A partner at a mid-market fund sits on five portfolio-company boards, and the EBITDA each business delivers at exit is the number their own carry is measured against. By the time the monthly board pack lands in their inbox, the operational figures inside it describe a company that existed three or four weeks ago. The median hold time for US private-equity exits ran close to six years through the first half of 2025, according to PitchBook, near the longest on record, so that same lag repeats across dozens of board cycles before the business is ever sold. Each cycle is a decision made on data that has already aged.
The blind spot sitting under the value-creation plan
Call this what it is: Data Blindness, scaled up from a single company to a whole portfolio. The operating partner carries an operational result they cannot watch as it happens, because the systems inside each portfolio company were built to run the business day to day, not to report upward on the cadence a sponsor works to. Finance closes the month, an analyst rebuilds the same spreadsheet they rebuilt last month, and the partner reads a version of events that is already weeks stale by the time it reaches the board.
A value-creation plan is a different document, and it answers a different question. We have written about which operational levers actually move the EBITDA bridge and which ones are expensive distractions in what operating partners miss in portfolio operations. This piece sits one layer underneath that plan, on a question the plan quietly assumes away: whether anyone running the company can see the levers moving week to week, or whether they find out a month later from a slide. A sound plan and a blind operator can live in the same portfolio company, and frequently do.
Why the lag costs more than it used to
Two things have made the latency more expensive than it was two cycles ago. The first is duration. With median hold times sitting near six years, close to the longest on record, more of the return has to come from operational improvement rather than multiple expansion or cheap debt, and operational improvement is precisely the part of the plan that depends on seeing the operation clearly and early. The second is scrutiny. The SEC’s 2023 private-fund adviser rules, even after a federal court vacated much of them in 2024, moved limited-partner expectations toward more frequent and more standardised reporting, and that expectation flows downhill from the fund to the portfolio companies that feed it.
An operating partner feels both pressures at the same desk. The capital is committed for longer, the reporting bar from LPs is higher, and the raw operational data still arrives through a monthly manual close that nobody designed for portfolio oversight in the first place.
There is also a bandwidth effect that compounds the problem. A partner covering five or six companies cannot personally chase down the real operational picture at each one every month, so they rely on what the board pack surfaces, and the board pack surfaces whatever was easy to assemble rather than whatever mattered most that quarter. The reporting ends up setting the agenda by default, and a blind spot in the reporting quietly becomes a blind spot in the partner’s attention.
What operational data does an operating partner actually need?
Not the finance-system month-end, which most partners already receive. The gap is in the operational layer beneath the financials: the numbers that explain why margin moved before the P&L gets around to confirming that it did. Order-to-cash cycle time, service backlog, utilisation, churn by cohort, the real cost-to-serve on the accounts that look profitable until someone counts the manual hours behind them. These are the readings that tell an operator whether a lever is working while there is still time to adjust it, rather than at the next quarterly review when the quarter is already spent.
The useful version of this data has one property the board pack usually lacks, which is that it is current enough to act on. A churn number that is six weeks old only confirms a loss, while the same number a week old is a retention call someone can still make.
The need sharpens at the two moments a hold is won or lost. In the first hundred days, the operating partner is setting a baseline and has to trust numbers from a company they have owned for only weeks, which is exactly when the underlying data is least organised. Approaching exit, those same numbers become the evidence a buyer’s diligence team will pressure-test, and a figure that took a week to assemble under current ownership will take longer still to defend under someone else’s scrutiny. Visibility that is good enough for a quiet middle of the hold is often not good enough for either end of it.
Why is portfolio reporting still built by hand?
Because the work that produces it is invisible on the org chart and no one owns the flow end to end. A finance analyst pulls figures from the ERP, someone in operations exports a report from the system that schedules the work, a commercial lead sends over a spreadsheet of pipeline, and a controller stitches all of it into the format the board sees. This is the Human API problem in a portfolio setting: people are the integration layer moving numbers between systems that were never connected, and the integration runs on their evenings at month-end.
Each of those handoffs is a place where a number can drift, and each one adds days to the age of the data. The portfolio company tolerates it because the business runs fine without the reporting being fast; the slowness only hurts the people above it, who are not in the building. So the manual close survives quarter after quarter, because the cost of it lands on the operating partner while the fix would land on the portfolio company.
What does stale operational data cost across the hold?
The direct cost is the analyst time, and it is larger than it looks. In the portfolio companies we have seen, assembling a board pack by hand consumes several finance and operations days every cycle, and across a six-year hold that is hundreds of person-days spent formatting history rather than running the business. You can put a rough number on your own version of it with the Manual Wall Calculator, which is built to price exactly this kind of recurring manual work.
