Advisor Fee Estimates vs Actuals: Closing the Gap

The number in the engagement letter is a forecast, not a price. When a financial diligence advisor writes "£180k" against a workstream at kickoff, what they are quoting is a confident guess about hours, seniority mix, and a scope they have only half seen. By the time the invoice arrives, that guess has met reality, and the two are usually not the same number. Advisor fee estimates set expectations; the actuals settle them, and the deal lead is the one holding the difference.

That gap is structural, not a sign anyone behaved badly. Most diligence is billed on time-and-materials, where the client absorbs cost variation as scope evolves rather than the advisor. Estimates are built before the data room is fully open. And nobody re-prices the work as it drifts, so the variance accumulates quietly until it lands as a line on the final bill.

Why do advisor fee estimates miss the final invoice?#

Advisor fee estimates miss because they are priced against an unknown. At engagement, the advisor has read a teaser, maybe a management presentation, and the buyer's stated thesis. They have not yet found the three subsidiaries with no audited accounts, the revenue-recognition question that needs a second specialist, or the seller who answers data requests three days late. The estimate is an honest forecast of a scope that has not finished revealing itself.

The billing model then decides who pays for the surprise. On a time-and-materials engagement, the advisor invoices hours worked, so when scope expands the cost flows straight to the client. A fixed fee shifts that risk the other way, which is why providers asked to quote fixed build in a risk buffer of roughly 15% to 30% for the unknowns. Either the buyer pays for the overrun after the fact, or they pay a premium for certainty up front. Most mid-market diligence runs on T&M, so the buyer carries the variance.

There is decent academic ground for treating these estimates as systematically optimistic. In the largest study of its kind, Bent Flyvbjerg and colleagues at Oxford analysed 5,392 IT projects and found cost overruns follow a power-law distribution with a long fat tail of extreme blowouts, rather than the tidy normal distribution managers assume. Diligence is not IT, and I would not stretch the analogy further than the shape of the problem. But the lesson transfers: when work is complex and interdependent, "managers may be unwittingly exposing their organizations to extreme risk by severely underestimating the probability of large cost overruns." A single thread you have to pull, one messy jurisdiction, can pull the whole estimate apart.

What actually drives the variance?#

Three things drive most of the estimate-to-actual gap: time-and-materials overruns, in-scope work that simply took longer, and out-of-scope work nobody re-priced. They compound, and they are easy to confuse on an invoice that arrives as one number.

Time-and-materials overruns are the baseline. The engagement quotes a blended rate and an hours assumption; if the team books more hours, or a partner does work a manager was meant to do, the cost rises with no change to the agreed scope at all. In-scope overruns happen when the agreed work is harder than priced, the classic example being a quality-of-earnings analysis that has to unpick a year of manual adjustments before it can even start normalising EBITDA. Out-of-scope work is the real surprise: a new workstream, an extra jurisdiction, a vendor diligence the seller did not flag. This is where scope creep quietly inflates advisor fees, because the work expands before anyone issues a change request.

The chart below shows how the same engagement letter can land. Each workstream was budgeted at kickoff; each came in higher, for a different reason. The pattern, not the exact figures, is what recurs.

Notice that commercial came in almost on estimate, while financial and tax ran 30% to 50% over. That asymmetry is the point. Aggregate "advisor spend was 18% over budget" tells the deal lead nothing actionable. Variance by workstream tells them tax needs a conversation and commercial does not.

How big is the gap on a typical deal?#

On mid-market deals the gap commonly runs 20% to 40% above the original estimate, concentrated in the workstreams where investigation can deepen without a hard stop. That range is consistent with how the fee structures themselves are built and re-priced, and with the broader pattern that complex professional-services work overruns more often than it lands on budget.

Start from the fee levels. According to the Firmex M&A Fee Guide 2024-2025, produced with Axial, DealCircle and Divestopedia, sell-side success fees on mid-market deals decline from roughly 5.5% on a $5M transaction to about 2.1% on a $100M one, with monthly work fees of $5,000 to $10,000 on top. Those are sell-side advisory numbers rather than buy-side diligence, so treat them as a sense of scale, not a diligence benchmark. The same guide found advisors struggled to push fees up in a softer market, with only 34% of firms raising at least one fee type in 2024, down from 47% the year before. When headline rates cannot move, the variance shows up elsewhere: in expenses (76% of advisors now receive some reimbursement) and in scope.

The estimating literature suggests the optimism is the norm, not the exception. Beyond the Oxford power-law work, the broader pattern in complex engagements is that actuals routinely exceed forecasts once interdependencies bite. The honest read for a deal lead is that a diligence estimate is a planning anchor with a known upward bias, and the job is not to demand a perfect estimate but to watch the drift against it in real time.

