In the typical portfolio company, the top 20 to 25 percent of customers produce 105 to 150 percent of total profit — which means the rest of the customer base, taken together, loses money. 80/20 segmentation is the analysis that proves this from the company’s own invoice data, sorts every customer and product into profit quartiles, and maps the business into four quadrants that each demand a different play. It is the single highest-leverage analysis a sponsor can run on a portfolio company, because it reorders the entire value creation plan around where the profit actually is.
What Does 80/20 Segmentation Actually Measure?
The analysis measures profit contribution at the level where profit is actually created: the individual customer and the individual product, and ultimately the intersection of the two. It starts from invoice-line data — every transaction over the trailing 12 to 24 months, with customer, product, volume, realized price, and cost attached — and computes true contribution for every account and every SKU. Not revenue. Not gross margin percentage. Contribution dollars, after the real costs each customer and product drags with it.
That last clause is where most in-house attempts fall short. Standard costing spreads freight, expediting, changeovers, engineering time, returns, and service intensity evenly across the book, which systematically flatters the difficult business and penalizes the easy business. Real segmentation pushes those complexity costs to the customers and products that cause them. When the smearing is undone, the picture that emerges is dramatically more concentrated than anyone in the building believes — which is precisely why the analysis has to be run on data rather than asserted from experience.
The output is deliberately simple: every customer ranked by contribution and cut into quartiles by revenue, every product likewise, and a two-by-two of the combinations. Management can argue with an opinion. It is much harder to argue with a ranked list of its own invoices.
How Do You Run a Quartile Analysis by Customer and Product?
Sort customers by revenue, largest first, and split the list into four quartiles of roughly equal revenue — not equal counts. Quartile 1 is typically a handful of large accounts; quartile 4 is usually hundreds or thousands of small ones. Then lay contribution against each quartile. Repeat the identical exercise for products. The whole build takes days, not months, given a clean transaction extract — this is the same dataset that sizes the pricing corridor and the EBITDA bridge, extracted once and used everywhere.
The canonical result looks like this: customer quartile 1 delivers the large majority of contribution; quartile 2 is solidly positive; quartile 3 hovers near breakeven; and quartile 4 — the long tail of small accounts — is outright negative once its true costs are attached. On the product side the shape repeats, often more sharply: a core catalog earning nearly all the money, and a tail of low-volume SKUs, custom variants, and legacy items that consume disproportionate operational effort per dollar of revenue.
Then comes the step that changes the room: cross the two rankings. Quartile-1 customers buying quartile-1 products is the engine of the company. Quartile-4 customers buying quartile-4 products is the anti-engine — small, complex, custom, expedited, and unprofitable at every layer. Most companies have never seen this intersection, because no standard report produces it.
What Is the Classic Finding — and How Can One Quartile Exceed 100 Percent of Profit?
The arithmetic that startles every management team: the top quartile routinely produces more than 100 percent of total company profit — 105 to 150 percent in the typical analysis we run. The excess over 100 is not a rounding artifact. It is the precise measurement of how much profit the bottom of the book destroys. If quartile 1 earns 130 percent of reported profit, then the other three quartiles combined burn 30 points of it, and the company’s reported EBITDA is the net of a great business and a bad business sharing one P&L.
Stated differently: inside the average portfolio company is a smaller, dramatically more profitable company covering the losses of a shadow company grafted onto it. A $200M revenue business earning $24M of EBITDA is often a $110M core earning $31M, financing a $90M periphery losing $7M. No diligence process that stops at company-level financials will ever see this — which is why two identical-looking targets can hold entirely different value creation potential.
This is also why profit concentration is a sponsor’s best friend rather than a risk disclosure. The concentration means the value creation plan does not require transforming the whole company. It requires protecting and growing a core that already works, and systematically fixing or exiting a periphery that demonstrably does not.
Why Do Averages Hide All of This?
