Commercial Due Diligence 2026: Bought or Earned Growth

The four questions a commercial due diligence has to answer

Every commercial diligence reduces to four questions, and a structure that misses one of them leaves the buyer exposed on that side.

1. Is the market growing, and where? By segment and by geography, because a flat market usually contains a growing pocket and a shrinking one.
2. Is the share defensible? Who the competitors are, what switching costs exist, and what the target would lose if a large customer retendered.
3. Is the revenue repeatable? Contracted, recurring, repeat or one-off, in that order of quality.
4. Is the earnings figure real? Which adjustments and add-backs survive contact with the next owner.

Question four belongs to a financial diligence team in most deals, and the commercial team still has to know the answer, because it changes the multiple being paid. Alvarez & Marsal separates the work explicitly in a single job posting: analysts get involved "in multiple financial due diligence engagements (including buy-side and sell-side work), refinancing due diligence, restructuring related work", and its transaction advisory team "can work alongside operational and commercial professionals in combined project teams", checked 30 September 2026 at careers.alvarezandmarsal.com.

The revenue build, from the bottom

A mid-market fund is looking at an industrial calibration business. It services pressure, flow and temperature instruments at customer sites. Reported revenue is 58 million dollars and adjusted EBITDA is 9.3 million.

Rebuild the revenue from drivers before accepting it. The company serves 3,100 customer sites. Each site takes an average of 2.6 calibration visits a year, so 3,100 times 2.6 is 8,060 jobs. Average job value is 7,200 dollars, so 8,060 times 7,200 is 58,032,000 dollars. The build closes on the reported figure, which means the three drivers are the right ones to interrogate.

Now split the revenue by how repeatable it is.

Sixty percent of 58.032 million is 34,819,200 dollars, thirty percent is 17,409,600 and ten percent is 5,803,200. A buyer underwrites the first line at close to face value, discounts the second by whatever retention evidence exists, and treats the third as absent from year one unless the pipeline is shown.

Separating bought growth from earned growth

The teaser says revenue grew at 7.4 percent a year for three years, against a market growing 3.1 percent. Read that way the target is taking share at more than four points a year, which is the kind of claim that supports a premium.

Work the bridge. Revenue three years ago was 58.032 million divided by 1.074 cubed. 1.074 cubed is 1.2388, so the base was 46.84 million dollars, and total growth over the period was 58.03 minus 46.84, which is 11.19 million.

Of that, three bolt-on acquisitions contributed 5.9 million dollars of revenue. Organic growth is therefore 11.19 minus 5.9, which is 5.29 million dollars, on a base of 46.84 million. That is a cumulative 11.3 percent over three years, and the cube root of 1.1129 less one is 3.63 percent a year.

So the target is outperforming its market by about half a point a year, against 4.3 points on the headline figures. Everything else about the business may be excellent and the growth story has changed shape completely: this is a roll-up with a modestly healthy core, and a buyer pays a different multiple for that than for an organic compounder.

The follow-up question writes itself. Acquisitions contributed 5.9 million of revenue over three years, so what did they cost, and is there a pipeline of similar targets left in the region. A roll-up whose acquisition pipeline is exhausted grows at 3.63 percent from the day it closes.

Quality of earnings changes the multiple

Adjusted EBITDA is 9.3 million dollars, which is 16.0 percent of revenue. The asking price is 9.0 times that, which is 83.7 million dollars.

Diligence finds that the adjusted figure adds back 0.6 million dollars of costs described as one-time: a systems migration that has now run for three consecutive years and appears in next year's budget. Costs that recur are not one-time, so the sustainable figure is 9.3 minus 0.6, which is 8.7 million dollars.

The price does not move on its own. At 83.7 million against 8.7 million of real EBITDA, the buyer is paying 83.7 divided by 8.7, which is 9.62 times. A deal approved at nine times has quietly become a deal at 9.62 times, and the change came from one line item.

Put the two findings together and the negotiating position is arithmetic rather than rhetorical. Organic growth is 3.63 percent and sustainable EBITDA is 8.7 million dollars. Holding nine times against the corrected earnings gives a price of 78.3 million dollars, which is 5.4 million below the ask.

A published diligence case, and the share it gets wrong

Oliver Wyman publishes a case in which "Your client is a private equity firm. They have reached out to Oliver Wyman to decide whether to invest in a dairy farm in Illinois", checked 30 September 2026 at oliverwyman.com. Its sample framework has three branches: farm profitability, local market dynamics, and compatibility with the client's own portfolio.

Two things are worth noticing. The third branch is about the buyer, which most candidate structures omit entirely. And the structure contains no valuation node: price appears later, under execution risks, where the firm writes that "Uncertainty about the acquisition price may result in the client overpaying for the dairy farm, reducing the expected return on investment". A candidate who follows the published structure never reaches the multiple.

The case also carries an arithmetic point worth publishing. Its question one gives farm one 10,000 cows at 100 units of milk each and farm two 5,000 cows at 80 units. The published conclusion reads: "Farm one has more than 70% of the market share in terms of the number of cows and milk production volume."

We checked both bases by hand on 30 September 2026. By milk volume, 1,000,000 of 1,400,000 units is 71.4 percent, so the statement holds. By cow count, 10,000 of 15,000 is 66.7 percent, so it does not. One threshold has been applied to two bases that do not give the same answer. Both readings are arithmetically sound and we publish them side by side rather than deciding on the firm's behalf. The practical lesson for a diligence is the same either way: name the base every share is taken on, because a buyer will.

What juniors actually do on a commercial due diligence

The clearest public description of the work comes from a job posting rather than a methodology page. CIL, a growth strategy firm whose analysts sit on this kind of engagement, publishes an analyst role whose first two duties are to "Grasp new business models quickly, typically working on a different project every 4-6 weeks" and to "Take responsibility for testing commercial hypotheses through collecting and interrogating primary and secondary research", checked 30 September 2026 at careers.cil.com.

That is the job in two lines. A commercial hypothesis, tested with primary research, on a four to six week clock. Primary research here means customer interviews: twenty to forty calls with the target's customers, its lost customers and its competitors' customers, asking who they would switch to and at what price. That firm also prints a salary to the unit and a dated deadline on each of its six analyst postings, and those deadlines run two months apart by office.

Scale matters too. Bain names three diligence service lines under its private equity practice and claims experience of "more than 22,000 diligence and 6,500 portfolio projects", checked 29 September 2026 at bain.com. The commercial one is not among the three named on that page, which is a reminder that the same word covers several different exercises depending on who is speaking.

Where diligence work is scored

The weights below are ours, read from the rubric code our own grader runs. No consulting firm publishes a marking scheme, so none of this is Oliver Wyman's reading of its own dairy case.