An investment committee paper has one job: give a committee everything they need to make a specific decision, and nothing that distracts them from it.
Most IC papers fail at the second half of that sentence. They are comprehensive, well constructed, and structured so that the committee spends its first twenty minutes locating the recommendation and its last twenty debating a chart.
Here is a structure that avoids both, section by section, with what each part has to contain.
Page 1: the recommendation
State it first. Committee members read for the recommendation regardless of where you put it, so placing it on page fourteen just makes them skim thirteen pages impatiently.
Page one carries:
- The recommendation, in one sentence.
- The amount, the instrument and the ownership position.
- The headline valuation and the implied multiples.
- The two or three factors the decision genuinely turns on.
- The specific approval being requested, including any conditions.
That fourth item is the one that separates a good paper from a thorough one. Every deal has forty considerations and two that decide it. If you cannot name the two, the committee will pick their own, and it will be the ones that are easiest to argue about.
Section 1: the investment thesis
One page. The argument in a form somebody could repeat accurately after reading it once.
We are acquiring X because Y. Value is created through A, B and C. We believe we can achieve this because of D. The exit route is E, at a multiple supported by F.
Then the counter case, honestly stated. What would have to be true for this to fail, and why you believe it is not.
A thesis page without a counter case reads as advocacy. Committees discount advocacy and trust papers that have visibly stress tested themselves.
Section 2: the business and market case
What the company does, how it makes money, who buys from it and why, its position in the market, and the market's direction.
The rule that keeps this section useful: every claim about the market carries a source and a date. Market growth assumptions drive the model, and a growth rate from a three year old report is a different input from one supported by current data. The committee needs to know which it is.
Include the competitive picture honestly, including the competitors who are performing better than the target.
Section 3: the financial case
Historicals, base case, and downside. Not a spread of five scenarios, which produces discussion of the scenario framework rather than the deal.
Requirements:
- Every figure resolves to a source, model version and as of date.
- The base case assumptions are listed explicitly on the page, not in an appendix nobody opens.
- The downside is a genuine downside, not the base case with slightly lower growth.
- Bridges are shown, so a committee can see what drives the change between periods.
- The returns calculation shows the components: entry multiple, growth, margin, debt paydown and exit multiple.
That last one matters more than anything else in this section. A return that depends primarily on multiple expansion is a different investment from one that depends on operational improvement, and a committee should be able to see which they are approving in ten seconds.
Section 4: diligence findings and open items
The section committees actually turn to first once they trust the paper.
Present findings by workstream, with severity, and be specific. "Commercial diligence confirms customer concentration risk" is not a finding. "Top three customers represent 47 percent of revenue, with the largest contract renewing in fourteen months and containing a ninety day termination right" is.
Then open items, explicitly:
| Item | Owner | Resolution date | Impact if unfavourable |
|---|---|---|---|
| Pension scheme valuation | Legal and actuarial | Two weeks before signing | Up to a defined value adjustment |
| Key customer contract assignment | Commercial | Pre signing | Material to the thesis |
Hiding an unresolved item to keep the paper clean is the fastest way to lose a committee's trust permanently. They find out. They always find out, and then everything else in the paper becomes suspect.
Section 5: the risk register
Specific risks with an assessment and a mitigation. Not a generic list that could apply to any transaction.
For each: what could go wrong, how likely, what the impact would be, what mitigates it, who owns the mitigation, and what residual risk remains after mitigation.
The honest test for this section: does it contain a risk that could genuinely stop the deal? If every risk listed is comfortably mitigated, the section is decoration and the committee will treat it accordingly.
Section 6: structure, terms and conditions
Deal structure, consideration and its form, key terms, conditions precedent, warranties and indemnities in outline, financing structure and covenants, and the governance arrangements post completion.
Keep it factual. This section is reference material and should be scannable rather than narrative.
Section 7: value creation plan
What you will actually do after closing, with a timeline, an owner for each initiative, the resource required, and the value attributed to each.
Two disciplines that improve this section considerably:
Attribute value specifically. If the plan claims a margin improvement, say which initiatives deliver it and how much each contributes. A plan whose value is unallocated is a hope.
Name the owner. Not "the management team". The person. If the plan depends on hiring someone who does not yet exist, that is a risk belonging in section five.
Section 8: exit
The exit route, the likely buyer universe, the timing, the multiple assumption with its basis, and what has to be true at exit for the assumption to hold.