Every company-level metric is a weighted average, and averages are where profit concentration goes to hide. A 34 percent blended gross margin can be — and usually is — a 52 percent core averaged against an 18 percent tail. Average selling price, average cost-to-serve, average order size: each one blends two populations that have nothing in common and reports a number that describes neither. Management teams run on averages because systems report averages, and so they make pricing, capacity, and investment decisions calibrated to a customer who does not exist.
The damage compounds through resource allocation. Sales effort, engineering hours, and service capacity get distributed roughly by revenue or by whoever asks loudest — which systematically overserves the unprofitable tail (small accounts generate the most transactions, exceptions, and noise per dollar) and underserves the profitable core (big accounts are quiet right up until a hungrier competitor calls on them). The averages conceal a daily transfer of the company’s best resources from its best customers to its worst.
Segmentation replaces the fictional average customer with the actual distribution, and decision quality changes immediately. The same leadership team, the same market, the same assets — different map, different choices.
What Are the Four Quadrants — and Why Does Each Need a Different Play?
Crossing customer size against profitability yields four quadrants, and the discipline of the method is that each quadrant gets its own play, run deliberately, rather than one blended strategy applied to all four.
- Quadrant 1 — big and profitable. The engine. Protect it, overserve it, and grow it — this is where the company’s best people and shortest lead times belong.
- Quadrant 2 — big but unprofitable. The repair shop. Large accounts losing money on bad pricing or expensive-to-serve behavior. Fix the terms; the relationships usually survive.
- Quadrant 3 — small and profitable. The seedbed. Efficiently served small accounts, some of which are tomorrow’s quadrant 1. Serve them simply and grow the best ones.
- Quadrant 4 — small and unprofitable. The drag. The long tail consuming operational capacity while destroying profit. Price it up, simplify it, or respectfully exit it.
The quadrant map also disciplines the organization chart. Most companies unconsciously organize to give every customer the same experience, which guarantees the expensive experience goes to accounts that cannot pay for it. The quadrants make service differentiation explicit and defensible: not worse service for small accounts — different service, designed to be profitable at their scale.
What Is the Play for Quadrant 1 — Big and Profitable?
Overserve, deliberately and visibly. These accounts fund the entire company, and the analysis usually shows they are being treated no better than accounts that lose money — same lead times, same service queue, same sales coverage. The play inverts that: named senior coverage, priority capacity, first access to new products, and executive relationships maintained on a cadence. The cost of overserving quadrant 1 is trivially small against the catastrophic cost of losing one of its members.
Then grow it from within and from lookalikes. Share-of-wallet analysis inside quadrant 1 accounts almost always reveals adjacent categories they buy elsewhere, and the profile of a quadrant-1 customer — size, industry, buying behavior, product fit — becomes the targeting spec for new business development. This is where the share-gain lever of the EBITDA bridge gets its named-account list: prospects that look like the engine, pursued with the capacity freed up from the tail.
One caution the data enforces: quadrant 1 is where pricing restraint belongs. The corridor work that raises prices across the loose bottom of the book should tread carefully here — these accounts often earned their pricing through volume and simplicity, and the margin they deliver comes from how cheap they are to serve, not how much they pay.
What Do You Do With Quadrant 2 — Big but Unprofitable?
Quadrant 2 is where the largest, fastest EBITDA recovery usually lives. These are substantial accounts — often top-ten customers — that lose money for reasons that are specific, diagnosable, and fixable: pricing grandfathered from a decade ago, freight terms nobody revisited, order patterns that shred production efficiency, service intensity far beyond what the contract contemplates. The account is not bad. The deal is bad.
The play is a structured commercial reset, run account by account. Bring the data: here is your realized pricing against the corridor, here is the cost your order pattern creates, here are the three paths forward — price adjustment, behavior change (consolidated orders, standard products, longer lead times), or some of each. Run with respect and evidence, the large majority of these conversations succeed, because the customer is usually receiving genuine value and paying a price set by a salesperson three organizations ago. A minority walk, and the capacity they release flows straight to quadrant 1.
