarterly, and payment can land in the same year, the following year, or six months out—so a target that clears the threshold on a strong first half is paid out of a later year's cash, whatever that year turns out to be. Arcvue carries the measurement period and the payment timing as terms and settles the payment in the year it actually falls, because that is the year the money has to be there.

Look at where the corner cell from stop 07 sits. It is not the downside. Downside pays nothing—the target misses the threshold and no earnout is owed. The $2,000,000.00 is the upside cell, at the lowest threshold on the matrix.

Which inverts how this risk is usually carried in someone's head. Deal risk is modeled as what if it goes badly, and an earnout is the one term where the exposure runs the other way: the better the business performs, the more cash you owe—and you owe it exactly when you would rather be reinvesting.

The threshold began as a number across a table that committed you to nothing. It is now the only term in the deal whose blast radius has been measured—and the measurement says the covenant is not the exposure and the cash is.

That changes what there is to negotiate. Cap the earnout lower and the corner disappears. Pay it in installments and it stops colliding with a single year's liquidity. Fund it from the revolver rather than cash and it becomes a leverage question—where, per stop 06, there is 1.79× of room. None of those are concessions on price. They are structure, and structure is what you can trade when you know precisely which cell hurts.

And the earnout was one term of one deal. The same baseline carries the debt itself. Every instrument you hold is scheduled month by month—principal and interest, at the rate you supplied and the date it took effect—and that interest lands in the forecast rather than beside it. Retire an instrument at close and the schedule is replaced, not annotated. Covenant headroom is then carried forward against the agreement's own thresholds, so a test that will fail four quarters out is visible four quarters out.

A refinance is the same balance sheet asked a different question, and it is answered the same afternoon.