Insights · Buyer's guide
The stack is fragmented. The IC decision falls through the cracks.
A practical map of private markets tooling, and the decision integrity layer most stacks are missing.
Private markets firms typically run tools across market data, sourcing, CRM, generative AI, extraction support, portfolio monitoring, and fund accounting. Each layer does its job well, but none owns the investment committee decision as a structured record. The thesis lives in a memo, the risk signals live in a spreadsheet, and the link between them is rebuilt by hand each quarter. Capital Refinery takes the rest of the stack's output and turns it into a record that survives the deal.
The stack is fragmented
Private markets firms rarely operate on one platform. They run a stack of tools across data, sourcing, CRM, generative AI and fund operations. Each layer is useful on its own. Together they leave the IC decision sitting in a shared drive nobody reopens after the next quarter closes.
The problem is not that each tool fails at its job. The problem is that the IC decision lives in the seams between the tools, and nothing carries it forward.
The layers of the private markets stack
What each layer does well, and where it stops.
Market data & screening
Market context, comps and sector research. Informs the decision; does not defend it.
PitchBookS&P Capital IQPrivCoMoody’s CreditLensCB InsightsPreqinCoStar
Sourcing & relationship workflow
Deal discovery and relationship mapping. Finding the deal is upstream of underwriting it.
GrataAxialAffinity
Deal CRM & workflow
Process discipline, approvals and pipeline visibility. The IC memo and the post-close drift live elsewhere.
DealCloudDealpathAltvia
Diligence & extraction support
Document hosting and review and extraction acceleration inside the diligence window.
DatasiteIntralinksDataSnipperDocuClipper
Generative AI for private equity
Faster drafts and document chat. The output is still unstructured once the deal closes.
HebbiaBlueFlameAlphaSenseEndex.aiEillaMicrosoft CopilotMacabacus
Portfolio monitoring
KPI ingestion, dashboard rendering and LP reporting for the post-close window.
iLEVELAllvueChronographCobalt LP
Accounting & fund operations
The books, the waterfall and the LP statements. Mature, audited, and below the decision layer.
FundCounteFrontStandard Metrics
Decision integrity
The missing layer. Takes the output of the rest of the stack and turns it into a record that survives the deal.
Capital Refinery
How the stack breaks in practice
A typical workflow: the opportunity is found in a sourcing tool, context is gathered in a data platform, the workflow is managed in a CRM, analysis is accelerated with an AI tool, and the reporting is rebuilt by hand in slides and memos. Each handoff is a place the decision basis can go missing.
- Underwriting assumptions are not preserved after the deal memo is filed.
- Downside scenarios live in isolated spreadsheets, disconnected from the live spread.
- Post-close monitoring runs independently of the original deal logic.
- Committee outputs are rebuilt by hand instead of generated from a live record.
What each layer does, and where decision integrity has to come from
| Capability | Layer · tools | Capital Refinery |
|---|---|---|
| Data and intelligence | PitchBook, Capital IQ, PrivCo | Informs the decision; does not defend it |
| Sourcing and relationships | Grata, Axial, Affinity | Finds the deal; does not own underwriting |
| CRM and deal workflow | DealCloud, Dealpath, Altvia | Manages process; the memo and the risk signals live elsewhere |
| Generative AI tools | BlueFlame, Endex.ai, Microsoft Copilot, Hebbia | Produces faster drafts; the output is not a structured record |
| Extraction and review | DataSnipper, DocuClipper | Handles a slice of the document problem |
| Portfolio monitoring | iLEVEL, Allvue, Chronograph, Cobalt LP | Ingests and reports KPIs; does not hold the IC thesis |
| Decision integrity layer | Capital Refinery | Memo sections built from cited evidence; decision basis anchored at approval, drift tracked after close |
Where Capital Refinery sits in the stack
Capital Refinery does not replace the rest of the stack. It reads the output of the layers above it, spreads statements with figures traced to source where a locator resolves, and assembles memo sections from that evidence while the thesis and decision stay with the deal team. The decision basis is anchored at approval, and each compliance certificate after that is tested against its covenant, with headroom and drift.

Questions
What tools make up a typical private markets software stack?
Most firms run separate tools for market data (PitchBook, Capital IQ), sourcing (Grata, Axial), CRM (DealCloud, Affinity), generative AI (Hebbia, BlueFlame), portfolio monitoring (iLEVEL, Allvue) and fund accounting (FundCount, eFront).
Why doesn't a CRM like DealCloud hold the IC decision?
A deal CRM manages process: pipeline stages, approvals and scheduling. The IC memo is typically an attached file rather than a structured record the platform can test against new data.
Does generative AI solve the decision integrity gap?
Generative AI tools speed up drafting and document review, but the output is still an unstructured document once the deal closes. Speed does not create a structured record.
Where does Capital Refinery fit relative to iLEVEL or Allvue?
Capital Refinery does not replace portfolio monitoring. It sits alongside it, reading each reporting pack to its covenant terms and keeping the original IC memo connected to that data.
What is the decision integrity layer?
It is the layer that holds the investment committee's thesis and conditions as a structured record, and keeps it connected to the operator data that arrives after close.