Redacted sample proposal · April 2026
Important note: This discussion note is adapted from a prior bespoke client engagement. The private capital firm described here is fictionalized. It is intended as a representative diagnostic note, not as a description of work performed for any named PE firm.
Keeping the Thesis Live
A Note for Senior Partners of a Lower-Middle-Market Private Capital Firm
Prepared for senior partner discussion
The Platform You’ve Built
The firm has built something difficult: a private capital platform where judgment is still close to the work.
That is not automatic as a firm grows. Many firms either remain small enough that every important call lives in a handful of partners’ heads, or they scale into process-heavy institutions where the sharpness of individual judgment gets diluted into templates, committee materials, and reporting routines.
You have taken the harder path: growing the partnership, expanding the portfolio, adding operating input, deepening sector focus, and increasing deal volume while still relying on the kind of partner judgment that made the firm successful in the first place.
That judgment is the asset.
It shows up in underwriting taste, management read, sector pattern recognition, price discipline, value-creation realism, and the ability to know which risks actually change the deal. It is not reducible to the memo, the diligence report, the model, or the committee discussion. Those artifacts matter, but they are not the thing itself.
The question is what happens when that judgment has to travel further.
The firm now carries more live complexity than it did when a smaller group of partners could stay close to every deal by proximity, conversation, and memory. There are more partners with distinct pattern recognition, more portfolio companies, more active workstreams, more diligence inputs, more operating advisors, more financing constraints, more board cycles, and more live decisions happening in parallel.
The firm does not need every partner to think the same way. The diversity of partner judgment is part of the platform’s value. But as the platform grows, the firm also needs a way to preserve coherence around its house judgment: what risks it takes, what signals it trusts, what breaks a thesis, where it is willing to stretch, and what kind of value creation it believes it can actually underwrite.
Today, too much of that live judgment is carried in people’s heads.
The platform exists. The question is whether the thesis can stay live as the platform scales.
The Private Capital Thesis
The first wave of AI in private capital is moving quickly at the task layer.
Market scans, company research, CIM review, diligence summarization, customer-call synthesis, model support, memo drafting, portfolio reporting, and KPI analysis will all become faster. Some of this is already useful. Much of it will soon be table stakes.
That investment is necessary. It is not sufficient.
Faster task work does not automatically improve investment judgment. In many cases, it increases the pressure on it. More processed output moves upward. More findings arrive faster. More summaries, analyses, and reports accumulate around the same senior investors who already have to decide what actually matters.
The bottleneck has never been the existence of information. The bottleneck is the live state of the thesis.
What do we believe?
What does that belief depend on?
What changed since the last partner discussion?
Which assumption is no longer safe?
Which diligence finding actually reprices the deal?
Which IC question is now live?
Which risk is manageable, and which one changes the nature of the investment?
Where would another partner’s pattern recognition improve the call?
AI is getting faster below the judgment layer. The judgment layer is getting more compressed.
That is where the next advantage sits: not in producing more analysis, but in making the firm’s best investment judgment travel further, more consistently, without flattening it into process.
The scarce object is not the memo.
It is not the diligence output.
It is not the reporting package.
The scarce object is the live thesis.
The Judgment Bottleneck
On live deals, senior investors lose too much high-value attention to reconstruction.
Before a deal call, before IC, after a diligence update, after a lender conversation, after a management meeting, after a weekend of new materials, the partner has to rebuild the picture: what changed, what still matters, which assumption broke, which issue interacts with which other issue, and what decision now requires judgment.
That reconstruction is not clerical. It is not simply “catching up.” It is the mental work of recovering the live investment state.
The same problem appears across the partnership. One partner may have a sharp read on founder psychology. Another may have seen the same customer-concentration pattern before. Another may know when a margin-expansion plan is real and when it is just underwriting optimism. Another may have a scar from a similar debt structure or integration plan.
But those inputs are often hard to activate at the right moment because the shape of the live thesis is not easily visible. Partners do not need to expose their entire internal reasoning to one another. They do need a way to make the decision architecture legible enough for the platform to challenge, contribute, and learn.
