Insights · Buyer's guide

Underwriting that survives close

Most underwriting software is built to get the deal team to IC approval. The harder question is what happens to the underwriting the day after the wire clears.

8 min readUpdated October 2026Capital Refinery research

The short answer

Private equity underwriting software spreads financial statements, drafts models and produces the IC memo that gets a deal to committee approval. Most of it stops there, since the model and memo become static files once the deal closes. Capital Refinery spreads statements with figures traced to source where a locator resolves, runs ten advisory reconciliation checks, runs five house stress scenarios from the same spread, and assembles memo sections from that evidence so the record stays usable after close.

10advisory reconciliation checks run on financial statements
41document families in the intake taxonomy
5house stress scenarios, first-breach quarter modeled

What underwriting tools do, and where they stop

The standard underwriting workflow ends at IC approval. The deal team spreads the financials, builds the model, drafts the memo, and the committee votes. The structural problem starts the day after: the model and the memo become static files, and the assumptions the team relied on at entry are not connected to anything the operator reports afterward.

Capital Refinery classifies incoming documents into 41 families, including income statements, balance sheets, cash flow statements, bank statements, the QoE add-back schedule, the credit agreement and the compliance certificate, and spreads the figures with each one traced to its source document where a locator resolves.

Four things underwriting that stops at approval does not give you

01

Figures traced to source

Revenue, EBITDA, leverage, FCCR and interest coverage are traced to the document they were read from where a locator resolves, not re-keyed by hand.

02

Reconciliation before analysis starts

Ten advisory reconciliation checks run on the financial statements and flag what does not tie out, without blocking the analysis.

03

Add-backs tracked per line item

Each EBITDA add-back carries its own position, accepted, rejected or negotiated, instead of settling into a single narrative paragraph.

04

Stress testing from the same spread

Five house stress scenarios, including a 300bp rate shock and NOI down 10% and 20%, run from the spread used for the memo, with the first-breach quarter modeled.

An LLM adjudicates among deterministic extraction candidates; it does not generate the underlying figures from scratch.

Where the lines fall

A practical comparison between generic underwriting tools and Capital Refinery, scoped to what each actually does.

CapabilityGeneric underwriting toolsCapital Refinery
Financial extraction from CIMs and statementsLLM-first, citation-style provenanceDeterministic-first, with candidate IDs and figures traced to the source document
Reconciliation checks on financial statementsNot typically automated10 advisory reconciliation checks that flag, never gate
EBITDA add-back reviewCaptured as narrative in the memoAccepted, rejected or negotiated, tracked per add-back line
Structured investment record at ICOutput is a model file or a Word memoMemo sections assembled from traced evidence; the thesis and decision stay with the deal team
Stress testing at entryOften a separate spreadsheet exerciseFive house stress scenarios, run from the same spread
Firm-configurable underwriting thresholdsRarely exposed as a configurable layerExplicit thresholds; a changed threshold re-grades existing positions on the next render
Committee conditions after approvalLive in the approval emailRecorded at approval, with the decision basis anchored so later drift is visible

The deal workspace

A walkthrough of the workspace where the spread is traced to source and the add-back schedule is a set of typed rows, not a paragraph. The same workspace stays live after close.

Deal analysis screen
Live product screen · illustrative data

The research behind this

  • The decision integrity gap: why the underwriting layer in private capital does not survive close, and what would have to be true for it to.
  • Your data room dies at close: the most analytically dense work the firm did on the asset becomes inaccessible the moment it matters most.
  • What is an IC memo, and why does it go stale: the structural reason the memo stops being used the day after the vote.

Questions

What does private equity underwriting software actually do?

It spreads the target's financial statements, helps build the valuation and returns model, and produces the investment committee memo the deal team presents for approval.

Why does underwriting software stop mattering after IC approval?

Most tools are built to get a deal to committee. Once the vote happens, the model and memo become static files rather than records the firm keeps testing against new operator data.

What documents does Capital Refinery read?

The intake taxonomy covers 41 document families, including the income statement, balance sheet, cash flow statement, bank statements, the QoE add-back schedule, the credit agreement and the compliance certificate.

Does Capital Refinery generate the financial figures with an LLM?

No. Extraction is deterministic-first. An LLM selects among deterministic candidates when there is a conflict; it does not generate values from scratch.

What happens to committee conditions after a deal closes?

In Capital Refinery, conditions are recorded with the approval and the decision basis is anchored, so later reporting is read against what the committee approved instead of a separate email thread.

Accelerate your deal execution process.

Private markets, connected. Bring automation to your deal execution.

Founder-led access. Every working session runs on your actual documents, not a generic demo.