Diligence
Normalised EBITDA
Normalised EBITDA is your statutory earnings adjusted for things that will not recur. It is the number your facility is sized on, and every adjustment in it is tested on its own.
Also called adjusted EBITDA · EBITDA add-backs · underlying EBITDA · run-rate EBITDA · pro forma EBITDA
The bridge is where the facility is really sized. Statutory earnings rarely equal the number a lender lends against.
| Step | Running EBITDA |
|---|---|
| Statutory EBITDA | £1.6m |
| Owner remuneration | £1.85m |
| One-off costs | £2.03m |
| Pro forma acquisition | £2.15m |
| Less struck by diligence | £2m |
Illustrative bridge to the £2m post-diligence EBITDA example used across these pages. Derived arithmetic.
Why the adjusted number exists
Statutory EBITDA describes what happened. A lender is trying to work out what will keep happening, because that is what services the debt. Where the two differ, the adjusted figure is an attempt to describe the sustainable earnings of the business rather than the accidents of one reporting period.
That is a legitimate exercise and lenders expect it. The difficulty is that it is also the number the facility is sized on, so the borrower has an obvious incentive to make it large, and the lender knows it. What follows is not a debate about accounting principle but a line-by-line test of whether each adjustment describes something that will not recur.
The bridge, and why it belongs in the pack
The bridge is the schedule reconciling statutory EBITDA to the adjusted figure, one line at a time, with the evidence for each. It is the single most-read page of an information memorandum and the one borrowers most often supply last.
Supplying it upfront changes the dynamic. A credit team given a clear bridge tests the adjustments. A credit team given an adjusted number and asked to trust it starts by assuming the gap is optimistic and works backwards, which is a worse starting position and takes longer.
A typical bridge at £3-15m runs from a statutory figure to an adjusted one through owner remuneration, genuine one-offs, and pro forma treatment of anything acquired mid-period, with diligence removing whatever cannot be stood behind.
Add-backs are not equal. Some are near-automatic, some are argued, and some are struck almost every time.
| Adjustment | How it fares |
|---|---|
| Unevidenced owner adjustments | Struck |
| Recurring 'one-offs' | Struck |
| Run-rate synergies | Capped |
| Pro forma acquisition | With support |
| Owner remuneration rebase | Usually accepted |
Market convention at £3-15m. Every adjustment is tested on its own merits.
Owner remuneration
The most common adjustment in an owner-managed business, and usually the most defensible. Where an owner has been taking remuneration well above or below a market rate for the role, rebasing to what a hired manager would cost describes the business as it will run after the transaction.
It survives when it is evidenced: a market salary benchmark for the role, a clear statement of what the owner will do post-transaction and on what terms, and consistency with the management structure described elsewhere in the pack. It fails when the rebased figure assumes the owner keeps working full time for a nominal salary, or when the adjustment quietly removes the cost of a role that still needs filling.
Related adjustments, such as personal expenses run through the business or above-market rent paid to a property the owner holds, follow the same logic and the same evidence test.
One-offs that are, and one-offs that are not
A genuine one-off is a cost that happened once and will not happen again: a specific piece of litigation now concluded, a site relocation, professional fees on an aborted transaction.
The category that fails is the recurring one-off. A business that adds back restructuring costs in three consecutive years is not describing an exception, it is describing how it operates. Diligence looks across several years precisely for that pattern, and finding it damages the credibility of the adjustments that were legitimate.
The practical test before you put an item in the bridge is whether you would be comfortable if the lender asked to see the equivalent line in each of the last three years. If the answer is no, the adjustment is doing more harm than good.
Pro forma and run-rate adjustments
Two forward-looking categories, treated differently.
Pro forma treatment of a completed acquisition annualises the earnings of a business you already own but have only held for part of the period. It is generally accepted where the acquired accounts support it, because it describes something that has happened rather than something planned.
Run-rate synergies describe cost savings expected but not yet delivered. They are accepted more sparingly, usually capped as a percentage of EBITDA and time-limited to savings that will be realised within a defined window, with evidence of the actions already taken. An unactioned saving is a forecast, not an adjustment, and lenders treat it as such.
Where a facility funds an acquisition, how the definition annualises acquired earnings matters twice: once for sizing, and again for the covenant, because a bolt-on funded mid-year adds debt before it adds twelve months of earnings.
Every pound struck out costs three. At 3x leverage, £200k of disallowed add-backs is £600k of facility.
| Leverage | Facility lost |
|---|---|
| At 2.5x | £500k |
| At 3.0x | £600k |
| At 3.5x | £700k |
Worked example at 3x leverage, extended across the bank senior 2.5-3.5x band. Derived arithmetic.