Not sharing the monologue. Sharing the model.
As AI accelerates the task layer, this problem gets worse, not better. A diligence team can now process more material in less time. Advisors can produce more synthesized output. Internal teams can generate more analyses. But the partner still has to answer the hard question:
Which of this changes the thesis?
More throughput below the partner does not reduce the need for senior judgment. It concentrates it.
The firm should not solve that by adding process for its own sake. The goal is not to standardize partner judgment. The goal is to make the shape of that judgment recoverable, shareable, and capable of compounding.
The Live Thesis Model
The Live Thesis Model is a maintained structure that sits above task-layer AI and ordinary deal materials.
It gives the senior investor a current decision picture of each live deal: not more material, but the state of the thesis.
The core innovation is not a new software platform. It is the thesis model itself: a shared, maintained structure that captures what the firm believes, what the belief depends on, what has changed, what would change the firm’s mind, which decisions are live, and where partner judgment is required.
Once that structure exists, AI can help maintain it. Without it, AI simply produces more summaries.
The Live Thesis Model tracks the investment thesis as a set of live assumptions, dependencies, risks, decisions, and unresolved questions. It does not replace the IC memo, the financial model, the diligence tracker, or the advisor workstreams. It sits above them and asks a different question:
Are we still buying the company we thought we were buying?
That is the Investment Thesis View.
It is maintained by a VP or principal with AI assistance and validated by the senior investor on a short rhythm. The VP or principal knows the deal. AI handles reconstruction work across materials, updates, notes, and prior decisions. The partner spends a few minutes reviewing and correcting the thesis state rather than rebuilding it from scratch.
The value is not that AI becomes the investor.
The value is that the investor re-enters at the level where judgment matters.
The Investment Thesis View
The Investment Thesis View is the first use of the Live Thesis Model.
It is designed for live deal work: diligence, IC preparation, management conversations, financing decisions, price movement, risk assessment, and go/no-go calls.
Consider a founder-led B2B services or vertical software business. The initial thesis is attractive: durable revenue, pricing opportunity, operating leverage, a credible path to professionalizing the organization, and room for add-on acquisitions.
Then diligence starts moving.
Customer calls suggest the revenue is stickier than expected in some segments and more services-dependent in others. The quality of earnings work shows margin upside, but only if implementation capacity improves. The second layer of management looks thinner than the original founder narrative suggested. A lender conversation tightens the debt case. A potential add-on could strengthen the thesis, but only if management can absorb integration without distracting from the core plan.
None of these findings alone is necessarily fatal.
The question is how they interact.
Does the services intensity undermine the software-like revenue story?
Does the management-bench issue change the pace of the value-creation plan?
Does the tighter debt package reduce the margin for error?
Does the add-on opportunity make the platform thesis stronger, or does it introduce an execution burden the company is not ready to carry?
Does the price still compensate the firm for the actual company now emerging through diligence?
That is the work senior investors do. The problem is that the live state of that work is often distributed across calls, files, notes, models, advisor updates, partner conversations, and memory.
The Investment Thesis View keeps the thesis current.
It does not ask the partner to read everything again. It shows what moved, what matters now, and which decisions deserve senior judgment.
Core Capabilities
Re-entry Brief
When a partner returns to a deal after an interruption, a new diligence update, or several days of other work, the Re-entry Brief provides the current investment state: what changed, what matters now, which assumptions moved, which questions are open, and which decisions require partner judgment.
It is not a general summary. It is a decision brief.
The partner should be able to enter an IC prep session, management call, lender discussion, or internal debate from orientation, not catch-up.
The practical test is simple: does the partner begin using the Re-entry Brief before live deal conversations instead of asking the team for a verbal download?
If yes, the model is doing real work.
Assumption Map
Every investment thesis depends on assumptions.
Some are explicit: revenue growth, margin expansion, churn, working capital, leverage, management capacity, add-on availability, pricing power.
Others are implicit: the founder will accept a different operating cadence, the organization can professionalize without losing commercial energy, the market will continue to reward the company’s positioning, the management team can execute two changes at once.
