Lenders size loans on trailing EBITDA: the last twelve months, adjusted for items the documents show will not recur. They add an acquired company's own trailing earnings when its statements are verified, and they sometimes credit a run-rate change that is already visible in results and backed by a contract. Cost savings not yet achieved are usually capped, time-limited or left out of sizing altogether, and forecasts are used to test headroom, not to size debt. Every adjustment you want credited needs paper behind it, and a file that separates proven adjustments from upside is read more generously than one that does not.
- What lenders size on
- Trailing twelve months, adjusted for documented one-time items
- Acquired company's earnings
- Added when its own statements are verified
- Run-rate changes
- Credited in part, when already in results and contracted
- Synergies and cost savings
- Capped, time-limited, or excluded from sizing
- Projections
- Test covenant headroom; rarely size the loan
Six versions of the same company's EBITDA
The same business can honestly produce several EBITDA figures, each answering a different question. Lenders are not hostile to any of them. They simply give each a different weight, based on how much of it has already turned into cash and how much of it the documents can prove.
| Version | What it measures | What proves it | How lenders usually treat it |
|---|---|---|---|
| Reported (trailing twelve months) | What the statements show for the last twelve months | P&L, balance sheet, tax returns | The starting point for every loan |
| Adjusted | Reported, plus add-backs for owner costs, one-time and non-cash items | Invoices, payroll records, settlement papers | Credited item by item, where documented |
| Run-rate | Recent results annualized, or a change not yet in a full year: a price increase, a new contract, a closed loss-making site | Signed contracts, months of results since the change | Credited in part, and only for changes already visible |
| Pro forma for an acquisition | Combined earnings as if an acquired company had been owned all year | The target's own trailing statements, ideally a quality of earnings report | Usually credited; it is historical, just combined |
| Pro forma with synergies | The above, plus cost savings expected from combining | Actions taken, contracts signed, a plan | Capped, limited to a set period, or excluded from sizing |
| Projected | Next year's forecast | A budget and its assumptions | Used to check headroom and repayment, not to size the loan |
The key terms are defined in LTM and TTM, EBITDA add-backs and pro forma EBITDA. What follows is how lenders move from one to the next.
Why lenders size to what has already happened
A loan is repaid from cash the business actually produces. Historical earnings are evidence that the cash exists; a forecast is a claim that it will. When a lender sizes a loan at 2x or 3.5x EBITDA, every unit of EBITDA that does not arrive leaves two to three and a half units of debt with no earnings behind them. Crediting a forecast that misses is how a lender ends up with a loan the company cannot service in its first year.
That is also why the question is not whether an adjustment is reasonable, but whether it is proven. An owner's salary above market is reasonable to add back, and a lender will credit it once payroll records show what was paid and the model shows what a replacement will cost. A new customer is reasonable to annualize, and a lender will credit it once a signed contract and several months of invoices show the revenue is real and recurring. The same adjustments, asserted without paper, are left out.
Lenders do not argue with your forecast. They just do not lend on it.
An adjustment ladder, worked through
A company presents EBITDA of 1,850 to support a refinancing and an add-on acquisition. Here is how a cautious lender might rebuild it, in plain numbers:
| Line | Borrower's figure | Lender credits | Why |
|---|---|---|---|
| Reported EBITDA, last twelve months | 800 | 800 | Ties to the statements and tax returns |
| Owner salary above a market replacement | 150 | 150 | Payroll records and a replacement cost in the model |
| One-time legal settlement | 50 | 50 | Settlement agreement and invoices |
| Owner's personal expenses in the P&L | 50 | 0 | No itemized support; the ledger mixes business and personal costs |
| New customer contract, annualized | 100 | 50 | Signed contract, but only a few months of invoices |
| Acquired company's trailing EBITDA | 300 | 300 | The target's latest full-year statements, verified |
| Cost savings from combining the two | 150 | 0 | Not yet achieved; left out of sizing |
| Forecast growth next year | 250 | 0 | A projection |
| Total | 1,850 | 1,350 |
At 3.5x EBITDA, the borrower's figure would support 6,475 of senior debt; the lender's supports 4,725. At a more conservative 2x, the gap narrows in absolute terms but not in proportion. A borrower who sized the transaction on 1,850 has a funding hole it has to fill with equity, seller paper or a smaller deal. One who expected 1,350 has no surprise. How much debt a business can carry shows how the leverage test and the coverage test combine once the figure is settled.
Caps and conditions on pro forma adjustments
Where lenders do credit pro forma items, particularly cost savings from an acquisition, the credit agreement usually fences them in. The common conditions are:
- A cap. Credited savings are limited to a share of EBITDA, so no amount of projected savings can make up more than a set portion of the figure.
- A time limit. The actions producing the savings must be taken, or expected to be realized, within a stated period after closing. Savings that never arrive drop out.