Sequenced early in the hold, quadrant 2 resets often deliver one to two points of EBITDA inside the first year — the fastest lever on the bridge after list-price corridor work, and one of the most durable, because a repaired deal tends to stay repaired.
How Do You Handle Quadrant 3 — Small and Profitable?
Quadrant 3 is easy to ignore and dangerous to mishandle. These accounts are profitable precisely because they are cheap to serve — they order standard products, accept standard terms, and make no special demands. The wrong move is to treat them as a growth engine and lavish expensive coverage on them; the added cost-to-serve destroys exactly the economics that made them attractive. The right baseline play is protective efficiency: keep the service model simple, keep them digital or inside-sales covered, and do not fix what is not broken.
The selective move is graduation. Within quadrant 3 hide a handful of accounts with quadrant-1 potential — growing companies, entry purchases from large parents, new sites of existing engine customers. The play is a light screening pass to identify them, then deliberate investment in that shortlist only. This is where profitable growth is cheapest to buy, because the relationship and the economics already work; the company only needs to show up with intent.
What Do You Do With Quadrant 4 — Price Up, Simplify, or Respectfully Exit?
Quadrant 4 is where segmentation earns its reputation — and where it is most often executed badly. The tail of small, unprofitable accounts typically represents 40 to 60 percent of customer count, under 10 percent of revenue, and negative contribution in aggregate, while consuming a wildly disproportionate share of transactions, exceptions, expedites, and quality claims. The instinct to “fire the bottom quartile” with a form letter is wrong: crude exits burn reputation, panic the channel, and occasionally torch an account that is small only because it is one plant of a large parent.
The disciplined sequence has three steps, applied in order. First, price up: move quadrant 4 to list price with minimum order quantities and standard lead times — no negotiation, no exceptions. A meaningful fraction accepts, and those accounts become profitable overnight. Second, change the service model: route what remains to distribution, e-commerce, or inside sales, where the cost-to-serve matches the revenue. Third, for the residue that neither pays nor migrates, exit respectfully — generous notice, a transition path, an introduction to an alternative supplier. Done this way, the exit reads as professionalism, not abandonment.
The prize is larger than the losses recovered. The operational capacity released — production slots, engineering hours, customer-service bandwidth — is redeployed to quadrants 1 and 2, which is how segmentation converts a subtraction into growth. Companies routinely discover that removing the worst 5 percent of revenue creates the capacity that unlocks the best 15 percent of growth.
How Does Segmentation Reorder the Whole Value Creation Plan?
Run first, segmentation becomes the source document for every lever on the EBITDA bridge. Price: the corridors are widest in quadrants 2 and 4, and the analysis names the exact accounts and SKUs to reprice. Mix: the quartile data defines the shift — which segments get capacity and coverage, which get simplified out. Share: quadrant 1 lookalikes become the named-account target list. Cost: the complexity attached to quartile-4 products and customers defines what is safe to remove — and protects the quadrant-1 service spine from ever appearing on a cost-out list. M&A: targets get screened for how their book overlays the platform’s quadrant map.
It also reorders the calendar. The value creation plan stops being a portfolio of parallel initiatives competing for attention and becomes a sequence: segment in the first 30 days, reset quadrant 2 and reprice quadrant 4 in the first two quarters, shift mix and reinvest the freed capacity through year one, then grow from a repaired core. This sequencing is why segmentation belongs at the very front of the hold — every quarter it waits, the company keeps investing in the wrong 80 percent.
And it reorders the conversation with management. Before segmentation, strategy debates are contests of anecdotes. After it, the leadership team is looking at a ranked list of its own customers and products, and the debate shifts from whether to act to how fast. In our experience that shift — from opinion to evidence — is worth as much as any single lever, because it is what makes the rest of the plan executable.