The Assumption Map makes those dependencies visible.
It tracks what the firm believes, why it believes it, what evidence supports it, what would weaken it, and what decision would change if the assumption moved.
This is where partner judgment becomes legible without becoming bureaucratic. A partner does not need to narrate every instinct. But the model can show the shape of the thesis: where confidence is high, where confidence is borrowed, where a risk is being accepted, and where the firm is close to underwriting hope instead of evidence.
That visibility allows another partner to contribute without having to reconstruct the whole deal.
Decision Compression
Private capital firms do not need more summarized output. They need compression to decision relevance.
A deal may generate hundreds of diligence points. Most belong in the record. A few change price, risk, timing, structure, ownership plan, or go/no-go judgment.
Decision Compression separates those.
It asks:
Which issues actually reprice the investment?
Which interactions between issues matter more than the issues themselves?
Which questions require partner judgment rather than team execution?
Which findings should change the IC discussion?
Which risks can be owned, and which risks alter the thesis?
The goal is not to reduce complexity by hiding it. The goal is to preserve the full record while bringing only the decision-relevant structure into the senior investor’s view.
Everything else remains available. Only what matters reaches the partner’s attention.
Partner Pattern Input
As the partnership grows, one of the firm’s most valuable assets is the pattern recognition distributed across partners.
A partner who has seen a similar founder transition may recognize a risk the deal team has normalized. A partner who has lived through a failed roll-up may see integration burden earlier. A partner with a sector scar may know that a customer-concentration issue is acceptable in one market structure and dangerous in another.
The Live Thesis Model creates a practical way to activate that pattern recognition.
Because the model exposes the shape of the thesis — assumptions, dependencies, confidence, open questions, and decision points — another partner can engage without needing to absorb the entire data room or re-run the deal from first principles.
This is not committee process. It is not consensus-seeking. It is targeted judgment input at the moment it can still change the decision.
The firm does not need every partner to think alike. It needs a way for different partners’ judgment to compound.
Testing It
The test should run for eight weeks on two active deals.
The right senior investors are not necessarily the most enthusiastic about AI. They are the ones currently feeling the re-entry problem most acutely: carrying multiple live deals or assets, moving between IC discussions and board responsibilities, frequently interrupted, and aware of how much attention they spend reconstructing thesis state.
They should volunteer because they feel the pain, not because they want to pilot a tool.
The right deals are not the simplest deals in the pipeline. They are deals where multiple elements of the thesis are moving at once: management quality, customer concentration, debt capacity, add-on logic, margin expansion, pricing, market growth, regulatory exposure, or exit path.
Each deal should have a VP or principal responsible for maintaining the Live Thesis Model with AI assistance. The partner validates it on a short rhythm. The model should be used before real deal moments: IC prep, management calls, diligence readouts, lender discussions, price conversations, and internal go/no-go debates.
Timeline
Weeks one and two: select two deals, define the thesis model, identify the live assumptions, establish the update rhythm, and agree on how the partner will validate the model.
Weeks three through six: live running. The VP or principal maintains the model. AI assists with reconstruction. The partner uses the Re-entry Brief and reviews assumption movement before live deal moments.
Weeks seven and eight: assessment and synthesis. Compare where the Live Thesis Model changed attention, surfaced issues earlier, improved partner input, or altered the quality of the investment discussion.
What Success Looks Like
There are four signals, in order of strength.
First: do partners start using the Re-entry Brief before deal calls or IC preparation instead of asking for a verbal download? That is a behavioral shift.
Second: do assumption breaks or decision collisions surface earlier than they would have through ordinary deal process? Partners will know.
Third: does the model change a real conversation about price, risk, timing, structure, ownership plan, or go/no-go judgment?
Fourth: does either partner voluntarily extend the model to another deal beyond the test scope?
No one adds process unless it helps.
Voluntary extension is the strongest proof.
Why This Is a Private Capital Test
The Live Thesis Model is not a productivity tool.
It is a mechanism for making the firm’s investment judgment compound.
