An AI acquisition memo is not a substitute for a banker. Bankers are still doing the relationship work, the price negotiation, and the regulatory math. AI's actual role in M&A is to compress the analysis cycle: the 3-week scramble of synthesizing target financials, comps, synergy math, integration risk, and strategic fit into a board-ready document. With the right structure, that cycle drops to one focused day. The board memo gets sharper, not weaker, because the analyst running the process spends 80% of their time on judgment rather than formatting.
Skip ahead to our free acquisition memo generator if you want the working tool. Below is the 9-section structure boards consistently respond to.
Goes on page 1, never longer than one page. Five elements: target name, proposed price + structure, strategic rationale (3 bullets max), key risks (3 bullets max), recommendation. Board members read this page; they may or may not read the next 30. The executive summary has to stand alone.
Why this target, why now, why this price. Three to five paragraphs. The hardest part: distinguishing a real strategic rationale from a tactical opportunity. "Their cash position is weak and we can get them cheap" is tactical. "Combining their data network with our distribution closes the gap to the category leader by 18 months" is strategic. Boards approve strategic rationales. Tactical rationales get tabled.
Business model, revenue, growth rate, gross margin, headcount, geographic footprint, customer concentration. AI is excellent here — feed it the target's S-1 (if public) or data room (if private) and it produces a clean, fact-dense overview in 15 minutes. The human review focus: customer concentration and revenue recognition. Both are where bad acquisitions hide.
Three methods, never one: comparable company analysis (trading multiples), precedent transactions (M&A multiples), and DCF (discounted cash flow). Each method produces a valuation range. The recommended price sits inside the overlap. AI is good at producing the spreadsheets and the comp tables; the analyst tunes the multiples for stage, geography, and quality discount.
Revenue synergies and cost synergies, separately, with realistic timelines. The single biggest acquisition-memo failure mode is overstating synergies — boards have seen it too often and discount it automatically. AI flags synergy claims that don't have a named operational owner. If no one's accountable for the synergy line, it's modeled as zero in the recommended scenario.
Day 1, Day 30, Day 100. What changes, who owns it, what stays separate. Boards approve memos with thin integration plans only when the acquirer has a strong M&A track record. For first-time acquirers, the integration plan is the section that gets the most scrutiny.
Eight categories: customer concentration, key-person retention, regulatory, IP, financial reporting, integration culture, technology debt, and competitive response. AI can populate the first seven from the data room; the eighth (competitive response) requires human judgment.
Pro-forma combined P&L for the next 3 years. EPS accretion/dilution. Leverage impact. Return on invested capital. This is the section the CFO will defend in the board meeting, so accuracy matters more than narrative. AI produces the model; the CFO validates every number.
Specific: "Recommend the board approve an acquisition of [Target] for $X in [structure], subject to [conditions]." Not "consider," not "discuss." Boards want a decision-ready recommendation.
Identify 5–10 publicly traded companies with similar business model, stage, geography, and growth profile. Pull their EV/Revenue, EV/EBITDA, EV/Free-Cash-Flow multiples. Apply a median or weighted average to the target's metrics. AI is excellent at building the comp set and pulling the multiples; the analyst's job is to argue for or against each comp.
Identify 5–10 M&A transactions in the target's category over the last 24 months. Pull deal value, revenue/EBITDA multiples paid, strategic vs. financial buyer mix, deal structure (cash, stock, earnouts). Apply to the target. This method usually produces the highest valuation, because acquisition premiums include synergy assumptions. Boards know to discount it.
Project the target's free cash flows for 5–10 years, terminal value, discount at WACC. Most quoted, often least useful, because the inputs (long-range projections) are largely guesses. The value of DCF is the sensitivity analysis: what does the deal look like at WACC +1%? At growth -200bps?
The traditional cycle: analyst spends 2–3 weeks pulling data, building models, writing draft, iterating with the deal team. AI compresses this to one focused day if structured right.
It overstates synergies. The model defaults to optimistic synergy assumptions because the training data is full of M&A press releases that promised synergies. Always strip 30–50% off AI's first synergy pass before showing the deal team.
It underestimates integration risk. Cultural integration, key-person retention, and post-close attrition are systematically underweighted by AI. Half of acquisitions destroy value not at the deal level but at integration. The risk register has to be human-led on these three lines.
It hallucinates comp transactions. If you ask the model "what comparable M&A transactions happened in this category in 2025," it may invent deals that didn't happen, with multiples that aren't real. Every named transaction needs to be verified against a real source (PitchBook, S&P, Bloomberg, or the original press release).
It produces generic strategic rationales. The model has read 10,000 corporate-strategy decks and defaults to phrases like "create scale advantages" and "unlock new market segments." Always rewrite the strategic rationale section in your own voice. If a board member can't tell whether the rationale was AI-generated, the rationale is too generic.
Any memo where the valuation range is presented without sensitivities. Boards want to see what happens if growth is 200bps lower, if churn is 100bps higher, if WACC ticks up. AI can produce these in seconds; the analyst should never present a static valuation.
Any memo without a "no-go" condition. "We recommend this deal" without naming the specific conditions that would change the recommendation is a memo without a spine. Sophisticated boards trust memos that include "we would recommend against this deal if [target's top customer concentration exceeds X%] or [retention of the founding CEO can't be secured]."
Our free AI acquisition memo generator runs the 9-section structure, produces the 3-method valuation, builds the risk register, and outputs a board-ready first draft. Designed for compression, not for replacing the deal team.
QADIR OS — local-first AI for strategic finance. M&A analysis, board memos, valuation models. Your deal data never touches a third-party cloud.
Join Early Access →