How to Build a Debt Covenant Tracker (and What Actually Happens If You Trip One)

Most covenant breaches aren't a surprise to the lender. They're a surprise to the borrower.

The lender has been watching the trend in your compliance certificates for three quarters. You've been watching revenue. Then the quarter closes, someone finally runs the calculation, and the number is 3.9x against a 3.75x limit. Now you're calling your relationship manager with no plan, no forecast, and 30 days to file a certificate that says you failed.

That call goes very differently if you make it six months earlier.

The math on how little room you have

As of March 31, 2026, Kayne Anderson BDC reported that its private middle market portfolio companies carried weighted average leverage of 4.4x and an interest coverage ratio of 2.4x, at a loan-to-enterprise value of about 43% (SEC filing). Those are healthy borrowers, not the watch list.

Now do the arithmetic. If your interest coverage is 2.4x and your covenant floor is 2.0x, EBITDA can fall about 17% before you fail the test. Not 17% over a couple of years. 17% on a trailing twelve month basis, which means one soft quarter plus one mediocre quarter gets you close.

That's the whole reason this matters at $5M–$50M in revenue. A single lost customer or a margin slip on a big contract can move the number enough to matter, and the trailing twelve month window means the damage sticks around for a full year after the quarter that caused it.

What a maintenance covenant actually tests

A maintenance covenant is a financial test you have to pass on a schedule, usually every quarter, for as long as the loan is outstanding. That's different from an incurrence covenant, which only gets tested when you do something specific like take on more debt or pay a dividend.

The common ones in a lower middle market credit agreement:

• Total leverage ratio. Funded debt divided by covenant EBITDA, with a ceiling that usually steps down over the life of the facility.
• Fixed charge coverage or interest coverage. EBITDA (sometimes less capex and cash taxes) divided by what you owe in interest and scheduled principal, with a floor.
• Minimum liquidity or maximum capex. Less universal, but common in sponsor deals and asset-heavy businesses.

If you're on an ABL revolver, your fixed charge coverage covenant may be springing, meaning it only gets tested when excess availability drops below a threshold. That sounds like a break. It isn't. It just means the test shows up on the exact day your borrowing base is already tight.

The compliance certificate is typically due 30 to 45 days after quarter close, signed by an officer. If your close takes three weeks, you're computing a legally binding ratio in the last few days of that window. That's how errors happen.

The number in your credit agreement is not the number in your P&L

This is where most companies get burned, and it has nothing to do with performance.

Covenant EBITDA is a defined term. It lives in the definitions section of your credit agreement, it runs a page or two, and it does not equal your GAAP EBITDA or the adjusted EBITDA from your last quality of earnings report. The add-backs you're allowed are enumerated, and several of them are capped, often at some percentage of EBITDA in total. Restructuring costs might be permitted up to a limit. Pro forma cost savings from an acquisition might be allowed for four quarters and then expire. That expiry alone has pushed plenty of borrowers into breach in a quarter where the business did fine.

Same with debt. "Funded indebtedness" in the leverage definition is often narrower than everything sitting in the liabilities section. Some agreements net cash against debt, up to a cap. Some don't net at all. If you're using your balance sheet total, you may be reporting a worse ratio than you actually have.

So pull the definitions and rebuild the calculation line by line, once, against the actual document. Not the term sheet, not the summary your banker sent. The executed agreement. It's a painful afternoon and then you have a template that lasts the life of the facility.

Building the tracker

One tab. Four columns per covenant, and it doesn't need to be fancier than that.

The test. Name it, cite the section, and note the exact schedule of levels including step-downs. A 4.0x ceiling that steps to 3.5x in Q2 next year is two different problems.

The build. Every input, tied to a source. Covenant EBITDA computed from your TTM P&L with each add-back on its own line, showing the cap and what you've used against it. Funded debt from the definition, not the balance sheet subtotal.

The result. The ratio, versus the level, for the current quarter.

The headroom. This is the column people skip and it's the one that gets read. Don't just show 3.6x against 4.0x. Show how far EBITDA can fall before you fail: about 10% in that example. Percentages are what a CEO and a board can actually react to. Ratios are what a lender reads.

Then roll it forward. Run every covenant against your forecast for the next four to six quarters, not just against last quarter's actuals. A covenant tracker built on actuals tells you that you already failed. One built on the rolling forecast tells you that Q1 gets tight if the new hires land in November, which is a decision you can still make differently.

How much headroom is enough? Below 15% you should be talking about it monthly and building the story. Below 10% you should have already spoken to your lender.

What actually happens when you breach

Nobody shows up at your office. A financial covenant miss is a technical default, and technical defaults are a negotiation, not a repossession.

Your lender's realistic menu: waive the test for the quarter, amend the agreement to reset the levels going forward, suspend the covenant for a period, or let the sponsor put in an equity cure. Most of the time it comes with a price. An amendment fee, a spread bump of maybe 50 to 100 basis points, tighter reporting like monthly financials and a 13-week cash flow, sometimes a tighter capex or distribution limit.

Equity cures are worth understanding before you need one. If you're sponsor-backed, the credit agreement usually lets the sponsor inject equity that counts as EBITDA for the covenant calculation. But cures are limited by design: often no more than two in any four consecutive quarters and a handful over the life of the facility, and frequently capped at the amount needed to just clear the test. It's a real tool. It is not an unlimited one, and burning one early costs you the option later.

The variable that changes the outcome most is timing. A borrower who calls in month two of the quarter with a forecast, a diagnosis, and a plan gets a waiver and a fee. A borrower whose lender finds out from the certificate gets a full repricing and monthly reporting for a year, because the lender has now learned something about the borrower that has nothing to do with EBITDA.

Run it monthly

Quarterly testing does not mean quarterly monitoring. Recalculate every covenant during your monthly close, on trailing twelve months, and put the headroom number in the management report next to the budget variance. It takes about twenty minutes once the template exists.

The point isn't compliance paperwork. It's that debt capacity is a real constraint on your operating plan, and if you can only see it four times a year you'll keep making hiring and capex decisions that quietly spend headroom you didn't know you were spending.

We build and run these trackers as part of outsourced FP&A, alongside the monthly close and the rolling forecast, so the covenant math updates itself every month instead of getting rebuilt in a panic the week the certificate is due.