For CFOs, controllers, and financial analysts, few post-acquisition tasks carry as much complexity—and as much risk—as purchase price allocation (PPA). Under ASC 805 and IFRS 3, every business combination demands that you identify and measure tangible assets, intangible assets, and assumed liabilities at fair value. The process is notoriously painful: you must coordinate with valuation specialists, reconcile deal models with target company data, allocate goodwill, and document every assumption for auditors. A single misstep—misidentifying a customer relationship intangible, applying the wrong discount rate, or failing to support a contingent consideration estimate—can trigger restatements, impairment charges, or SEC scrutiny. The friction is real: weeks of back-and-forth, expensive advisor fees, and sleepless nights during audit season.
AI tools, particularly large language models fine-tuned for financial analysis, are transforming this workflow. Instead of manually sifting through hundreds of pages of target company financials, contracts, and valuation reports, you can now instruct an AI to extract relevant data, apply the appropriate valuation methodologies, and generate draft allocation schedules. The AI does not replace your judgment—it accelerates the mechanical work, flags inconsistencies, and produces a defensible first draft. For a controller managing multiple acquisitions in a quarter, this means cutting PPA cycle time by 40–60% and reducing reliance on external valuators for routine allocations. The result: faster close, lower costs, and stronger audit documentation.
But the key to unlocking this value lies in prompt engineering. A generic request like “help me allocate the purchase price” yields generic output. You need structured, context-rich prompts that feed the AI the right reference materials, define success criteria, and enforce your firm’s accounting policies. Below are two production-ready prompts designed for Claude (or similar advanced models) that will guide you through the entire PPA process—from data extraction to final allocation memo.
Why Standard Prompts Fail for PPA Work
Most financial professionals treat AI like a search engine: ask a question, get an answer. But PPA is not a question-and-answer exercise—it is a multi-step analytical process. The AI needs to understand the deal structure, the target’s business model, the relevant GAAP guidance, and your specific valuation approach. Without explicit instruction, the model will hallucinate asset values, misapply the residual method, or produce output that looks plausible but fails audit scrutiny. The structured prompts below solve this by forcing the AI to read your source documents, confirm its understanding, and produce output in a format that your audit team can review line by line.
First, read these files completely before responding:
[purchase_agreement.md] — The signed purchase agreement including consideration structure, earnout terms, and closing adjustments
[target_balance_sheet.md] — Target company’s most recent pre-acquisition balance sheet with asset and liability details
[target_contract_list.md] — Summary of all material customer contracts, licenses, and leases held by the target
[valuation_assumptions.md] — My firm’s standard discount rates, tax rates, and useful life assumptions for intangible asset classes
Here is a reference for what I want to achieve:
I am attaching a sample PPA memo from a prior acquisition (reference_ppa_memo.md). This memo was accepted by our external auditors without comment.
Here’s what makes this reference work:
– Each intangible asset class (customer relationships, technology, trade names) is identified with a clear basis for recognition
– Fair values are calculated using the multi-period excess earnings method for customer intangibles and relief-from-royalty for trade names
– Goodwill is presented as the residual, with a reconciliation to total consideration
– All assumptions (growth rates, attrition rates, royalty rates) are footnoted with sources
– The memo includes a sensitivity table showing impact of 10% changes in key assumptions
Here’s what I need for my version / SUCCESS BRIEF:
Type of output + length: A complete PPA allocation memo in structured markdown, approximately 3-5 pages when printed. Include a summary table, detailed asset-by-asset valuation, goodwill calculation, and assumption footnotes.
Recipient’s reaction: My audit partner should be able to read this and say “this is consistent with our prior methodology and sufficiently documented.”
Does NOT sound like: A generic textbook explanation of PPA. Avoid theoretical discussions of valuation methods—just show the calculations with real numbers.
Success means: The memo can be submitted directly to the engagement team as a working draft, requiring only partner-level review of assumptions.
My context file contains my firm’s accounting policies manual, specifically the sections on business combinations and intangible asset useful lives. Read it fully before starting.
DO NOT start executing yet. Ask clarifying questions first.
Give me your execution plan (5 steps max) before you begin.