An Illustrative Example With Numbers
Consider a $180M industrial distributor-manufacturer with $18M of EBITDA and 2,400 active customers. The quartile analysis lands like this: quartile 1 (26 customers, $82M revenue) contributes $23.5M — 131 percent of company profit. Quartile 2 ($46M revenue) adds $4.7M. Quartile 3 ($33M) is roughly breakeven at $0.6M. Quartile 4 (1,900+ customers, $19M revenue) loses $10.8M once freight, expediting, small-order handling, and service intensity are attached to the accounts that cause them. Reported EBITDA of $18M is the net of a $28.8M engine and a $10.8M leak.
The plan writes itself from the table. Quadrant 2 resets on 14 large money-losing accounts recover $2.6M over three quarters. Quadrant 4 pricing and service-model moves — list price plus minimums for the tail, 600 accounts migrated to distribution, roughly 500 respectfully exited — recover $4.1M of the $10.8M loss and release about 15 percent of operational capacity. That capacity plus a lookalike target list drives quadrant-1 growth worth $3.0M over 18 months. Net effect: EBITDA from $18M to roughly $27M in under two years, on slightly lower revenue — and a company that is simpler, faster, and easier to diligence at exit than the one the sponsor bought.
Every number above is illustrative, but the shape is not. It is the same shape we find in the substantial majority of industrial and industrial-service companies we analyze, at every size from $40M to well past $1B in revenue. The only genuinely open question for any given company is the exact depth of the concentration — and that question is answerable from data the company already has.
See Your Own Concentration Curve
The fastest way to test everything in this article against your own portfolio is to run one company’s numbers through our Profit Concentration Analyzer — a free tool that takes a customer-level revenue and margin extract and returns the quartile picture: how much of the profit the top of the book really produces, what the tail really costs, and how the four quadrants lay out. It takes minutes, and it turns this article’s claims into your company’s facts.
For the sponsors and operating partners who want to go deeper — how the quadrant plays get sequenced, how segmentation feeds the EBITDA bridge, and how the whole system gets installed and measured through a hold — we run a regular working Workshop that walks through the full method on real (anonymized) portfolio company data. Bring a company you are holding or underwriting. Details and the analyzer are at Private Equity — and if you only do one thing after reading this, pull the customer profitability file and look at the top quartile. The number will not be what you expect. It will be bigger.
Frequently Asked Questions
What Is 80/20 Customer and Product Segmentation?
It is a profitability analysis built from invoice-line transaction data that ranks every customer and product by true contribution — with complexity costs like freight, expediting, and service intensity assigned to the accounts that cause them — and cuts the results into revenue quartiles. The typical finding is that the top 20 to 25 percent of customers produce 105 to 150 percent of total profit, meaning the rest of the book loses money in aggregate.
How Can the Top Quartile of Customers Produce More Than 100 Percent of Profit?
Because the bottom of the book destroys profit. If the top quartile earns 130 percent of reported profit, the other quartiles combined lose 30 points of it, and reported EBITDA is the net of the two. The excess over 100 percent is a direct measurement of how much value the unprofitable tail consumes — and therefore of the recovery opportunity.
What Are the Four Quadrants in an 80/20 Analysis?
Crossing customer size against profitability yields: quadrant 1, big and profitable — protect and overserve it; quadrant 2, big but unprofitable — reset pricing and terms account by account; quadrant 3, small and profitable — serve it efficiently and graduate the best accounts; and quadrant 4, small and unprofitable — price it up, move it to a lower-cost service model, or exit it respectfully.
Should a Company Just Fire Its Unprofitable Customers?
No — crude exits burn reputation and sometimes cut accounts that are small only because they are one site of a large parent. The disciplined sequence for quadrant 4 is: first move to list price with minimum order quantities (a meaningful fraction accepts and becomes profitable), then shift remaining accounts to distribution or inside sales, and only then exit the residue with generous notice and a transition path.
Why Do Company Averages Hide Profit Concentration?
Every blended metric averages two populations with nothing in common — a high-margin core and a money-losing tail — and describes neither. A 34 percent blended margin is often a 52 percent core averaged with an 18 percent tail. Decisions calibrated to the average customer systematically overserve the unprofitable tail and underserve the profitable core, which is why the concentration only becomes visible at customer and product level.