A quarterly report is a historical document. That is not a criticism of it — it is the job. The report exists so that a number can be fixed, agreed, reviewed and filed, and so that everyone who looks at it afterwards is looking at the same thing. Fixing a number takes time, and the time is precisely what makes the document historical.
The trouble begins when the same document is asked to do a second job it was never designed for: telling a general partner what to do next.
What the quarterly cycle was designed for
The quarterly rhythm is inherited from accounting and from the audit, and its logic is sound. A period closes. Books are ruled off. Figures are reviewed, adjusted, approved and published. The value of the output comes from its finality — from the fact that it will not quietly change next week.
Everything that makes a quarterly report trustworthy also makes it slow. It waits for the period to end, because a period has to end before it can be reported. It waits for review, because review is what turns an estimate into a record. There is no version of the audit cycle that is also a management cycle.
So a fund ends up holding a document that is authoritative and late, and then makes decisions with it that are neither.
What breaks when management runs on the record
The failure is never a reporting failure. The report is fine. What degrades is the quality of four decisions that all depend on knowing where companies stand now.
Reserves
A reserve decision allocates a finite pool across companies whose needs are moving. Doing it well means knowing which companies are burning faster than planned, which are close to a milestone that would support a higher price, and which are heading for a bridge. Those are month-level facts. Read them off a quarter-old report and you are allocating this year's capital against last quarter's needs, then learning the shape of the real needs when the next cycle lands.
Follow-ons
A follow-on has a deadline someone else controls: a round closing on the lead's timetable. The fund's own view of the company should already exist when that deadline appears, rather than being assembled in response to it. When the internal picture is stale the work gets done in a hurry — usually from the founder's narrative rather than from the fund's own record of what was promised and what was delivered.
Exit timing
Exit windows open and close for reasons outside the portfolio. Being ready to act means holding a current view of trading, cash and the ownership position. A fund that needs three weeks to reconstruct a company's numbers has effectively shortened every window it will ever be offered.
The company in trouble
The expensive case is the simplest one. A company misses plan in the second month of a quarter. The miss surfaces in the quarter-end pack, reaches the fund some weeks later, and is discussed at the following portfolio review. By then the cash position has moved on again, and the options that existed at the moment of the miss — a bridge on reasonable terms, a hiring pause, an introduction made while the company still had leverage — have narrowed to the ones nobody wanted.
The lag is not a reporting problem. It is a collection problem wearing a reporting problem's clothes.
The bottleneck is collection, not analysis
It is natural to look for the delay in analysis, and so to try to fix it with better dashboards. In most funds analysis is not the constraint. Analysis is fast once the data sits in one shape. The delay is upstream, at the point where information enters the fund at all.
Consider what actually arrives. One company sends a paragraph of prose in an email: revenue up nicely, hiring on track. Another sends a board deck as a PDF, with the numbers inside a chart rather than a table. A third sends a spreadsheet in its own layout, with revenue net of refunds this time and gross last time, and no note explaining the change. A fourth sends nothing until it has been chased twice.
Then a person reads all of it and retypes it into a tracker. That step is the bottleneck, and it has three properties that make it worse than it looks. It cannot finish until the slowest company has replied. It introduces transcription errors that are invisible, because there is nothing to check them against. And it is the only place where inconsistent definitions get reconciled, which means the definitions live in one colleague's head rather than in the record.
Nothing downstream can run faster than that step. A dashboard built on top of it is not a live view of the portfolio; it is a picture of when the retyping finished.
What changes when collection is templated and continuous
Two words carry the weight. Templated means the fund defines the fields once — each with a unit, a period and a written definition attached — and every company reports into that shape. Continuous means the request goes out on a cycle chosen for management value rather than inherited from the audit.
Get those two right and a series of things follow, none of which require anything clever:
- Fields are comparable across companies by construction, so a portfolio view is a query rather than a project.
- Variance against plan is visible while it is moving, not after it has resolved itself one way or the other.
- A restated figure is visible as a restatement, with the original preserved underneath it.
- Chasing becomes a scheduled function, not a week of somebody's month.
- The quarterly report is generated from the same fields rather than assembled from attachments, so the LP pack and the internal view cannot disagree.
The claim here is about ordering, not about products. Continuous collection is what makes portfolio monitoring a management instrument rather than a filing exercise, and the fastest wins usually come from removing human keystrokes at the boundary — accepting a structured upload, or reading straight from a company's accounting system where the company is willing to connect it. Every keystroke removed at the boundary is a day removed from the cycle.
The hard part is not technical. It is agreeing what the fields mean. Revenue, bookings, gross margin, burn and runway each have several honest definitions, and companies will use whichever their own reporting already uses. A fund that writes down one definition per field, publishes it with the request, and holds it steady for years gains something more valuable than speed: a history that can be compared with itself.
The objection that deserves an answer
Founders do not enjoy investor reporting, and the resentment is not irrational. A company with no finance function experiences a data request as an unfunded obligation. When several investors each ask for slightly different things on slightly different dates, the burden is real and it lands on the person you least want spending a day on formatting. Asking for the same information more often can make a good relationship worse. A fund that pretends otherwise will get compliance without candour, which is the worst outcome available: numbers on time and no idea what is actually happening.
The answer is not to ask less often. It is to make each ask smaller than the one it replaces.
- Ask for less. A short set of fields you genuinely use every month beats a long form that gets filled in carelessly. If nobody has looked at a field in a year, remove it.
- Ask for the same things every time. Most of the burden is not volume, it is re-interpretation — working out what this quarter's version of the question means. A stable template turns reporting into a habit rather than a task.
- Make it two minutes, not two hours. Pre-fill the previous period, accept the company's own file, and never ask for a number the fund already holds.
- Give something back. Send the company its own trend, and anonymised context on how comparable companies in the portfolio are performing. A reporting cycle the founder finds useful is one they will keep.
Handled that way, frequency stops being an imposition and becomes a shared record. The relationship improves for the same reason the data does.
The argument, briefly
Reporting and insight are not competing priorities. They are two products built from the same underlying data, and the fund only controls one thing about that data — how it arrives. A fund that changes its dashboard has improved the last mile of a slow pipeline. A fund that changes how company data enters has changed the pipeline, and the report at the end of it becomes a by-product of work that was already being done.
Nobody will call that real-time insight. They will call it knowing where things stand. It is the same thing.