daappa/Who we serve/LPs and Fund of Funds
LPs and Fund of Funds

Hundreds of GP reports.
One governed data set.

LPs and Fund of Funds managers receive quarterly reports from dozens or hundreds of fund managers -- each in a different format. Extractor AI processes them automatically. DataHub centralises and governs the data. Analytics delivers consistent portfolio visibility across every manager.

Start a proof of conceptSee how it works
Quarterly reports
Automated extraction from GP reports, capital account statements, financial statements
Consistency
Normalised schemas across all managers regardless of reporting format
Analytics
Portfolio analytics including IRR at FoF, GP, and portfolio company levels
Audit trail
Every value traceable back to source document location
The volume problem

The quarterly reporting cycle at scale

A Fund of Funds managing 50 LP commitments receives 50 quarterly reports -- each in a different format, with different table structures, different calculation conventions, and different metrics. Processing them manually is not a reporting problem. It is an operational one.

Every report contains data that looks similar but isn't. MOIC and TVPI mean different things to different managers. Fund-level and portfolio-level data are mixed. Reconciliation to NAV is manual. The result is that by the time you have a consolidated view, the quarter is nearly over.

Extractor AI was built specifically for this problem. It processes each document regardless of format, normalises the data to your schemas, and populates DataHub automatically. What took weeks takes days.

99% less
Manual time on quarterly document processing with Extractor AI
99%
Extraction accuracy, human-checked on exceptions
How it works for LPs and FoF

From document receipt to portfolio analytics

Four steps from GP report receipt to governed, analytics-ready data.

1
Documents arrive from GPs
Quarterly portco reports, capital account statements, financial statements, and LP letters are uploaded or emailed directly to Extractor AI. PDF, scanned, native, multi-language -- all formats handled.
▼
2
Extractor AI processes each document
Computer Vision AI, Generative AI, and parallel LLM validation extract structured data from each document. High accuracy. Every value linked back to its source location in the original PDF.
▼
3
DataHub normalises and governs
All extracted data is normalised to your schemas, validated, and stored in DataHub -- your single governed data warehouse. Data from 50 managers in 50 different formats becomes one consistent data set.
▼
4
Analytics delivers portfolio visibility
Performance metrics, IRR at FoF and GP level, vintage analysis, KPI trends, and exception reporting -- all from the same governed data set. No manual model maintenance.
Platform capabilities for LPs and FoF

What the platform covers

Every component connects to DataHub. Nothing is siloed.

Extractor AI
Automated extraction from GP quarterly reports, capital account statements, audited financials, and LP letters. Works with your existing schemas. Tier 2 managed validation available with 72-hour SLA.
Full detail
DataHub
Central governed data warehouse. All GP data normalised to consistent schemas across managers. Reconciliation rules applied. Single source of truth for all downstream analytics.
Full detail
Analytics
IRR at FoF, GP, and portfolio company level. Vintage analysis, manager benchmarking, exposure concentration, and performance trend reporting. Built for private markets data structures.
Full detail
Look Through
Ownership and exposure calculation across multiple layers of fund structures. Identify concentration and cross-exposure across GPs within your portfolio.
Full detail
Analyser
Self-service data extract and report builder. Query any DataHub data combination. Respond to investment committee requests and board queries without analyst involvement.
Full detail
Compliance Monitoring
Mandate exposure monitoring, concentration checks, ESG screens, and investment policy compliance across the portfolio. Flags breaches before they become reporting events.
Full detail
Common questions

LPs and FoF — frequently asked

Does Extractor AI work when every GP uses a different report format?
Yes. That is exactly what it is designed for. Extractor AI handles inconsistent formats, narrative-heavy documents, and mixed fund and asset level data. The onboarding process configures extraction templates per GP if needed. You do not need to ask your GPs to change how they report.
How does IRR get calculated at FoF and GP level?
IRR and other performance metrics are calculated in DataHub from the validated extracted data -- not during extraction. The separation of extraction, calculations, and analytics ensures that the calculation logic is consistent and auditable, and that the same governance standards apply to every metric regardless of how the source data arrived.
Can we start with a subset of our GP relationships?
Yes. The five-week proof of concept runs on a defined set of GPs and document types. You see results on your actual documents before committing to full rollout. Most clients start with their highest-volume or most complex GP relationships and expand from there.
How does the audit trail work across hundreds of documents?
Every extracted value is linked directly to its source location in the original document through four preserved pipeline stages: raw extraction, validated dataset, consolidated dataset, and final dataset. At any point you can trace any metric back to the specific cell, row, or paragraph it came from in the original GP report.

Process your GP reports in a fraction of the time

Tell us about your GP manager count, your current extraction process, and your reporting cycle. We'll design a proof of concept around your actual documents.

Start a proof of conceptSee Extractor AI