This first prompt is designed for the initial allocation—when you have all the raw deal documents and need to produce a structured memo. Notice how it forces the AI to ingest five specific files before generating anything. The “reference” section is critical: by uploading a prior approved memo, you effectively train the model on your firm’s preferred format, tone, and level of detail. The success brief ensures the output is practical, not academic. Before the AI writes a single line, it must ask clarifying questions—this prevents it from making assumptions about discount rates or useful lives that conflict with your actual policies.
Refining the Allocation with Sensitivity Analysis
Once the initial draft is complete, most PPA work moves into a refinement phase. The audit team will challenge certain assumptions—is a 15% customer attrition rate too aggressive? Should the royalty rate for the trade name be 2% or 3%? The second prompt below is designed for this iterative process. It takes the draft memo from the first prompt and forces the AI to run sensitivity scenarios, identify the most sensitive assumptions, and produce a variance analysis. This is where AI truly shines: it can recalculate the entire allocation across dozens of assumption combinations in seconds, something that would take an analyst hours in Excel.
First, read these files completely before responding:
[draft_ppa_memo.md] — The draft PPA memo generated from the initial allocation prompt (includes all asset values, assumptions, and goodwill calculation)
[audit_review_comments.md] — Specific questions and challenges raised by the audit engagement team during their initial review
[assumption_ranges.md] — My firm’s acceptable range for each key assumption (e.g., discount rate 10-14%, attrition 8-18%, royalty 1-4%)
Here is a reference for what I want to achieve:
I am attaching a sensitivity analysis workbook (sensitivity_reference.xlsx) that our external valuation firm provided for a prior acquisition. It shows a tornado chart, a table of scenario outcomes, and a narrative explaining which assumptions drive the most value change.
Here’s what makes this reference work:
– Each assumption is tested independently (one variable changes, others held constant)
– The impact is shown in absolute dollar terms and as a percentage of total goodwill
– The narrative explicitly flags assumptions where a 10% change causes more than 5% movement in goodwill
– The output includes a “defensibility score” for each assumption based on available market data
Here’s what I need for my version / SUCCESS BRIEF:
Type of output + length: A sensitivity analysis report in structured markdown, approximately 2-3 pages. Include a table of base case vs. each scenario, a list of top-3 sensitive assumptions, and recommended audit response for each.
Recipient’s reaction: The audit manager should be able to use this report to focus their testing procedures on the highest-risk areas.
Does NOT sound like: A generic risk management lecture. Avoid phrases like “it is important to consider.” Just show the numbers and the recommended action.
Success means: The audit team accepts the sensitivity analysis as sufficient documentation for assumption testing, reducing the number of follow-up requests.
My context file contains the specific materiality thresholds used by our audit firm (5% of goodwill is considered material). Read it fully before starting.
DO NOT start executing yet. Ask clarifying questions first.
Give me your execution plan (5 steps max) before you begin.
This second prompt is particularly powerful because it directly addresses the most common audit pain point: assumption support. Auditors rarely challenge the math in a PPA—they challenge the assumptions. By having the AI produce a formal sensitivity analysis with defensibility scores, you give the audit team exactly what they need to sign off. The prompt forces the AI to read the audit comments and the acceptable assumption ranges, ensuring the output is responsive to real feedback, not generic theory. The success brief makes clear that the goal is to reduce follow-up requests—a measurable outcome that every controller will appreciate.
Practical Tips for Implementation
To get the most out of these prompts, follow three rules. First, always upload reference documents. The AI’s output quality improves dramatically when it has an example of what “good” looks like for your specific firm and auditor. Second, never skip the “ask clarifying questions” step. If the AI does not ask questions, it is likely making assumptions that will produce unusable output. Force it to confirm the deal date, the consideration structure, and the applicable accounting standards before it writes anything. Third, use the sensitivity analysis prompt as a final checkpoint before submitting to auditors. If the AI flags a particular assumption as highly sensitive (e.g., a 10% change in discount rate shifts goodwill by 15%), you know exactly where to focus your documentation efforts. Over time, you can build a library of reference memos and assumption ranges that allow the AI to produce first-draft PPA work in under an hour—work that previously consumed three to five days of analyst time.
Try these prompts on your next acquisition. Start with a simple, low-risk deal—a small asset acquisition or a bolt-on with no contingent consideration. After one or two cycles, you will have refined your context files and reference materials to the point where the AI produces output that requires only minor adjustments. That is the moment when PPA shifts from a painful bottleneck to a streamlined, repeatable process in your M&A workflow.
Published on 10 June 2026 on growwithgpt.com