The EBITDA problems that make headlines are the obvious ones: a major customer churns, a key product line loses margin, a cost structure that has outgrown what the revenue base can sustain. These are real. But they're not where most middle market value is lost.
Most EBITDA leaks are subtle. They're a hundred small inefficiencies compounding quietly over months and quarters — invisible in a monthly P&L, invisible in a quarterly board report, and often invisible to the management team itself until someone goes looking for them systematically.
Here are the five most common hidden EBITDA leaks we see across middle market portfolio companies — and what it takes to find them.
1. Vendor Contract Drift
Most middle market companies have between 50 and 300 active vendor relationships. A meaningful percentage of those relationships were negotiated years before the private equity acquisition, have never been rebid, and are priced at rates that no longer reflect the company's scale or market conditions.
Vendor contract drift is the gap between what you're paying and what you should be paying given your current volume, your current market, and your current leverage as a buyer. It's not fraud. It's not mismanagement. It's just entropy — the natural result of contracts that renew automatically and never get scrutinized.
In a typical middle market company with $50M in revenue, vendor contract drift across telecom, logistics, software, and professional services can represent $500K–$1.5M in annual EBITDA. At a 7x exit multiple, that's $3.5M–$10.5M in enterprise value sitting on the table.
How to find it: Systematic spend analytics by vendor category, benchmarked against market rates. This is an automated process — it requires data aggregation across AP systems and a benchmarking layer. And the payback is almost always immediate.
2. Manual Reporting Overhead
Ask the CFO of a typical middle market company how many hours per month the finance team spends pulling, reconciling, and formatting data for internal reports. The answer is almost always shocking — often 200–400 hours per month across the organization.
That's not a small number. At fully-loaded cost rates, 300 hours of skilled finance and operations labor is $25,000–$50,000 per month — $300,000–$600,000 annually — spent on moving numbers from one spreadsheet to another. It's not creating value. It's maintaining the appearance of visibility while consuming the resources needed to act on it.
The most expensive data is the data that takes three days to get and arrives too late to change anything.
How to find it: Time-tracking analysis of the finance and operations function for a single month. The results are rarely comfortable, but they create an immediate business case for automation and a clear ROI on operating infrastructure investment.
3. Pricing Inconsistency
Pricing inconsistency is endemic in middle market companies — and almost entirely invisible without transaction-level data analysis. It shows up as: different customers paying different prices for the same product or service, with no systematic logic behind the variation.
The cause is almost always historical. Prices were set deal by deal, relationship by relationship, over years of sales activity. Nobody ever went back and rationalized the structure. Some customers are paying 2019 prices. Some are on volume discounts that were never tied to actual volume commitments. Some are on custom contract terms that made sense at the time and remain unreviewed.
In a services business, pricing inconsistency of 8–12% across the customer base is common. In a product business, it's often higher. Closing that gap — even partially — is pure margin expansion with no additional cost.
How to find it: Transaction-level revenue analysis by customer, product, and channel. You're looking for unexplained variance in realized price for equivalent offerings. The variance is almost always there. The question is how large it is and how much is recoverable.
4. Working Capital Inefficiency
We covered working capital KPIs in an earlier post, but it bears repeating here in the context of hidden leaks: most middle market companies have 20–40 days of unnecessary working capital tied up in their balance sheets.
The math is simple. A company with $50M in revenue and a DSO of 52 days, when peers run at 38 days, has roughly $1.9M in cash that belongs in the bank, still waiting to be captured. A company that pays vendors on 25-day terms when standard terms are net-45 has voluntarily accelerated $2M+ in cash outflows. Neither of these is a strategic decision. They're operational defaults that nobody challenged.
How to find it: Benchmarking DSO, DPO, and inventory turns against industry peers and against the company's own historical performance. Working capital improvement builds cash generation directly — separate from EBITDA improvement — and reduces the cost of capital. Both matter at exit.
5. Underperforming SKUs and Service Lines Nobody Is Watching
Every middle market company has a long tail of products or services that are consuming resources — sales effort, operational capacity, management attention — and generating disproportionately little return. They're rarely killing the business. They're just quietly diluting it.
The challenge is visibility. When you're running a company with 200 SKUs or 15 service lines, the bottom quartile rarely surfaces in a standard P&L. Revenue is revenue. Cost is cost. The margin contribution at the line level gets averaged out and buried.
Rationalizing the bottom quartile — whether through repricing, sunsetting, or operational restructuring — typically recovers 1–3% of gross margin. In a $50M revenue business at a 30% gross margin, that's $150,000–$450,000 in annual EBITDA. Multiplied at exit, it's a real number.
How to find it: Gross margin analysis at the SKU or service line level, sorted by contribution margin. Most companies have never done this analysis. The first time is always revealing.
The Common Thread
Every one of these leaks has the same root cause: the absence of systematic, granular data visibility. They're not management failures. They're data failures. Nobody fixed them because nobody could see them clearly enough to act.
The companies that find and close these leaks consistently are doing something systematic. They've built an operating infrastructure that makes the invisible visible — that surfaces the variance, benchmarks the performance, and creates the data-driven conversation that leads to action.
That infrastructure is the difference between a hold period where you hunt for value and one where value creation is systematic, repeatable, and compounding.