Selling servitization: from capital equipment to outcome contracts

# Selling servitization: from capital equipment to outcome contracts

Kaeser Kompressoren will sell you a compressor. It will also sign a Sigma Air Utility contract in which Kaeser keeps the compressor, owns and operates the compressor room, and bills you per cubic metre of compressed air delivered at an agreed pressure and dryness. Same engineering, same corner of the plant, a completely different thing on the invoice.

Once the invoice changes, the marketing built for the machine stops working. The spec sheet described horsepower and efficiency class; the contract describes availability over ten years. The proof that closed a capital sale is not the proof that closes a supply commitment. What follows is how to rebuild the value propositionvalue propositionA clear statement of the benefits your product delivers, the problems it solves and why customers should choose you over alternatives.View full definition →, the price story and the evidence when the thing being sold is a promise rather than a product.

What "as a service" actually means in manufacturing

Servitization moves the customer from CapExCapExCapital Expenditure (CapEx) is money spent to acquire, upgrade, or extend long-lived assets like equipment, property, or software that deliver value over multiple years.View full definition → (a one-time purchase sitting on their balance sheet as an asset) to OpEx (a recurring cost). Three models, least to most ambitious:

1. Product plus service. Sell the machine, attach a maintenance contract. Most of the industry is already here.

2. Usage-based. Charge per unit consumed: cubic metres of air, kilometres run on a tyre, months a tool is on site. The asset stays on your books.

3. Outcome-based. Guarantee a result and get paid on it. Rolls-Royce set the template with power by the hour, charging airlines per engine flying hour instead of selling the turbine and then selling spares against it.

One caveat has quietly weakened the classic CapExCapExCapital Expenditure (CapEx) is money spent to acquire, upgrade, or extend long-lived assets like equipment, property, or software that deliver value over multiple years.View full definition →-to-OpEx pitch. Since IFRS 16 took effect in 2019, lessees put most leases on their own balance sheet. If your contract is a disguised lease of one named machine, the customer's controller will treat it as one and your "no asset on your books" slide dies in the finance review. The argument survives only when you genuinely control the asset, can substitute it, and sell an output. A metered air contract passes that test. A rental with a service sticker usually does not.

Why buyers say yes (and why some say no)

• No upfront capital, so cash stays available for capacity and people.
• A predictable cost per unit instead of a budget shock when a bearing seizes.
• Risk transfer: if the machine fails, that is now your problem.
• You maintain it better than they do, because you hold the sensor data and the incentive.

The objections are just as consistent:

• "Over ten years, renting costs more than buying." Often true on sticker price. Reframe on lifecycle cost, where energy dominates: electricity is the large majority of what a compressed air system costs over a decade, and compressed air alone is roughly a tenth of industrial electricity use in Europe. A contract that pays you to lower specific energy beats a cheap box.
• "I lose control of my own equipment." A real fear in critical operations, and one you answer with data access, not reassurance.
• "What if you exit the business?" A recurring contract makes you a single point of failure. Step-in rights and a pre-agreed asset transfer price answer this better than adjectives.

If a customer runs a stable, high-volume operation with cheap capital and a strong maintenance crew, buying may genuinely be smarter. Saying so is what gets you the next conversation.

Pricing the outcome

Step 1: Know your true cost to serve

Model, before you quote: depreciation over the contract life, energy, maintenance labour and parts, travel to a site you may visit forty times a year, and the expected cost of missing the target.

Michelin is the cautionary tale here. Its pay-per-kilometre offer for truck fleets, launched around 2000, ran unprofitably for several years and had to be restructured. The tyres performed. The cost of serving small fleets, and a sales force still measured on tyres shipped, had been underestimated. Cost to serve is what kills these contracts, not engineering.

Step 2: Pick a pricing metric the customer trusts

The metric must be measurable by both sides, aligned with the value the customer receives, and controllable enough that their behaviour does not bankrupt you.

The edge case that decides the whole economics: leaks. A typical plant loses somewhere between a fifth and a third of its compressed air output to leaks. Meter at the outlet of the compressor room and the customer pays for their own leaks, which gives them a reason to fix them and gives you a leak audit to sell. Promise pressure at the point of use for a flat fee, and you have just bought a distribution network you did not design and cannot see.

Step 3: Structure the tiers

• A base fee covering availability and the installed asset.
• A variable fee per unit consumed.
• A performance clause: bonuses above target, penalties below, capped.

Cap the downside and exclude the customer's consequential loss. An hour of stopped production in an automotive plant runs into six figures, more than the annual value of most air contracts. Accept liability for lost output once and a single bad quarter erases years of margin.

