Skip to main content

Financial blind spots in your portfolio

A $200K vendor nobody flagged. A side project nobody tracked. A pile of expenses without a single document attached. Blind spots like these do not announce themselves. They build up quietly until they become expensive problems, and for investors managing several companies they are the default state.

This page walks through five blind spots we keep running into, and how the parts of Moonlight already in production try to address them. The product is in active development, so we are honest below about what is live, what is partially there, and what is on the roadmap.

Blind spot 1: the invisible project​

A portfolio company has three products. You invested because of Product A. Product A does well. The founder launched Product C eight months ago, and it has been losing $30K a month. You do not know about it because the quarterly report shows company-level totals.

Revenue: up. Expenses: up a little. Net: positive. At the company level, the picture looks healthy.

At the project level, Product A carries the company while Product C drains it. The founder plans to shut Product C down "next quarter", and they have been saying that for two quarters. By the time you hear about it, Product C has consumed $240K of the capital you thought was funding Product A's growth.

How Moonlight helps. Entries are grouped by project, so you see income and expenses for each one separately. Product C's negative trend is visible the month it starts, without waiting for the founder to disclose it.

Blind spot 2: the concentrated vendor​

Across your portfolio, three companies use the same software development agency. Each company pays them separately: $15K here, $25K there, $40K in a third. The total is $80K a month going to one vendor across your portfolio.

No individual company sees this as a problem. From their side, they have a reliable agency relationship. From yours, you have $80K a month of dependency on a vendor that could raise rates, get acquired or go out of business.

How Moonlight helps. Party-level analytics show vendor concentration inside each workspace, and the Sankey flow view maps money movement to specific parties. When the same vendor name shows up across multiple workspaces, the pattern is visible from the data structure itself. Cross-workspace party rollups for portfolio-wide concentration are on the roadmap; today the comparison is one workspace at a time.

Blind spot 3: the undocumented expense​

Company D shows $150K in Q2 expenses. You drill into the entries. Sixty percent have invoices, payment confirmations and accountant approval. Thirty percent have an invoice but no payment proof. Ten percent are bare line items: an amount and a description, nothing else.

That last 10% is $15K of unverified spend. It might be legitimate. It might be a data-entry error. It might be something worse. Without a system that tracks document completeness, nobody is flagging it.

How Moonlight helps. Every entry runs through a completeness model. The system knows which documents are required based on entry type and status, and it recalculates completeness as files are uploaded and approved. An entry missing its invoice shows a gap. A workspace with low overall completeness scores tells you the finance process needs attention before it shows up in an audit.

Blind spot 4: the stale data​

You received Company B's Q1 report in April. It is now late June. You have no idea what happened in Q2. The founder has not sent an update. You could ask, but you asked last month too and it took two weeks to get a response.

Between reports, you are operating on assumptions. Maybe expenses are stable, maybe they are not. There is no way for you to know. Your follow-on capital decisions, intros and escalations are based on data that is three months old.

How Moonlight helps. Entries appear in the workspace as they are recorded. Reports refresh as the underlying data changes. There is no reporting cycle because there is no report to prepare. Whether you check daily or once a month is your call, but the option to check is always open.

Blind spot 5: the inconsistent accounting​

Company A puts software subscriptions under "Operations". Company C calls them "Technology". Company E splits them between "R&D" and "General". When you try to compare software spend across the portfolio, you are comparing categories that do not line up.

Your master spreadsheet has a column where you reclassify these manually. It is tedious, error-prone and out of date the moment any company updates their books.

How Moonlight helps. Tags give you a consistent classification layer that sits across workspaces. Define "Software" as a tag, and any entry tagged "Software" in any workspace is comparable. Tags work alongside accounting categories, not instead of them. The company's accountant uses their own chart of accounts. You use tags for cross-entity analysis. Both can coexist.

The compound effect​

One blind spot is a nuisance. Five blind spots across five companies is a portfolio risk.

When you cannot see project-level performance, you misallocate follow-on capital. When you cannot track vendor concentration, you miss dependency risks. When you cannot verify that expenses have documents behind them, you find out about problems at audit time instead of in real time.

Blind spots rarely cause blowups on their own. They keep you from catching problems early. The cost is in the months of undetected damage before someone finally looks.

Building visibility as a practice​

Moonlight is being built as a portfolio practice rather than a one-off fix.

Each company you invest in gets a workspace. Each workspace follows the same financial structure. Each entry has a completeness model. Each workspace has reports. You hold a read-only role in every one of them.

Over time, your portfolio runs on the same visibility layer (each company keeps its own books in its own accounting system). The same questions get answered the same way across every entity. Blind spots shrink because the structure of the data stops letting them hide, even when nothing else about your analysis changes.