Quarterly LP report: structure GPs use that actually gets read

A senior LP at a fund-of-funds told me she reads the cover page, the fund metrics page, and the one-pagers for the deals she remembers from the last quarter. That’s it. The rest she skims at best. She’s reading thirty quarterly LP reports a quarter — that’s the actual budget. The implication for GPs writing these reports is direct: the two or three pages every LP reads have to be tight, and the rest is supporting evidence for the LPs who go deeper.

This piece is the structural version. Five sections every quarterly LP report has, what LPs skim versus what they read carefully, the data layer that feeds the document, and when automation starts to justify itself. The longer treatment lives at investor update automation.

LP reporting is a structurally conservative category. The format is largely industry convention plus a layer of fund-specific preference. Innovation in LP reporting is rarely rewarded; consistency quarter-to-quarter is. Which makes it an unusually good candidate for automation — same shape every cycle, fresh data each cycle, real cost in analyst hours when done by hand.

The five sections every quarterly LP report has

1. Portfolio performance

The narrative core. Per-company commentary on the quarter — material developments, financings, exits, write-downs, key hires, key losses. The default structure is a one-pager per portfolio company, organised by stage of investment or by size of position.

What LPs skim: the table of contents and the headline metrics on each one-pager (current valuation, last round, ownership, mark). What LPs read carefully: the commentary on companies they already know — the headline names, the recent exits, the recent down rounds. The art is making the commentary substantive without cherry-picking; an LP who notices the same kind of upbeat tone applied equally to a 5x company and a write-down loses trust quickly.

The data layer here is messy. Some of it lives in the fund-admin software (cap table, marks, ownership). Some lives in portfolio-tracking software like Carta, Aumni, or Standard Metrics. Some lives in the GPs’ own notes from board meetings. Some lives in nowhere — it’s in someone’s head and gets typed in fresh each quarter.

2. Capital activity

The accounting core. Capital calls in the quarter, capital distributions in the quarter, recallable capital, remaining commitments, fund expenses. The page reads as a small set of tables, often with a year-to-date and inception-to-date layer.

What LPs skim: whether there were calls or distributions this quarter, and how big. What LPs read carefully: the math against their commitment, especially around the end of the call period. This section is mechanical; the numbers either reconcile to the LP’s records or they don’t, and they need to.

The data layer is the fund admin — Carta Fund Admin, Standish, Gen II, the in-house spreadsheet at smaller funds. The reporting tooling needs to pull straight from there; any layer of human transcription introduces reconciliation errors that show up in LP-side queries within forty-eight hours.

3. Fund metrics

The page every LP reads. DPI (distributions to paid-in), TVPI (total value to paid-in), IRR (internal rate of return). Plus the GP’s preferred secondary metrics — RVPI, MOIC at the deal level, vintage benchmarks. Plus, if the fund is far enough along, comparison to prior funds in the same family.

This page is small (often a single page or a single spread) and dense. It is the page LPs look at first. The layout discipline matters: same metrics, same order, every quarter. An LP comparing Q3 to Q2 should see the same labels in the same position; the visual continuity is what makes the comparison readable. Funds that re-skin this page to look better when the numbers improve get caught — and lose credibility — within a couple of cycles.

The data layer is fund admin output, sometimes augmented by the GP’s own IRR calculations. The trap is that different fund-admin tools compute IRR slightly differently; the GP needs to commit to one definition and stick with it.

4. Market commentary

The optional section that some funds do well and most do poorly. A page or two on the macro environment, sector dynamics, or thematic observations from the GP’s vantage point. At its best, this is the page LPs forward to colleagues — original signal from a GP who has seen the deal flow that quarter. At its worst, it’s recycled VC-Twitter takes that LPs have already read three times.

The discipline is honesty about whether the GP has anything to say. A market commentary section that has nothing original in it is worse than no section at all; the LP notices, and the noise dilutes the rest of the report. Better to skip the section in quarters when there’s nothing original than to fill it with content for the sake of length.

5. Outlook

The forward-looking close. Pipeline (without naming names where confidentiality requires), planned capital calls, expected exits, hiring or strategy shifts at the GP level. Often paired with a list of upcoming LP events — annual meeting, advisory board, regional gathering.

LPs read this for the operational signals — when is the next call, when is the AGM, what’s the pace of deployment for the rest of the year. It’s the section that affects their own planning. The numbers here don’t need to be precise (and shouldn’t pretend to be) but they need to be defensible — an LP who hears “we expect three exits in H2” and gets one will remember.

What LPs skim vs read

Three pages get the most read time: the cover summary, the fund metrics page, and the one-pagers for the named deals.

Three pages get skimmed: the market commentary, the outlook, and the one-pagers for the smaller positions.

The rest — the table of contents, the appendix, the disclosures, the methodology — is reference material an LP returns to when they have a specific question. The structural implication is that the report’s design hierarchy should match the read hierarchy: the cover and the fund metrics get the most design attention, the commentary gets the next layer, the appendices get template-driven utility design.

This is also why “more pages = better LP report” is wrong. A tight twenty-page report read in full beats a sprawling fifty-page report skimmed for ten minutes. The brevity discipline is the GP’s, not the fund admin’s.

The data layer

LP reporting sits on a particular stack:

Fund admin software (cap table, marks, capital activity, fund expenses): Carta Fund Admin, Standish, Gen II, Aduro, in-house at smaller funds.

Portfolio company data (operating metrics, KPIs, board materials): Standard Metrics, Aumni, Visible.vc, or direct from the portfolio companies on a quarterly request.

Market data (benchmarks, comparables, IRR comparisons): PitchBook, Cambridge Associates, Preqin, or the GP’s internal benchmarks.

GP-internal narrative (commentary, hiring updates, strategy notes): nowhere structured. Lives in the partners’ heads, gets written each quarter.

The fragility of LP reporting comes from this stack — four data sources, three of them owned by external software, one of them owned by no one. The automation work is in binding the four into a single document layer that regenerates on a schedule. See airtable document automation for one pattern that handles the GP-internal narrative layer; the rest is API integration with the fund-admin tooling.

When automation justifies itself

The breakeven for LP reporting automation depends on three things: number of portfolio companies, fund vintage, and report cadence.

Below ten portfolio companies, with a single fund on a quarterly cadence, the report is small enough that one analyst can build it in a week. Automation pays back, but the payback period is long enough that most Fund I GPs reasonably defer it.

Past ten portfolio companies, or past Fund II, or past quarterly cadence (some firms now run monthly), the report becomes a recurring multi-day project. Multiple data sources, multiple authors, multiple review cycles. The error rate in hand-built reports starts to climb — wrong vintage label, stale logo, mismatched ownership percentage between two pages, prior-quarter numbers that didn’t get refreshed. Each error is small; cumulatively they erode LP confidence in the GP’s operational rigour.

This is the inflection where LP reporting automation pays. Not because the tooling is exciting — it isn’t — but because the cost of a mismatched ownership percentage on page eleven of a forty-page LP report is operational credibility, and that’s expensive.

The pattern most mid-stage funds converge on: the fund admin handles the financial pages (capital activity, fund metrics) directly from their software’s reporting layer; a separate document-automation pipeline handles the portfolio-company one-pagers and the narrative commentary, pulling from the portfolio-tracking software and the GP’s own structured notes; the document is assembled and branded in the GP’s preferred template tool. Three sources, one pipeline, one document.

For the founder-side equivalent (different audience, different shape), see monthly investor update template. For the longer architectural treatment, read the investor update automation.