Messaging: sell the outcome, not the iron

Your old marketing sold horsepower, efficiency ratings and warranty length. Servitization messaging sells certainty.

| Old message (product) | New message (outcome) |

|---|---|

| "Industry-leading energy efficiency" | "Your air bill drops and stays predictable" |

| "5-year warranty" | "You never pay for a breakdown" |

| "Remote monitoring included" | "We fix problems before your line stops" |

The proof has to change with the message. Nobody audits a brochure claim about efficiency class, but a procurement team will absolutely ask what "99.5% availability" meant on your last twelve contracts, measured how, and by whom. Reference contracts with metered results, third-party verification of the meter itself, and claims documented to the standard the substantiation lesson sets out are the assets that close these deals. A testimonial is not evidence when the customer is signing away a decade.

For a primer on the broader shift, MIT Sloan Management Review's articles on servitization and outcome-based business models are a credible starting point.

🎬 [VIDEO: "What is Servitization?" - youtube.com - a short explainer on how manufacturers shift from selling products to selling outcomes and services]

De-risking the uptime guarantee

Instrument everything

You cannot guarantee what you cannot measure. Sensors stream pressure, temperature, vibration and flow to a monitoring platform, which is the base for predictive maintenance: replacing a part before it fails, on data, rather than reacting to a stopped line.

if vibration > threshold AND temperature rising:
    flag bearing wear
    schedule service before failure
    → downtime avoided, penalty avoided

That loop is what makes an outcome contract profitable instead of terrifying.

Define the guarantee precisely

Nail down what counts as downtime (planned maintenance normally excluded), the measurement window (production hours, not 24/7), excluded causes (customer power quality, misuse, force majeure), and the remedy (service credits, capped).

Do not forget volume risk

Technical risk is the one everyone models. Demand risk is the one that hurts. Rolls-Royce is paid by the flying hour, and in 2020 large engine flying hours fell to well under half of 2019 levels, turning a proven service model into a multi-billion-pound cash outflow through no fault of the engines. If your revenue moves with the customer's output, put a minimum volume, a take-or-pay floor or an indexation clause in the contract before you sign it.

Selling it internally before you sell it externally

Servitization changes your company more than your customer's, and marketing cannot carry it alone.

Revenue looks worse before it looks better. A machine sale books a large number today. A ten-year air contract books a twelfth of it a year and consumes working capitalworking capitalWorking capital is the difference between a company's current assets and current liabilities, measuring short-term liquidity and the funds available to run daily operations.View full definition → to fund the asset. Sales people paid on booked revenue will quietly steer every deal back to the transaction, so compensation changes first or the model dies.

The cycle stretches. The committee and the timeline the mapping lesson describes get longer still once a multi-year financial commitment brings the CFO into a decision the plant engineer used to own, so you need a cash-flow story and an uptime story running side by side.

Cannibalisation is real. A per-kilometre tyre contract rewards you for making the casing last and retreading it, which reduces the units the product P&L is measured on. Somebody senior has to decide that the contract margin outranks the volume line, in writing.

The back office becomes the product. Hilti's fleet management sells tools on a monthly fee over terms of roughly three to five years, covering repair, replacement and theft, across hundreds of thousands of tools. That required tool tracking, repair logistics and a claims process before it required a campaign. The relationship after delivery, and the renewal and expansion measures the installed-base lesson defines, become marketing's scoreboard rather than a post-sale afterthought.

A quick framework for your first pilot

1. Pick one product with high downtime cost and good sensor coverage. Compressors qualify.

2. Pick one friendly customer who will tolerate rough edges and let you publish the results.

3. Start usage-based, not outcome-based. Learn your cost curve before you guarantee anything.

4. Instrument heavily and log everything, including the disputes.

5. Compare actual cost to serve against your model, then reprice.

6. Write the exit clause on day one: what happens to the asset, the data and the service if either side walks.

The pilot's output is not profit. It is the evidence base that lets you make a bigger promise and defend it.

Key Takeaways

• Sell the outcome, not the machine. Buyers want air at pressure, kilometres, or tools on site. Every message reframes a feature as a worry removed.
• Cost to serve kills these contracts, not engineering. Michelin's pay-per-kilometre business needed years of repair for that reason.
• Choose the metering point deliberately. Meter at the compressor room and leaks are the customer's problem; promise pressure at the point of use and they are yours.
• Cap penalties, exclude consequential loss, and price volume risk. Rolls-Royce showed how fast a flying-hour model turns when the hours disappear.
• Fix compensation and the back office first. The model usually fails inside your own company before a customer ever objects.