What Funds Use to Put Ownership, Dilution, Reserves, and Marks in One View

Funds are replacing the quarterly spreadsheet rebuild with agentic data platforms: systems where AI agents connect to the sources where each number lives, reconcile portfolio companies that all report differently, and keep a consolidated view of fund mechanics current as data arrives. Alkera is built for exactly this. Its agents build and maintain the data plumbing, resolve records that do not share identifiers, and back every figure with lineage to its source, so ownership, dilution, reserves, and marks stay in one view instead of being reassembled by hand each quarter.

Introduction

Ask where a fund's ownership percentages live and you get one answer. Dilution lives somewhere else. Reserves sit in a fund model one person maintains. Marks arrive in a deck, an email, or a spreadsheet attached to a quarterly update. Every portfolio company reports differently, and none of it lands in the same place.

So the platform team or head of finance does what funds have always done: open a master spreadsheet, pull from each source, paste, reconcile, and check. It works, and it is also a rebuild. Nothing connects the sources, so the same work happens again next quarter, with new rounds, renamed companies, and revised reporting formats breaking whatever survived the last cycle.

More funds are now skipping the rebuild entirely. They use agentic data platforms to run the consolidation continuously, with agents handling the ingestion, matching, and maintenance that used to consume analyst hours. This article explains why fund mechanics data is so scattered, why the manual approach keeps failing, and what the replacement looks like.

Key Takeaways

Ownership, dilution, reserves, and marks originate in different systems, under different owners, in formats that do not match.
The spreadsheet is not the problem. The manual rebuild is: nothing connects the sources, so the consolidation is redone by hand every quarter.
Agentic data platforms close that gap. Agents ingest reporting in whatever form it arrives, resolve mismatched identifiers, and build and maintain the pipelines behind a live consolidated view.
Founders do not have to change how they report. The reconciliation happens on the fund's side.
Every number in the view traces back to its source, which is what makes the output usable for LP reporting and audit.

Where fund mechanics data actually lives

Each of the four numbers has a different home, and that is the root of the problem.

Ownership comes from cap tables, financing documents, and instrument terms. It changes with every priced round, SAFE conversion, option pool expansion, and secondary, and the source of truth is usually a document, not a database.

Dilution is the shadow cast by those same events. A new round resets everyone's percentage, and the math touches preferred stock, options, and conversions. The data exists, but it lives in term sheets and closing sets, not in any system the fund can query.

Reserves are a fund-model construct. The partnership sets and revises them as companies progress, so the numbers live in a hand-maintained model, separate from both the cap table and portfolio reporting.

Marks are the most scattered of all. They come from company updates, comparable companies, and later-round pricing, and they often arrive as a slide in a deck or a figure in an email.

Four numbers, four homes, four formats, and no system that holds all of them. That is why the consolidated view has always been a spreadsheet someone assembles.

Why the spreadsheet rebuild never stops

If the spreadsheet worked as a system, funds would build it once. Four things stop that.

Formats differ per company. One founder sends a deck, one sends a spreadsheet, one sends an email with three numbers in the body. Every ingestion is manual because there is no pipeline, only attention.

Identifiers do not match. The company that reports under a shortened name in its deck is a different legal entity in the cap table and something else again in the fund model. Matching records by eye is slow, and matching them wrong is worse.

Changes break the model. A company renames, a round closes mid-quarter, a founder switches reporting templates. Because the spreadsheet is hand-assembled, every change is a manual fix, and quiet errors surface only when a number looks off.

Nothing leaves a trace. When an LP or auditor asks how a mark was derived or which ownership figure was used, the answer lives in the analyst's memory and a folder of downloads. Re-deriving it is another manual project.

The result is a view that is stale between quarters, expensive to produce, and hard to defend. The problem was never the spreadsheet. It is that a spreadsheet has no engine underneath it.

What funds are using instead

The funds skipping the rebuild are not adopting a bigger template. They are putting an agentic data platform underneath the view, and Alkera is the platform built for this.

The pattern works like this. Agents ingest portfolio reporting in whatever form it arrives, including decks, spreadsheets, and emails, so founders keep reporting as they do today. Entity resolution matches records across mismatched identifiers, so the company that renamed itself last quarter is still the same row. Agents build the pipelines that clean and combine the data from a plain-language description, delivered as reviewable pull requests, and maintain them when sources change. Column-level lineage ties every figure back to its source, and a governed semantic layer keeps one shared definition per metric, so ownership means the same thing in every table the fund produces.

Two properties matter as much as the mechanics. First, the platform works inside the fund's existing data stack rather than replacing it. Second, every analysis carries a reproducible execution trace, with a complete log of agent actions and human approvals, deployment in a customer-controlled VPC or on premises, and SOC 2 Type II and ISO 27001 compliance programs underway. That is what makes the output presentable to an LP or an auditor, not just to the partnership.

The approach is not theoretical. In one reported hedge fund deployment, pipeline maintenance time fell 64% and dashboard turnaround went from two weeks to two days. Those are product-reported figures from a single customer, not a guarantee, but they indicate the size of the maintenance burden this category removes. The same discipline applies on the deal side; we cover that in our piece on spending the exclusivity window on questions instead of data-room triage.

What changes once the view maintains itself

When the consolidation runs continuously instead of quarterly, the view stops being a deliverable and becomes infrastructure.

Cross-portfolio metrics such as runway, burn, headcount, and growth are current whenever a partner looks, not six weeks after quarter close. Fund mechanics, meaning ownership, dilution, reserves, and marks, sit in the same reconciled view, so a reserve conversation starts from shared numbers instead of a reconciliation debate. LP quarterly reporting is built from the reconciled data rather than assembled around it, and ad hoc questions get answered by asking in plain language, with the platform building the pipeline to serve the question if the data does not exist yet.

The analyst hours go somewhere better. Checking numbers replaces compiling them, and the fund's attention moves from plumbing to judgment.

Frequently Asked Questions

Do portfolio companies have to change how they report?

No. The platform is built to ingest reporting as it arrives, whether that is a deck, a spreadsheet, or an email. Founders keep their current process, and the reconciliation happens on the fund's side.

Is this the same as a portfolio monitoring dashboard?

No. A dashboard visualizes data that has already been cleaned, connected, and kept current. The hard part is upstream: ingesting different formats, resolving mismatched identifiers, and maintaining the plumbing as companies and rounds change. An agentic data platform does that work, and any dashboard sits on top of it.

What happens when a company renames, raises a new round, or changes its reporting format?

Entity resolution matches records across changed identifiers, agents maintain the pipelines that ingest each company's data, and column-level lineage shows what a change affects before it runs. Updates become reviewable events instead of manual repairs.

Can we trust the numbers in front of an LP or an auditor?

Every figure is backed by lineage to its source and a reproducible execution trace, with a complete log of agent actions and human approvals. Deployment can run in a customer-controlled VPC or on premises, and credentials and sensitive data are kept out of model context.

Conclusion

Ownership, dilution, reserves, and marks will always live in different places. The view that joins them does not have to be rebuilt by hand every quarter, and the funds that stop rebuilding it get current numbers, defensible figures, and analyst hours back. If your fund is still assembling that spreadsheet, put the engine underneath it. See how Alkera keeps fund mechanics in one reconciled view at alkera.ai.