More than a third (39%) of device manufacturers that have launched AI features don’t charge for them. Not a single dollar.
This stat, which I recently heard from Chris Rommel, Executive Vice President, IoT and Industrial Technology, at VDC Strategy, highlights the scope of a problem that I see on a regular basis. Device manufacturers excel at building products, but they fall short when it comes to monetizing them.
In 2026, ⅔ of devices are designed to be IoT-connected, which is double the amount from just 3 years ago, as found by VDC’s studies. This pace of innovation demands that manufacturers focus on how to close the monetization gap, particularly as they build more software and data solutions to enhance the value of the data a device creates, but that drives up cost and eats into profits.
Those who provide AI functionality face challenges capturing revenue from it. The majority of software producers already offer AI-enabled products or features, but delivery costs are eroding profitability. Many are planning new monetization models to offset cloud costs, the biggest blocker to growth.
From my experience working alongside hardware companies transitioning to digital business models, businesses must step up to monetize the software, data, and artificial intelligence in their IoT, industrial, and embedded devices. Done well, this presents a great opportunity for revenue capture. While that approach can be complicated when looked at in isolation, it usually works well when it’s part of an overall strategy to transition to recurring revenue models and digital offerings.
In the era of AI, software-led monetization of devices requires a new approach—with infrastructure to support it. Here’s how.
Understand how AI changes the market of what’s being sold.
A device is more than hardware; it’s more than the software running it. A one-and-done sale is no longer sufficient. AI products are increasingly being priced based on units of labor or productivity, as recently reported by Goldman Sachs.
This shift highlights the need for flexible pricing models, moving away from per-user or seat-based entitlements. Over the coming three years, Rommel predicts growth of hybrid monetization models that combine subscription, usage, and feature-based pricing. Our own research similarly shows that 56% expect usage-based revenue to grow by 2027 as suppliers seek to deliver flexible options to end-users.
Any company that incorporates AI functionality in smart devices must adjust their pricing and monetization initiatives to unlock the potential revenue of each device—and keep pace with new financial pressures. Even monetization models like subscription must be re-evaluated and offered alongside consumption-based models to reflect the expense of offering artificial intelligence as part of a product or feature.
Bring devices into the revenue flywheel.
There’s a chance that not charging for AI features is part of a strategic approach, such as to drive adoption. More likely, however, I believe it reflects a blindspot around usage, value, and pricing.
New approaches are absolutely necessary in order to monetize premium features and usage spikes, including in air-gapped, high-throughput environments. Bringing software-defined hardware into the revenue flywheel—with predictable, high-margin recurring revenue, across any connectivity—requires alignment of price with value, unifying entitlements as the system of record, and embracing revenue analytics for effective product decisions that support sales and renewals.
Identify where you are mid-transformation.
Most device companies are in the middle of their digital transformation. When I speak to device manufacturers, it’s clear that they need greater definition of their goals. Beyond simply shipping assets, they must identify and implement the operational infrastructure necessary to deliver effective business models for contemporary devices that function as ongoing platforms.
A manufacturer needs to be clear about how features are being used. Pricing and product decisions rely on this usage data. Clear strategy requires a robust entitlement system of record, spanning devices and on-premises and cloud deployments. Many organizations face hurdles to monetization and pricing initiatives because of back-office systems that were designed to support older, seat-based or perpetual monetization models for hardware. A robust monetization platform is essential for supporting all facets of offering hybrid monetization models, including licensing and entitlement management, usage metering, and flexible pricing tiers.
Organizational silos must be eliminated in order to align monetization strategy with growth goals. Nearly ⅘ (78%) of decisions about IoT and AI functionality are driven by engineering-led teams, as reported by Rommel. When digital transformation is initiated by product development organizations, rather than top-down from the CIO or C-suite, manufacturers must take concerted efforts to eliminate the organizational silos that may exist between engineering operations, engineering leaders, product managers, finance and pricing teams, and senior leaders.
Measure usage in order to price it.
Intelligent devices don’t stop delivering value when they ship. They provide the opportunity to deliver upgrades and data-driven services—but only if the manufacturer or software supplier has clarity into how the products are being used.
“Usage-based” is defined by various metrics and is known by various names, including consumption, metered, and pay-per-use, with credits or tokens serving as the pricing currency and units of measurement. Nearly ¾ (74%) of companies now implement usage-based monetization models (including both pre- and postpaid options) at least moderately, as reported in the Monetization Monitor.
Industry leaders, such as OpenAI’s CEO Sam Altman, are highlighting the significance of the shift toward tokens and usage-based metering for AI, in particular. In order for a company to remain profitable with such shifts, accurate pricing that’s based on usage data is necessary.
Software usage data should be used to analyze trends such as active users, device activations, feature/version usage, and data volume. Tracking usage patterns and amounts, including bursts, is necessary for pricing decisions that protect revenue margins that can be eroded by the cost of offering AI.
Turn product usage data into profit.
Once you see your usage data, how will you move with the times to monetize your software?
Software usage analytics can inform packaging, expansion, and revenue assurance (by minimizing leakage and churn), while lowering barriers to adoption.
Clarity into what’s being used and underlying costs is more important than ever. Knowing that a capability is underutilized may call attention to the need to improve functionality; knowing that it is utilized beyond what is profitable may indicate an upsell opportunity or the time to create a new pricing tier.
Intelligent device monetization in today’s market is, at least in part, a design decision. Device manufacturers who launch and support products with appropriate monetization approaches will be able to succeed. The value lifecycle of an intelligent device can go well beyond the initial sale, if properly designed, supported, and implemented.