- Supportability. The savings must be identifiable and factually supportable, and an officer usually certifies them.
- No double counting. Once savings show up in actual results, the pro forma credit for them falls away.
There is a distinction borrowers often miss. The EBITDA used to size the loan and the EBITDA defined in the credit agreement to test covenants are not always the same. A lender may leave savings out of sizing and still allow them, capped, in the covenant definition, giving the company some room. The negotiated definition is covered in how EBITDA is defined in a credit agreement, and the room it creates in covenant headroom.
What counts as documentary support
| Adjustment | Evidence lenders accept | What weakens it |
|---|---|---|
| Owner compensation | Payroll records; the cost of a market replacement in the model | Family members on payroll who also do real work |
| One-time expense | Invoices, a settlement agreement, an insurance claim | The same kind of one-time cost appearing every year |
| Discontinued product or closed site | Lease termination, results by site or product line | Shared overhead that did not go away with it |
| Price increase | Customer notices, invoices at the new price, retention since | A few months of data against a history of discounting |
| New contract | The signed contract, invoices to date, the customer's credit quality | A contract terminable on short notice |
| Acquired company's earnings | Its own latest full-year statements and tax returns, a quality of earnings report | Seller-prepared figures that do not tie to returns |
| Cost savings | Headcount changes already made, signed vendor contracts | Savings that depend on future negotiations |
Two points apply to all of them. First, the figures have to reconcile: an add-back from a P&L that does not tie to the tax returns starts from a number the lender does not trust; see when the seller's statements do not match the returns. Second, for acquisitions, lenders size on the target's latest full year of figures, never an older year. A quality of earnings report is the most efficient way to settle a long list of adjustments at once.
When earnings are rising fast
Trailing twelve months is hardest on a company whose earnings are climbing. If the last quarter was far stronger than the same quarter a year earlier, the trailing figure understates what the business earns today, and a loan sized on it will be smaller than the company can carry. Lenders know this, and the better ones have ways to meet a growing company partway without lending on a forecast.
- Weight on recent quarters. Where revenue is contracted and recurring, some private credit lenders give weight to the most recent quarters rather than the full trailing year, provided the monthly figures show the step-up holding.
- Debt that grows with earnings. A delayed-draw term loan or an accordion lets the company borrow more later, once the higher earnings show up in trailing results, instead of sizing the whole loan today.
- A pricing grid. Pricing that steps down as leverage falls rewards growth after closing, so a company that expects earnings to rise can accept a starting leverage level and let the rate improve with results.
- A current set of figures. A year-to-date P&L through last month-end lets the lender see this year's results rather than last year's, which is often worth more than any argument about run-rate.
What none of them does is credit next year's budget. A company that needs the debt now, sized on earnings it has not yet produced, is asking for equity risk at a lender's price, and the answer is usually equity or growth capital structured for that risk.
Presenting adjustments so they survive credit committee
The strongest files show the bridge from reported to adjusted EBITDA line by line, with the document behind each line named, and they separate what they expect a lender to credit from what they consider upside. That separation costs the borrower nothing. Credit committees trust a sponsor or owner who has already taken out the weak adjustments, and they look harder at every line of a file that has not.
Midas Partners's financing model carries that bridge, with the downside case beside it, and the underwriting memo explains each adjustment and its support; how we underwrite sets out the approach. Software does the analyst work of building the bridge from the ledger, and a senior banker checks every line before the client approves it. Once the documents are in, the full lender package is built in a day, where by hand the same package takes at least a week. For acquisitions, the add-on case is covered in add-on acquisition financing, and the second test a lender runs, on value rather than earnings, is in how lenders value a business.
Common questions
- Will a lender lend on projected EBITDA?
- Rarely for sizing. Lenders use projections to check that the company can repay and stay inside its covenants, but the loan amount is set on trailing results plus documented adjustments.
- What is the difference between run-rate and pro forma EBITDA?
- Run-rate EBITDA annualizes a recent change in the same business, such as a new contract or a price increase. Pro forma EBITDA restates earnings as if an acquisition or other transaction had happened at the start of the period, sometimes with expected cost savings added.
- Do lenders credit acquisition synergies?
- Some do, with a cap on how much of EBITDA they can make up, a time limit for achieving them and a requirement that they be supportable. Many lenders leave them out of sizing entirely and allow them only in the covenant definition.
- Can I annualize a strong recent quarter?
- Only if you can show the change is permanent, such as a signed contract or a price increase customers have accepted. A strong quarter in a seasonal business is not a run-rate, and lenders look at the same months in prior years.
- Will a lender credit a price increase I just put through?
- In part, if customers have accepted it. Lenders want invoices at the new price and evidence that customers stayed, usually over several months. A price increase announced but not yet billed is a forecast, and is treated as one.