What a rejected add-back costs
This is the arithmetic that makes the bridge worth the effort. Facility size is broadly EBITDA multiplied by the leverage the lender will underwrite, so every pound removed from EBITDA removes several pounds of borrowing capacity.
At 3.0 times, £200,000 of disallowed add-backs is £600,000 of facility. At the top of the bank senior band it is £700,000, and at the bottom £500,000. On a raise where the funding requirement is fixed, that gap has to be found from equity, from a deferred element, or by shrinking the transaction.
It also runs the other way. An adjustment that is properly evidenced and accepted is worth three times its face value in capacity, which is why the time spent assembling support for the bridge has a better return than almost any other preparation task.
The evidence standard
The working test is whether a diligence accountant would put their name to it. That is a higher bar than whether it is arguable, and a lower one than whether it is certain.
In practice it means a document rather than an explanation: an invoice, a contract, a benchmark, a board minute, a signed agreement. Adjustments supported only by management assertion are struck, not because anyone doubts the management, but because a credit paper cannot rest on an assertion the lender cannot verify.
Where the raise is large enough to warrant it, a vendor-commissioned quality of earnings report does this work in advance and in the lender's own language. It costs money and it removes a whole category of argument from the process.
What the pack has to survive
The figure gets normalised either way. The only question is whether it happens on your evidence or on a diligence accountant's, and a bridge built in advance with every line evidenced settles that before it is contested.
Dropping weak adjustments is counterintuitive and correct. A bridge with six well-evidenced lines is worth more than one with twelve where four are struck, because the strikes cast doubt on the eight that survived. Credibility across the whole pack is worth more than the marginal capacity from an adjustment that was always going to fail.
Then hold the definition. The adjustments accepted at close are the ones that roll forward into the covenant EBITDA for the life of the facility, so the bridge is not only about how much you borrow, it is about which number you are tested on every quarter afterwards.
Common questions
What is normalised EBITDA?
Statutory EBITDA adjusted for items that will not recur, so the figure describes the sustainable earnings of the business rather than the accidents of one period. It is the number a lender sizes the facility on, and every adjustment in it is tested on its own merits.
What add-backs will a lender accept?
Rebasing owner remuneration to a market rate and removing non-recurring costs are usually accepted where evidenced. Pro forma earnings from a completed acquisition are accepted with supporting accounts. Run-rate synergies are accepted sparingly and usually capped. Costs that recur each year in a different form, and adjustments supported only by management assertion, are usually struck.
How much does a rejected add-back cost me?
Roughly the multiple. At 3.0 times leverage, £200,000 of disallowed adjustments removes about £600,000 of facility. At the top of the bank senior band it is nearer £700,000. On a raise with a fixed funding requirement, that gap has to come from equity, from deferred consideration, or from shrinking the deal.
What is the difference between adjusted and pro forma EBITDA?
Adjusted EBITDA removes non-recurring items from the period as it happened. Pro forma EBITDA annualises something structural, most often a business acquired part-way through the period, so the earnings match the debt that funded them. Lenders generally accept pro forma treatment of a completed acquisition where the acquired accounts support it.
Why did diligence strike out my add-backs?
Usually because they could not be evidenced to the standard a diligence accountant would sign, or because the same category appeared in several consecutive years, which makes it a cost of doing business rather than an exception. A pattern of recurring one-offs also damages the credibility of the adjustments that were legitimate.
Should I include every adjustment I can argue for?
No. A bridge with six well-evidenced lines is worth more than one with twelve where four are struck, because the strikes cast doubt on the survivors. Drop the adjustments you cannot support before the pack goes out; credibility across the whole document is worth more than the marginal capacity from one that was always going to fail.
Do I need a quality of earnings report?
Not always, but where the raise is large enough to justify it, a vendor-commissioned report does the normalisation work in advance and in the lender's own language. It removes a category of argument from the process and shortens diligence, at the cost of the fee.
Does the normalised figure matter after completion?
Yes, and this is often missed. The adjustments accepted at close usually define the EBITDA used in the covenant for the life of the facility, rolled forward on a trailing twelve-month basis. So the bridge decides both how much you can borrow and which number you are tested against every quarter afterwards.
The full treatment sits in the guide: what lenders look for in your accounts.
This page explains a term as it is used in the UK lower-mid-market. It is general information, not advice on any particular facility. Terms vary between lenders and between deals, and the drafting in your own agreement governs.