Software Monetization Webinar
From Subscription to Tokens (and Beyond): Designing Hybrid Software Monetization with Confidence
Forrester's Lily Varon explores how software companies are blending subscription, usage, and outcome-based models to monetize AI with confidence — and what it takes to navigate the messy middle of the quote-to-cash stack.
Original Air Date: May 18, 2025
Overview
Software monetization is in the middle of its biggest shift since the move from perpetual licenses to subscriptions — and AI is accelerating it. Forrester data shows 84% of product managers say AI is now embedded in their products, and 82% expect generative AI to have a major impact on their portfolios in the next twelve months. The challenge isn't building AI features; it's figuring out how to charge for them when usage is unpredictable, costs are volatile, and the link between a human seat and the value delivered is dissolving.
In this Software Monetization Webinar, Michael Goff, Principal Product Marketing at Revenera, sits down with Forrester Principal Analyst Lily Varon to unpack what's actually working. Drawing on fresh Forrester survey data and Revenera's Monetization Monitor research, they explore why 50% of product teams have already moved to a hybrid approach combining subscription with usage or outcomes — and why AI-native companies in particular are anchoring their strategy in consumption pricing while diversifying across cost-plus, value-based, and competitor-based models.
The session moves beyond strategy into the practical: how the "messy middle" of the quote-to-cash stack creates overlap between billing, usage data management, and entitlement management vendors, and how product leaders can decide which category actually solves their problem. Lily introduces a working framework for classifying these solutions and a companion framework — Forrester's AI maturity pricing horizon — that ties pricing flexibility to value attribution and product autonomy.
If you lead product, pricing, or monetization at a software company embedding AI, this discussion will give you a clear-eyed view of where the market is heading, why monetization is harder than it looks, and how to plan investments without pricing ahead of your measurement capabilities.
Key Takeaways
- 84% of product managers say AI is embedded into their products today. (Forrester)
- 82% expect generative AI to have a major impact on their products over the next twelve months. (Forrester)
- 50% of product teams are already monetizing GenAI using a hybrid approach — subscription combined with usage or outcomes. (Forrester)
- AI-native companies are roughly one-third more likely to use outcome-based pricing models than their non-AI peers. (Lily Varon)
- Several public software companies blew through their annual AI infrastructure budgets within the first few months of the year. (Lily Varon)
- "Don't price ahead of your measurement capabilities." (Lily Varon)
Recap
84% of Product Managers Already Have AI Embedded — and They Expect More
AI embedding is no longer aspirational. When Forrester surveyed product managers and portfolio marketers, 84% said AI is embedded into their products, and 82% expect generative AI to have a major impact on their products over the next twelve months. The number was called out as "astounding given the fact that this is a concept that wasn't really part of the mass awareness, as of three years ago." The takeaway: AI isn't a side experiment to price for later — it's already inside the portfolio, and the pricing conversation has to catch up to engineering.
Why Agentic AI Severs the Link Between Seats and Value
The move from generative to agentic AI is "simply about the autonomy with which the software can operate." In an agentic scenario, "the AI shoots off another set of code that can take action independently and autonomously" — by definition, independent of a human user. The consequence is that one human can take a lot more action inside the domain of the AI, and "there's less correlation between the human seat… and the value of the solution." That's the structural reason per-seat pricing is breaking down, and the reason AI-native companies are moving toward consumption and outcomes as the natural anchors for value.
Subscriptions Aren't Going Away — They're Being Joined by Usage and Outcomes
It's important not to overstate the shift: "Subscriptions aren't going away, but we're definitely seeing this sort of shift toward more hybrid monetization." When Forrester asked product managers how they're monetizing GenAI today, 50% said they're using a hybrid approach — subscription combined with usage, or subscription combined with outcome. There's an explicit parallel to the last big transition: subscriptions originally broke the confines of the perpetual license, and now usage and consumption are breaking the confines of a rigid, flat-rate subscription. The frame for product leaders is supplement, not replace.
AI-Native Companies Diversify Pricing — and Anchor on Consumption
When the Forrester data is cut by companies that have GenAI in all of their products — "AI companies" for shorthand — two patterns jump out. First, AI companies are roughly one-third more likely than peers to monetize on outcome-based models, often starting with binary outcomes like "was this customer service ticket resolved, yes or no?" Second, their overall pricing strategy is highly diversified — value-based, consumption, cost-plus, and competitor-based pricing all show up — but the center of gravity is consumption. It was a "surprise that cost-plus wasn't higher" given the token economics, but the takeaway holds: consumption builds the foundation for everything else, because it generates the telemetry needed to credibly attribute value.
