"EBITDA was down $119K on prior year." That sentence appears in a great many board packs and it contains almost no information.
Down because revenue fell? Because margin compressed? Because costs grew? Because all three moved and partially offset each other? Management cannot act on the headline because the headline does not identify the mechanism.
An EBITDA bridge answers the question properly. It decomposes the movement into named drivers that sum exactly to the total change — and that exactness is what makes it defensible.
For most operating businesses, EBITDA movement resolves into three components:
Separating the first two matters more than it might appear. Revenue and margin are frequently discussed together, but they are different management problems. A revenue decline is a demand or retention issue. A margin decline at flat revenue is a pricing or cost-of-delivery issue. Collapsing them into "gross profit was down" loses the distinction.
Take a simple case. Prior year revenue $83.5K at 85.5 percent gross margin, operating expenses $18.3K. Current year revenue $89.3K at 85.5 percent gross margin, operating expenses $18.5K.
| Component | Calculation | Effect |
|---|---|---|
| Prior year EBITDA | ($83.5K × 85.5%) − $18.3K | $53.1K |
| Revenue contribution | +$5.8K revenue × 85.5% prior margin rate | +$5.0K |
| Margin effect | 0.0pt rate change × $89.3K current revenue | −$0.0K |
| Operating expense movement | −($18.5K − $18.3K) | −$0.2K |
| Current year EBITDA | Sum of the above | $57.9K |
The bridge ties. Every component is derived from figures that exist in the accounts, and the total reconciles without a plug.
The rule: a bridge with a residual line labelled "other" or "unexplained" is not finished. The residual is where the actual insight usually lives.
Four causes account for most residuals.
The revenue contribution uses the prior period margin rate. The margin effect uses the current period revenue. Reversing either produces a cross-term that lands in the residual. The convention matters less than applying it consistently, but mixing them guarantees a gap.
For a business with foreign currency operations, translation movement will contaminate every component unless it is broken out as its own bucket. A revenue bridge showing new, expansion, contraction, and churn but no FX line is either a single-currency business or an incomplete bridge.
A legal settlement inside the operating expense movement makes the cost line look structurally worse than it is. One-time items should sit in their own component so management can see underlying trend separately from noise.
If revenue is shown gross of discounts in one component and net in another, the bridge will not close. This is more common than it should be when figures are pulled from different reports rather than a single reconciled ledger.
Once the top-level bridge ties, each component can be decomposed further. This is where a bridge becomes genuinely actionable rather than merely correct.
The revenue contribution breaks into new logos, expansion, contraction, churn, and FX. A business showing $5.8K of net revenue growth might be composed of $11.5K in new logos plus $12.7K expansion against $3.8K contraction and $14.8K churn. Net positive, structurally alarming.
The margin effect breaks into rate and mix. Margin can fall while every individual product's margin holds steady, simply because the sales mix shifted toward lower-margin lines. That is a commercial problem, not a cost problem, and the two demand different responses.
The operating expense movement breaks by cost category, ranked by contribution to the change. Listing the top four movers by chart of accounts group is usually enough to identify whether growth was deliberate investment or drift.
The bridge explains the past. The obvious follow-up question is whether the movement persists.
If the revenue contribution was churn-led, the relevant question is how much of the forward period is already contracted. A business with 93 percent of next quarter under deferred revenue schedules has time to respond. A business with a fully modelled forward book does not.
If the margin effect was mix-driven, the question is whether the mix shift is structural or a timing artefact of a large delivery in one period.
A bridge that stops at explaining the past is analysis. A bridge that connects to committed forward positions is management information.
The requirement that a bridge reconcile exactly is not accounting pedantry. It is the property that makes the bridge usable when someone challenges it.
In a board meeting, a director will ask about one component. In due diligence, a buyer's accountant will select a component and trace it. If the bridge contains an unexplained residual, the entire decomposition becomes suspect — because if one component is a plug, the reader has no way to know which of the others were derived and which were fitted.
A bridge that ties by construction has a different standing. Each component was calculated from the accounts, the total reconciles, and any single line can be traced back to the entries that produced it.
An EBITDA bridge is a decomposition that explains the change in EBITDA between two periods by attributing it to named drivers — typically revenue contribution, gross margin effect, and operating expense movement. A correctly built bridge sums exactly to the total change with no residual, which is what makes it usable in board reporting and due diligence.
Start with prior period EBITDA. Add the margin contribution of the revenue change, holding the prior period gross margin rate constant. Add the effect of the gross margin rate change, applied to current period revenue. Subtract the change in operating expenses. The result must equal current period EBITDA. If it does not, a component is missing or a figure is being double counted.
The most common causes are: mixing gross and net figures between components; failing to isolate foreign exchange translation as its own bucket; applying the margin rate change to the wrong revenue base; and including one-time items in the operating expense line without separating them. Working through each component against the source accounts usually locates the residual quickly.
Variance analysis compares actual to budget or prior period across many lines and reports the differences. An EBITDA bridge is narrower and stricter: it explains one number's movement through a small set of drivers that must reconcile exactly. Variance analysis tells you which lines moved. A bridge tells you how those movements combined to produce the change in profitability.
Fynease builds EBITDA, revenue, and cash bridges from your reconciled close data. Every component reconciles to zero variance by construction. If it does not tie, it does not publish.
See Fynease Intelligence Start free trial