Portfolio-Company Profitability Diagnostics for Private Equity

The value-creation clock starts at close. McKinsey's work on private-capital value creation finds that the earliest phase of ownership carries a disproportionate share of the eventual gain, which is why operating partners fight so hard to front-load the plan. Yet the analytical foundation for margin work, a customer- and product-level P&L, is usually the piece that arrives last, if at all. Teams spend the first hundred days on revenue theses and headcount while the single richest source of near-term EBITDA, the profit the company already earns and then gives away, stays invisible.

A profitability baseline fixes that in the diagnostic window. The goal is not a perfect cost model but a decision-grade one: a defensible net-margin figure for every customer, product, and channel, built fast enough to shape the value-creation plan rather than audit it after the fact. Because TDABC needs only two inputs per activity, the unit cost of supplying capacity and the time each transaction consumes, a competent team can stand up a first model on existing ERP and payroll extracts in a matter of weeks. That speed is the point: a diagnostic that lands in month one reshapes pricing and service decisions for the whole hold period, while one that lands in month nine mostly explains history.

This is where a rigorous baseline pays for itself twice, first by sizing the prize and again by de-risking it. Speed without rigour produces a number no one trusts in a board pack; rigour without speed misses the window. The discipline is to compress the timeline without hollowing out the method, and that is exactly what a purpose-built costing engine such as CostCtrl is designed to do.

The engine underneath the curve is cost-to-serve, and the method that makes it fast and defensible is time-driven activity-based costing. Robert Kaplan and Steven Anderson's TDABC assigns the real cost of the work a company does, order handling, picking, delivery, returns, credit control, field visits, support calls, to the customers and products that actually consume it. You estimate a capacity cost rate for each resource pool (its annual cost divided by its practical capacity) and a time equation for each activity, then let the transaction data do the allocation. Kaplan and Anderson recommend setting practical capacity at roughly 80% to 85% of theoretical, leaving realistic room for breaks, changeovers and downtime, so that idle and used capacity are visible rather than smeared across output.

That last point matters commercially. Traditional allocation buries unused capacity inside product cost, making healthy products look expensive and hiding the true cost of service intensity. TDABC separates the cost of the work done from the cost of capacity sitting idle, which is precisely the distinction a buyer needs to price and to right-size. The worked example below shows how two customers with identical revenue and gross margin diverge once cost-to-serve is loaded.

Same revenue, same gross margin, opposite economics. Customer B is not a bad customer, it is a mispriced and mis-served one: it needs a minimum order size, a delivery surcharge or a channel shift, not a fire sale. Multiply that logic across the book and you have the raw material for the bridge. The cost-to-serve and activity-based management disciplines describe how to turn these figures into standing operating rules rather than one-off adjustments.

The failure modes in portfolio profitability work are predictable, and nearly all of them trade rigour for the appearance of it.

Averaging the truth away. Allocating overhead as a flat percentage of revenue or headcount reproduces the blended margin and destroys the very signal you are looking for. If the method cannot make a large account look unprofitable, it cannot find the prize.

Chasing false precision. The opposite error is spending three months building a model with hundreds of drivers to answer questions that a dozen well-chosen activities would settle. In a diagnostic, decision-grade beats audit-grade. Refine after the levers are chosen, not before.

Ignoring practical capacity. Costing at theoretical capacity hides the cost of idle resources and overstates unit costs, which corrupts every downstream pricing decision. Set capacity honestly at 80% to 85% and let unused capacity show.

Confusing gross margin with contribution. A customer can carry a strong product margin and still destroy value once cost-to-serve is loaded. Any diagnostic that stops at gross margin has stopped one level too early.

Cutting the tail blindly. Unprofitable accounts are usually mispriced or mis-served, not worthless. Firing them without first testing repricing, minimum orders, or a channel shift can strip contribution that was covering fixed cost and shrink EBITDA in the wrong direction.

Leaving no owner. Findings without a bridge, and a bridge without named owners and dated quarters, quietly decay into a slide nobody executes. The management team, not the diagnostic team, has to own the lines.