A small partnership can maintain coherence through proximity. Partners talk constantly. Everyone knows the live deals. The house view travels through conversation, memory, and repeated exposure to the same decision-makers.
A growing partnership cannot rely on that alone.
As the firm adds partners, sectors, portfolio companies, operating advisors, and deal volume, the challenge changes. The firm still wants partner distinctiveness. It does not want to turn investment judgment into a template. But it also needs a way to maintain coherence around what the firm believes, what risks it accepts, what evidence it trusts, and what kind of value creation it is willing to underwrite.
The Live Thesis Model gives the partnership a way to do that without bureaucratizing judgment.
It does not force a partner to reveal every internal thought. It does not require every investor to reason the same way. It does not turn the IC process into a compliance exercise.
It makes the shape of the thesis visible enough for others to challenge, contribute, and learn.
That is the platform move.
Once two partners are using Live Thesis Models on live deals, the natural next question becomes: where would another partner’s experience have changed the investment discussion earlier? Which assumptions would have been challenged sooner? Which risks would have been classified differently? Which value-creation claims would have been sharpened before IC?
That is how house judgment becomes more than an implicit culture.
It becomes a live institutional capability.
From Investment Thesis to Ownership Thesis
If the Investment Thesis View works, the natural extension is ownership.
The investment thesis should not die as an IC artifact. It should become the seed of the ownership thesis.
The Ownership Thesis View is the post-close view of the same logic: whether the company is actually becoming the asset the firm underwrote.
Too often, the moment of close creates a discontinuity. The IC discussion, diligence scars, partner debates, management doubts, pricing concessions, value-creation hypotheses, and open risks scatter into separate artifacts: closing materials, 100-day plans, board materials, operating workstreams, lender reporting, and partner memory.
The company becomes an asset, but the live logic of why the firm bought it does not always remain intact.
That matters even more when exit timing is less forgiving. If a firm cannot count on a clean exit window to validate the underwriting case, ownership has to do more of the work. The firm has to keep testing whether the value-creation logic is becoming true, whether the risks it accepted are behaving as expected, and whether the exit story is strengthening or weakening while the asset is still in its hands.
The Ownership Thesis View should pick up where the Investment Thesis View leaves off.
Its governing question is different:
Are we still creating the value we underwrote?
The same assumptions that shaped the buy decision become the operating questions the firm should watch during ownership. The same management concerns become board-cycle priorities. The same value-creation levers become testable hypotheses. The same risks become early-warning signals. The same exit logic becomes a living view of what the company must prove.
The two views have to talk to each other.
Investment Thesis View: are we still buying the company we thought we were buying?
Ownership Thesis View: are we still creating the value we underwrote?
Together, they create continuity across the private capital lifecycle: diligence to IC, IC to close, close to ownership, ownership to exit, and exit back into the firm’s pattern recognition.
This is where the firm’s judgment begins to compound across deals.
Not because every partner remembers everything.
Not because every lesson becomes a generic playbook.
But because the firm preserves the live structure of what it believed, what happened, what broke, what held, and what it should recognize sooner next time.
Getting Started
The test requires no new platform, no major budget approval, and no organizational redesign.
It requires two senior investors willing to try a different way of staying oriented on live deals, two VP or principal maintainers who know the work, and eight weeks.
The work is practical:
select two active deals;
build the Live Thesis Model for each;
establish the update rhythm;
use the Investment Thesis View before real deal moments;
track what changes in partner attention, assumption movement, and decision quality;
then decide whether the model deserves to extend.
We would be glad to work with the partners and teams who take this on: designing the Live Thesis Model, establishing the rhythm, helping the maintainers use AI for reconstruction, and interpreting what the test reveals about the broader platform opportunity.
Do not aim merely to make diligence faster.
Aim to make the firm’s best investment judgment a live capability — one that travels from partner to partner, from deal to deal, and from investment thesis to ownership thesis.
The next private capital advantage is not more information.
It is keeping the thesis alive.
Pricing
Engagements typically run from $95,000 to $175,000, depending on the situation, for an 8-week test on two live deals or portfolio situations.
Please reach out first to elissa@recompound.ai