Why don't spreadsheets close the gap?#

Spreadsheets record what an advisor cost last month; they cannot tell you what an advisor is about to cost. A fee tracker updated from invoices is a rear-view mirror. By the time a number lands in a cell, the work that drove it is already done and the cost is already committed, which is the worst possible moment to discover an overrun.

The deeper problem is that a spreadsheet has no concept of agreed scope. It holds an estimate and an actual, and it subtracts one from the other. It does not know that the actual includes two days of work the engagement letter never covered, or that a workstream is 60% billed against 30% of its deliverables. So the deal lead sees a total that is "a bit over" and has no way to tell whether that is in-scope effort running hot or out-of-scope work that should have triggered a change request and a conversation with the IC.

That blind spot is expensive in ways that outlast the deal. PwC's 2023 M&A Integration Survey found that 59% of companies spent 6% or more of deal value on integration, up from 38%, yet only 14% reported significant success across strategic, operational and financial measures. Costs that are invisible at the point of commitment do not just blow the diligence budget; they erode the return the whole deal was underwritten on. Closing the estimate-to-actual gap means tracking committed spend, not just invoiced spend, against scope you actually agreed, which is the core discipline behind managing due diligence costs across a deal.

How do deal teams close the estimate-to-actual gap?#

Deal teams close the gap by turning a single advisor estimate into a live, workstream-level budget they track against committed and actual spend, so drift surfaces as a decision rather than a discovery. The aim is not a more accurate estimate, which the unknowns make impossible, but a shorter distance between when scope expands and when someone prices it.

The point of the discipline is not to stop scope from expanding. Some expansion is the diligence working as intended, finding the thing you were paying to find. The point is that each expansion becomes a deliberate, priced choice, captured against the original estimate, rather than a number you reconcile in surprise at the end. The deal lead who can show their IC a clean estimate-versus-actual by workstream, with every variance traced to a logged decision, is not running a cheaper diligence. They are running one nobody has to apologise for.

The harder question is cultural, not technical. As long as the deal lead is the unpaid finance chaser who finds out what diligence cost when the advisor's accounts team sends the bill, the estimate-to-actual gap will keep being a year-end story instead of a live one. Pricing the drift while it happens is less a software feature than a decision about who gets to be surprised.

Frequently Asked Questions#

Why do advisor fee estimates differ from the final invoice?#

Advisor fee estimates are forecasts set before the full scope is visible, and most diligence is billed on time-and-materials, so the client absorbs cost variation as work expands. The estimate meets the data room, finds more than anyone priced, and the gap surfaces only at invoicing because no one re-prices the drift in between.

How much do diligence advisor fees typically run over estimate?#

On mid-market deals the overrun commonly runs 20% to 40% above the original estimate, concentrated in the financial and tax workstreams where investigation can deepen without a natural stopping point. Commercial and legal scopes tend to be more contained, so the variance is rarely spread evenly across the engagement.

Is a fixed fee better than time-and-materials for controlling advisor costs?#

A fixed fee gives budget certainty but the provider prices in a risk buffer of roughly 15% to 30% for the unknowns, so you pay a premium even when the work goes smoothly. Time-and-materials is cheaper when scope holds but pushes overrun risk onto the buyer, which is why live tracking against agreed scope matters more than the billing model itself.

How do you track advisor fees against estimate in real time?#

Break the engagement into workstream-level estimates, record committed and accrued spend rather than waiting for invoices, and log every scope expansion as an approved change request. Tools like AdviLink are designed to hold the agreed scope alongside live spend so variance is visible while it is still a decision, not after the bill lands.

Sources#

1. The Empirical Reality of IT Project Cost Overruns: Discovering A Power-Law Distribution. Bent Flyvbjerg, Alexander Budzier, Jong Seok Lee, Mark Keil, Daniel Lunn, Dirk W. Bester, 2022. Analysis of 5,392 IT projects showing overruns follow a fat-tailed power-law distribution, not a normal one.
2. M&A Fee Guide 2024-2025, Global Edition. Firmex, with Axial, DealCircle and Divestopedia, 2025. Mid-market success-fee ranges, monthly work fees, fee-increase rates (34% in 2024 vs 47% prior), and expense-reimbursement statistics.
3. 2023 M&A Integration Survey. PwC, 2023. Integration-spend levels (59% spent 6%+ of deal value) and success rates (14% significant success across all measures).
4. Time and Material vs Fixed Price Guide. Gain, 2026. Risk allocation between billing models and the 15–30% risk buffer providers add to fixed quotes.