Key takeaways
- 64% of LMM QoE engagements found material EBITDA differences from the seller's number, with a median gap of $180K, the EBITDA bridge is the highest-stakes document in your model.
- Sellers with documented revenue categorization (recurring vs. transactional) and customer-level detail received LOIs with 15–20% less valuation variance than sellers providing summary P&Ls.
- Every forward projection must be traceable to a specific operating driver, projections that can't be explained verbally don't survive a 5-hour diligence session.
- A 36-month forecast accuracy record is one of the most credible management capability signals in a PE process; build monthly variance documentation starting 18 months before launch.
How to use this before a process
For adjacent context, compare this with How to Prepare a Business for Sale: Why Transaction Readiness Starts Before the Process and Transaction readiness checklist for founder-owned businesses; the strongest operators connect these topics instead of treating them as separate workstreams.
36 months
Minimum historical period buyers want to see
3-5 years
Typical forward projection period
6x-8x
EBITDA multiple range where model precision matters most
$600K
Enterprise value impact of a 1x multiple on $100K EBITDA error
Readiness Snapshot
What buyers will ask
Which terms change economics after the headline price is agreed?; What conditions let the buyer delay, retrade, or walk away?; Which obligations survive close and how are they capped?
What to prepare
Marked LOI or purchase agreement term tracker.; Economic impact summary for escrows, holdbacks, notes, and indemnities.; Approval, covenant, and closing-condition checklist.
A financial model that cannot explain its own assumptions is a liability in an M&A process, not an asset. Buyers will build their own model regardless of what the seller provides, but the quality of the seller's model signals management sophistication, controls the opening anchor on valuation, and determines how much buyer diligence is required to get comfortable with forward projections.
The seller's financial model and the buyer's model serve fundamentally different purposes, and understanding that difference is one of the most underused preparation advantages available to founders. PE buyers underwrite at an entry multiple and target a 2.5–3x MOIC over a 4–6 year hold. Their model starts from the exit multiple and works backward to determine the maximum entry multiple that still achieves their IRR hurdle. When a seller understands how the buyer's model constrains their offer, they can structure their pitch to address those constraints directly, presenting not just EBITDA but the specific forward drivers that support the buyer's exit multiple assumption.
PE firms typically underwrite LMM deals at a target IRR of 20–25% and a 2.5–3x MOIC. On a $20M acquisition at 6x EBITDA ($3.3M EBITDA), a 5-year hold, and a 7x exit multiple, achieving 2.5x MOIC requires growing EBITDA to roughly $4.5M, a 36% increase. When the seller's financial model shows a credible path to $4.5M EBITDA with documented assumptions, it directly addresses the buyer's underwriting constraint. When it shows $5.5M with vague assumptions, buyers discount the entire projection. Sellers who understand the buyer's model can target their narrative at the right number.
Seller model vs. buyer model: the critical differences
Most sellers build their financial model as a management planning tool, top-down revenue assumptions, an operating budget, and a forward P&L. That structure is useful for managing the business. It is the wrong structure for an M&A process, because it addresses the questions the seller is asking rather than the questions the buyer is asking.
A PE buyer's model starts from a different set of questions: What is the normalized EBITDA at close? What is the quality of that EBITDA (recurring vs. transactional)? What are the realistic growth drivers over the hold period? What multiple can the business command at exit? And does the combined return math work at the proposed purchase price? Sellers who understand this framework can build a model that answers those questions rather than creating more of them.
In 2024, 64% of lower-middle-market QoE engagements identified at least one material difference between the seller's adjusted EBITDA and the QoE-adjusted EBITDA, with a median difference of $180K.
Sellers whose financial models included documented revenue categorization (recurring vs. transactional) and customer-level revenue detail received LOIs with 15 to 20% less variance in valuation ranges compared to sellers who provided summary P&Ls only.
The most common model deficiency identified by PE diligence teams in 2024 was insufficient documentation of growth assumptions, specifically, the absence of a bottoms-up revenue build showing how projected revenue growth would be achieved by customer segment, channel, or product category.
The model build sequence
Building the M&A financial model is a 30 to 60 day exercise when done properly. The most common mistake is starting with the projections rather than the historical data, buyers evaluate projections through the lens of historical performance, and a forward model not anchored to a clean historical baseline will not be credible regardless of how detailed the projection assumptions are.
Financial Model Build Sequence
Step 1: Assemble and clean 36 months of historical data
Pull monthly P&Ls, balance sheets, and cash flows. Normalize each period for accounting policy changes, one-time items, and presentation inconsistencies. This is the foundation for everything downstream.
Step 2: Build the EBITDA bridge
Construct the bridge from GAAP EBITDA to adjusted EBITDA, addback by addback, with documentation for each item. This is the highest-stakes document in the entire model.
Step 3: Categorize revenue
Separate revenue into recurring, transactional, and project categories at the customer or contract level. Build a retention and renewal analysis for the recurring revenue base.
Step 4: Build cost structure analysis
Separate fixed from variable costs. Calculate the contribution margin and the breakeven point. Show the operating leverage profile at different revenue levels.
Step 5: Build forward projections bottoms-up
By revenue category, document the specific assumptions driving each growth driver. Customer additions, retention rates, pricing changes, new product or service revenue. Every material assumption needs a supporting rationale.