The larger cost is the decisions that arrive late. When an operating partner can only see the operation a month in arrears, every corrective move is a month behind the problem, and in a plan measured against a fixed exit date, a month compounds. A pricing leak caught in week two is a memo; the same leak caught at the quarterly review has run for ninety days across the whole book. The plan rarely fails loudly in these cases; it just delivers less than it should have, and the shortfall stays invisible because no one can point to the meeting where the slow data cost the money.
Consider a single lever from a value-creation plan, say a push to lift utilisation in a services business. The plan sets the target and the operating partner expects to track it monthly. If the utilisation figure is assembled by hand from timesheets and a scheduling export, it reaches the board roughly five weeks after the period it describes. A dip that began in week one becomes visible in week nine, by which point the team has already booked two more months of work at the lower rate. The lever was sound and the target was right, and the operation still missed it, because the signal that would have triggered a correction landed after the window to correct it had closed.
How do you get current visibility without replacing the portfolio company’s systems?
The instinct is to buy a portfolio-monitoring platform and mandate that every company feed it. That is a multi-year rollout fighting the same integration problem from the top down, and it usually stalls at the portfolio company, which has no spare capacity to implement the sponsor’s reporting tool on top of its own roadmap. The faster path starts from the data that already exists and automates the stitching that people currently do by hand.
Concretely, that means mapping the handoffs that produce the board pack, then automating the ones that carry no judgment: the export and the reconciliation between systems. This is what we at Digital Forms built our Profit Leak Diagnostic to surface, the recurring manual work and the blind spots it hides, and an Operations Sprint is how the highest-value piece of it gets automated and live inside a fixed window, on the systems the company already runs. The company keeps its ERP and its scheduling tool, and what changes is that the numbers move between them without a person in the middle. Current visibility does not require a new system of record, only the handoffs between the systems already in place to stop depending on someone’s manual effort at month-end.
Who owns portfolio operational reporting?
Usually no one, which is the root of why it stays slow. The operating partner consumes the reporting but sits outside the company. The portfolio-company CFO owns the financial close but not the operational systems that feed it. The function that should own the flow of operational data, someone accountable for turning the company’s systems into something the sponsor can read on time, is the seat most mid-market portfolio companies never created. That missing seat is the CDO gap, and we have written about why AI and data mandates stall at the portfolio-company level without someone to carry them.
Filling it with a full-time hire rarely makes sense at a single mid-market company, where the work is real but not a permanent full-time load. The role tends to work better carried as a shared, outcome-accountable function across the hold, which is the model we take into portfolio companies when the data layer needs an owner rather than another tool bolted on top.
What this looks like without a named logo attached
Picture a services business inside a portfolio, around two hundred people, growing well enough that the sponsor is content and no one is looking hard at the back office. Its board pack takes a finance analyst the better part of a week every month to assemble, pulling from an ERP, a scheduling system, and a pair of spreadsheets that only one person fully understands. The operational metrics in it are a month old by the time the operating partner reads them, and the one that matters most that quarter, the cost-to-serve on the largest customer segment, is not in the pack at all, because no single system holds it.
None of this registers as a problem, because the company is growing and the reports do eventually arrive. The operating partner is making sound decisions on the information available to them. That information simply runs a month behind the business and happens to omit the number that would have changed the call. This is the pattern Data Blindness describes, and it earns less attention than it should precisely because it never announces itself.
What an operating partner can do about it
Start by pricing the current state before buying anything. Count the finance and operations hours that go into assembling the reporting each month, and list the operational metrics the board pack does not contain but a good decision would need. That inventory on its own usually moves the conversation away from “which monitoring platform” and toward “which handoffs are worth automating first,” which is a smaller and far cheaper question to answer.
From there, the sequence that works is the one we apply to any operational drag: find where the manual reporting work and the blind spots actually concentrate, then automate the highest-return handoff first on the existing systems before deciding whether a broader tool earns its place. Surfacing where the hours and the blind spots sit is what the Profit Leak Diagnostic returns, as a ranked list rather than a platform recommendation. The aim is to make the operation legible to the person accountable for it, at a speed that lets them act while the quarter is still live.
That same inventory doubles as a brief for whoever eventually owns the data layer, because it names the handoffs and the missing metrics in the company’s own terms rather than a vendor’s. It turns the first piece of automation into scoped work with a known return, not an open-ended monitoring programme that competes with the rest of the portfolio company’s roadmap.
Closing
An operating partner’s edge has always been judgment applied faster than the next fund’s. That edge is dulled by a month of latency on the one thing they are paid to watch, and no amount of operating experience closes a gap that is built into the plumbing rather than the person. The portfolio companies that compound best over a long hold tend to be the ones where the people above the business can see inside it in something close to real time. That visibility is buildable now, on the systems already in place, and the first move is simply to count what the current blindness is costing.