Committees discount exit sections heavily because they are the most speculative part of the paper. You earn credibility here by being specific about the assumption's basis rather than confident about its magnitude.
Section 9: the decision requested
Close by restating exactly what you are asking for. The amount, the conditions, the authority sought, the timeline, and what happens next if approved.
The appendix
Everything the committee might want and does not need to read: detailed financials, full diligence reports, market analysis depth, management biographies, technical detail, legal detail.
The appendix is not a dumping ground, though. Index it, so a member who wants the customer concentration analysis can find it in five seconds during the meeting.
The IC paper review checklist
- The recommendation is on page one, with the decision factors named.
- The thesis includes an honest counter case.
- Every market claim carries a source and a date.
- Every financial figure resolves to a model version with an as of date.
- Base case assumptions are visible on the page, not buried.
- The downside case is a genuine downside.
- The returns bridge shows what actually drives the return.
- Diligence findings are specific and quantified, not categorical.
- Every open item has an owner, a date and an impact assessment.
- The risk register contains at least one risk that could stop the deal.
- Value creation initiatives have named owners and attributed value.
- Numbers reconcile across the main body and the appendix.
- Confidentiality markings and distribution are correct for this committee.
- The appendix is indexed.
- A second reader who did not write it has checked the whole paper.
Where AI helps, and where it must not
The mechanical parts of an IC paper are substantial and automate well: market data assembly, comparable tables, financial exhibits from an approved model, standard structural sections, diligence finding summaries with citations, and reconciliation across the document.
Three hard limits.
The recommendation is human. So are the valuation view and the risk assessment. A committee approving a recommendation that no named person formed has a governance problem, and the paper should make it obvious who formed it.
Every figure resolves to a source. In an IC paper an unsourced number is not a quality issue, it is a decision made on fiction. The system must flag what it cannot source rather than producing something plausible.
Deal boundaries hold. Retrieval never crosses into another transaction, and the system must enforce that rather than relying on user discipline.
Where offgen fits
An IC paper is a structured document that repeats with every deal, which is exactly the shape a defined skill handles well. The Structured Skill Builder lets you define the section structure, which parts vary, which sources are permitted for each and which elements stay locked.
Output stays natively editable, which matters because IC papers get edited late and by several people, including committee members annotating in the meeting. Source references survive into the file, so a question about page eleven takes seconds to answer. Retrieval runs inside deal team boundaries.
The habit worth building: write page one first, before any other page exists. If you cannot state the recommendation and the two deciding factors in a paragraph, the analysis is not finished, and no amount of appendix will disguise that from a good committee.
Frequently asked questions
What should an investment committee presentation contain?
A recommendation on page one, the investment thesis, the business and market case, the financial case with the base case and downside, the diligence findings and open items, the risk register with mitigations, the structure and terms, the value creation plan, the exit view, and the specific decision requested. Everything else belongs in the appendix.
How long should an IC paper be?
Short enough that every committee member reads all of it. The main body should carry the decision, with depth pushed into appendices. A committee that skims a hundred page main body is making a decision on the first ten pages while believing otherwise.
What is the most common failure in IC papers?
Burying the risks. A paper that presents the case persuasively and lists risks generically in a late section invites a committee to approve on the strength of the narrative. The risk section should be specific, quantified where possible, and honest about what would make this deal fail.
Should the recommendation appear on page one?
Yes. Committee members read the recommendation first regardless of where you put it, so putting it anywhere else just makes them hunt. Page one states the recommendation, the amount, the key terms and the two or three factors the decision turns on.
How should open diligence items be presented?
Explicitly, with an owner, a resolution date and an assessment of what happens if the answer is bad. Hiding an unresolved item to keep the paper clean is the fastest way to lose a committee's trust, and they will find it anyway.
Can AI help produce IC papers?
It can assemble the mechanical parts: market data, comparable tables, financial exhibits from an approved model, diligence finding summaries with citations, and the standard sections. It should not form the recommendation, the valuation view or the risk assessment, and every figure must resolve to a source.
Sources
- 01Regulation (EU) 2016/679 (General Data Protection Regulation) — EUR-Lex, 2016-04-27. Accessed 26 August 2026.
- 02Editable templates — offgen. Accessed 26 August 2026.
- 03Structured Skill Builder — offgen. Accessed 26 August 2026.
Related articles

About the author
Florian Ploszczyk
Co-Founder and COO, MD
Florian writes about consulting workflows, professional presentations, company knowledge, and the controlled adoption of agentic AI.