Step 6: Stress test and sensitivity analysis
Identify the two or three key assumptions the projection is most sensitive to. Build a sensitivity table showing EBITDA at different outcomes. Buyers will do this, present it first.
The model build sequence starts with historical data, not projections. Buyers evaluate the projections in the context of historical performance. A three-year forward model that is not anchored to a credible, documented historical baseline will not be taken seriously regardless of the projection assumption quality.
$180K
Median QoE EBITDA adjustment in LMM transactions 2024
15-20%
LOI valuation range compression for sellers with revenue documentation
64%
QoE engagements finding material EBITDA differences
36 months
Historical period that most credibly supports forward projections
AI diligence angle
Run a short scan to identify reporting, data room, and workflow gaps that could affect diligence confidence.
Run an AI readiness scan →The five assumptions PE buyers always stress-test
Before running their own model, experienced PE buyers have a short list of assumptions they will always challenge regardless of what the seller presents. Knowing this list and building your model to address each assumption in advance converts a series of buyer challenges into confirmations of management credibility.
The Five Assumptions PE Buyers Always Stress-Test
1. Revenue growth rate
What buyers test: is the projected growth rate higher than trailing performance? What is the bottoms-up evidence base? Sellers who project above trailing growth should provide a pipeline analysis, signed contracts, or a new product/service revenue bridge with documented assumptions, not narrative.
2. EBITDA margin sustainability
What buyers test: is the current margin supported by cost discipline or by temporary cost suppression? Have any significant costs been deferred? Will owner-related addbacks hold under PE governance? Any margin near or above industry benchmarks requires documentation.
3. Customer concentration and retention
What buyers test: what happens to the revenue model if the top one or two customers reduce spend? Is the retention rate documented? For recurring revenue, what is the actual net retention rate over the last 12 and 24 months? Buyers haircut projections when concentration risk is underdocumented.
4. Capex and maintenance requirements
What buyers test: is the capex plan adequate to sustain current EBITDA, or has maintenance capex been deferred? A business reporting $500K in maintenance capex on aging equipment worth $5M will face a normalization adjustment. Buyers will compare capex to depreciation; if capex is below depreciation by a wide margin for multiple years, they will assume deferred maintenance and increase the assumed run-rate.
5. Management continuity assumption
What buyers test: does the projected EBITDA require the founder to remain operating? Which specific functions are currently performed by the founder with no backup? If projected revenue includes the founder's personal customer relationships, buyers discount the revenue they attribute to those relationships to reflect management continuity risk.
Making projections defensible
The most common model problem in founder-owned businesses is revenue projections that reflect ambition rather than evidence. A buyer will not accept 25% growth in year one when trailing history shows 8%, unless the model documents specifically what will change, why, and what evidence supports the assumption. On a $3M EBITDA business at 6x, an unsupported projection 15% above the credible baseline costs sellers $2.7M in LOI valuation when buyers haircut the projected EBITDA to the defensible level.
A $17M technology services business presented a forward model showing 22% revenue growth in year one, based on three enterprise contracts 'in late-stage negotiation.' The PE buyer's financial model accepted 8% growth, the trailing 3-year average, as the base case and discounted the $4.5M of 'in-negotiation' revenue to zero.
The resulting EBITDA delta was $720K.
At 7x, that was $5M of enterprise value difference between the seller's implied valuation and the buyer's LOI. The founder had never built a structured pipeline document that would have supported the revenue claim. The three contracts were real. The documentation did not exist.
Defensible projections are built bottoms-up by revenue category, with explicit assumptions about customer additions, retention rates, pricing changes, and product or service expansion. Every material assumption should have a supporting data point or a clearly documented rationale. Projections that cannot be explained in a management presentation will not survive a diligence Q&A.
Frequently asked questions
What time period should my financial model cover?
The standard in lower-middle-market M&A is 36 months of historical data plus a 3 to 5 year forward projection. Buyers typically underwrite based on LTM EBITDA but evaluate the trend over the full historical period. A business showing consistent growth is valued differently than one showing a recent spike, even at the same LTM EBITDA.
How detailed should the revenue build be?
At minimum, revenue should be categorized by type (recurring, transactional, project) and by major customer or customer segment. Sellers who provide customer-level revenue detail with contract tenure, renewal dates, and historical growth rates receive significantly less pushback on revenue quality during diligence.
What is the most important thing to get right in the model?
The EBITDA bridge. Every buyer will rebuild your EBITDA figure and test every addback. The bridge from GAAP EBITDA to adjusted EBITDA should be explicit, fully documented, and internally consistent with the addback schedule in the CIM. Inconsistencies between the model and the CIM are the most common cause of management credibility concerns in financial diligence.
Work with Glacier Lake Partners
Request a Financial Model Review
Most useful when preparing for a process or completing a management presentation.
Start a Conversation →AI diligence angle
See where AI can clean up readiness before buyers ask.
Run a short scan to identify reporting, data room, and workflow gaps that could affect diligence confidence.
Run an AI readiness scan →Research sources
Disclaimer: Financial figures and case-study details in this article are anonymized, composite, or representative examples based on middle market operating situations, and are not guarantees of outcome. Statistical references are drawn from cited third-party research; individual transaction and operational results vary based on business characteristics, market conditions, and deal structure. This content is for informational purposes only and does not constitute legal, financial, or investment advice. Consult qualified advisors for guidance specific to your situation.
Explore adjacent topics
