Skip to content

Plain-language guides to iot development

Date

Monday, November 16, 2026

Posted by

Date

IoT Business Models: Product, Subscription and Outcome-Based

iot business models

Choosing between iot business models can feel like betting your cash flow. The short answer: sell the product, charge a subscription, or price the outcome, and match the choice to your telemetry, contracts and balance sheet.

This guide sits inside the wider iot business strategy topic and focuses on how a connected product earns its money.

Item Value
Year Rolls-Royce began shifting from selling engines to selling miles flown 1962
Share of Rolls-Royce revenue from service (Forrester, Sep 2019) More than half
Rolls-Royce flying-hour revenue under long-term service agreements, 2023 £4.6bn, up 28% on 2022
Lighting electricity cut in the Schiphol Airport “Light as a Service” contract 50%
Tools under contract in Hilti’s Fleet Management programme by 2021 3 million tools
Typical transition from product sales to as-a-service One to five years

These figures come from Forrester’s IoT and as-a-service post and a Jacar write-up on industrial as-a-service. They can change, so check the sources before you rely on them.

The three iot business models at a glance

Every connected product ends up in one of three revenue patterns, and real offers often blend them. The differences come down to when you get paid and who carries the risk.

  • Product: the customer buys the device and owns it, so revenue arrives up front.
  • Subscription or equipment-as-a-service: the customer pays a recurring fee for access, use or time, and you keep ownership.
  • Outcome-based: the customer pays for a result such as light delivered or hours flown, and you carry the performance risk.

The Jacar write-up says three variants have proven traction in industry: equipment-as-a-service, outcome-based pricing and maintenance-as-a-service. You can read its summary on industrial as-a-service models.

Product sales: when selling the device still makes sense

A product model is the simplest to run, and it suits buyers who want to own and control what they install. The Jacar write-up notes that some customers prefer full control by buying, and that the as-a-service model does not suit them (source).

It also fits when you cannot yet carry the cost of keeping assets on your own books. The same source lists a lack of balance sheet to carry capex as a common reason as-a-service attempts fail.

  • Your customers want ownership and do not want a service relationship.
  • You lack reliable telemetry from the installed base.
  • Your finances cannot absorb a long payback period.

Whichever model you choose, measure it properly, because a hardware margin and a recurring revenue line need different measures. Our guide to iot kpis covers what to track.

Subscription and equipment-as-a-service

In equipment-as-a-service, the customer rents equipment per use or time, while the manufacturer retains ownership and includes maintenance (Jacar). That is the closest industrial cousin of an IoT subscription.

Forrester’s analyst argues that the shift is a positive one, and lists what it does for both sides (Forrester). It:

  • creates opportunities to build closer, deeper, stickier relationships between supplier and customer;
  • allows customers to pay only for what they use;
  • shifts money from capex to opex;
  • offloads the management of non-differentiating assets to someone else.

Scale is possible. Hilti’s Fleet Management programme passed 3 million tools under contract and 131,000 cumulative customers by 2021, according to the Jacar write-up.

How you set the fee is its own decision, and our guide to iot pricing models walks through it.

iot business models Blank board
A person holds a large blank white board in front of their upper body.

What does outcome-based pricing look like?

With outcome-based pricing, the customer pays for the result, not the product. Jacar gives examples such as lighting charged by light delivered, and logistics charged per moved unit rather than per vehicle (source).

Rolls-Royce is the long-running example. Forrester says it began moving toward selling miles flown in 1962, and that with TotalCare it earned more than half of its revenues from service as of that September 2019 post (Forrester).

The Jacar write-up reports that flying-hour revenue under Civil Aerospace’s long-term service agreements reached £4.6bn in 2023, up 28% on 2022 (Jacar). It also reports that the Philips “Light as a Service” contract at Schiphol Airport, running since 2015, cuts lighting electricity use by 50% against the earlier conventional installation.

What working models have in common

Jacar names five ingredients shared by working models (source):

  1. Deep telemetry, so you know the real state of every asset.
  2. Clear SLAs that define what counts as delivered, with measurable metrics.
  3. A structured financial model that says who pays capex and how profit is split.
  4. Aligned incentives, so you want the asset to work well, not fail.
  5. Risk assumed with skin in the game: you pay the consequences if it does not work.

Without real telemetry, the model stays in marketing, because you cannot manage what you cannot measure. The same source says OT cybersecurity of the connected asset is the technical dependency manufacturers underestimate most.

Behind those ingredients sits a stack: IIoT sensors and connectivity, digital twins, predictive maintenance, field service management, and a usage-based billing platform. Skipping that investment is why many attempts never leave the slide deck.

Where the shift goes wrong and how to start

Forrester warns that the organizational change may prove too disruptive for many suppliers. It requires different sales skills, different support skills, different product design assumptions and a very different approach to cash flow (Forrester).

Public companies may also struggle to persuade investors that the dip in revenue, as up-front payments give way to ongoing income from services, is short-term and transitional. Trust matters too: customers may suspect that a black-box algorithm recommending spare parts serves the supplier, not them, unless they can look inside.

Jacar lists these failure modes (source):

  • no real telemetry, so you cannot tell whether the asset is used as contracted;
  • SLAs that reality cannot meet;
  • a customer who simply does not want the model;
  • a weak financing model;
  • a sales team that keeps selling units, not services.

The pragmatic way in is small. Jacar’s runbook runs as follows:

  1. Pick one pilot product, not the whole line.
  2. Sign two or three early-adopter customers willing to experiment.
  3. Instrument fully: sensors, data pipeline and operational dashboards.
  4. Build an honest financial model with usage scenarios.
  5. Set up an operations team that understands service, not just manufacturing.
  6. Iterate, because the first contract will have issues.

Jacar puts the move from traditional sales at one to five years, with about one year from pilot to a second customer (source).

Frequently asked questions

Which of these iot business models should you start with?

Show answer

Start with whichever one your data can support. If you cannot yet measure asset use, a product sale is safer, and a single pilot is the sensible first step toward a service model (Jacar).

Is a subscription the same as outcome-based pricing?

Show answer

No. A subscription charges for access or time, while an outcome-based deal charges for a delivered result, so the supplier takes on more performance risk.

How long does it take to move away from selling units?

Show answer

Expect a multi-year change rather than a quick switch. Jacar’s range runs from one to five years, depending on how far the model becomes dominant (source).

Why might customers resist an as-a-service offer?

Show answer

Some want full control of what they own, and procurement teams do not always accept growing opex (Jacar). Others need transparency into how the supplier’s algorithms make decisions (Forrester).