Monetization Is Hard — and Harder for the Teams on the Frontier
A counterintuitive finding from the data: when product teams rated how difficult various tasks are, "the scale tips toward difficult or neutral," and companies monetizing on usage or outcomes were more likely than others to say monetization is hard. Company shape explains part of it — AI-native companies have a tighter aperture for value, while mature companies with diverse portfolios have to define value customer by customer ("the aperture is smaller"). And "the hard part of monetization is all the work you need to do before you are ready to implement that model" — understanding where users get value, what usage looks like, what adoption looks like. The difficulty is a signal of maturity, not a reason to retreat.
We've Been Here Before — and That's Why We Have the "Messy Middle"
No need to panic — "we've been here before. When we moved from the perpetual licenses to subscription, it broke the established operating model." Now usage and consumption are breaking the rigid flat-rate subscription model the same way. The side effect is the "messy middle" — front-office systems built for creativity (any package, any bundle, any entitlement) butting up against back-office systems that have to enforce accounting standards and compliance. Vendors are moving up and downstream into each other's territory to bridge the gap, which solves real problems but also creates a confusing landscape for buyers to navigate.
A Diagnostic Card for Cutting Through Vendor Overlap
To bring clarity to the messy middle, the session introduced a "handy dandy kind of index card" — four questions to ask any vendor in the monetization stack: What is the core definition of this category? When a vendor says they support usage-based monetization, what do they actually mean (what are the use cases)? When the solution isn't usage-friendly, what business problems surface? And when vendors do it well, what does that look like — and who is the buyer persona who has a stake in it? The diagnostic is meant to be applied to billing, usage data management, and entitlement management individually, so buyers stop comparing apples to oranges across categories.
Billing, Usage Data Management, and Entitlements Are Three Different Jobs
Three categories with overlapping vocabulary but very different jobs. Billing solutions are calculators of what is owed — rating logic is core, but enforcement isn't. Usage data management specialists provide the infrastructure to manage raw usage data: ingest it, normalize it, enrich it, apply treatments, and route it to the rest of your applications. Entitlement management — Revenera's category — is about defining and enforcing user permissions and access in accordance with the agreements that have been made. "Crucially, entitlements rules can be set and enforced" — that real-time, in-the-moment ability to allow or block access is what only entitlement solutions can do. Knowing which category solves which problem prevents stacking overlapping capabilities.
Rate Tables Are the Hottest Topic — and Usability Is Underrated
Rating is "the hottest topic at the moment" — every vendor in the monetization stack has some rate-table story. But when clients are pushed on what they actually ask about, the answer isn't sophistication, it's usability. "One of the features that was most satisfying for a group of customers was the ability to upload an Excel sheet as the first version of the rating logic. They're so used to using Excel to set that rating logic. They just want you to speak Excel so they can upload that right." The same surface needs to be configurable for guardrails, group-level visibility, and feeding usage data back into rate tables to iterate on pricing. If only a deeply technical user can modify the table, pricing agility dies in the implementation backlog.
Don't Price Ahead of Your Measurement Capabilities
The closing principle borrows from Forrester's AI maturity pricing horizon framework (developed by Lisa Singer) — a 2x2 with value attribution clarity on the y-axis and product autonomy on the x-axis. The further right and the further up you are, the broader the range of pricing models available to you. When attribution is fuzzy and autonomy is low (assistive Copilot-style features), a simple flat fee may be necessary until telemetry catches up. As attribution clarity and autonomy grow, consumption and outcome models become defensible. The takeaway, repeated as the headline of the talk: "Don't price ahead of your measurement capabilities" — base the model on what you can observe, validate, and tie back to the customer.
Speakers
Lily Varon
Principal Analyst
Forrester
Michael Goff
Principal Product Marketer
Revenera
Frequently Asked Questions
Hybrid software monetization combines two or more pricing structures — typically a subscription baseline plus usage-based or outcome-based components — to capture value that pure subscriptions can't. Where a flat subscription charges the same regardless of how the software is used, a hybrid model varies revenue with consumption, features unlocked, or outcomes delivered. Roughly half of software companies have already moved to hybrid for their AI products, because flat pricing under-captures value when usage and autonomy scale independently of seats.
Agentic AI breaks the historical link between a human seat and the value the software delivers, because autonomous agents can take action independent of any single user. As a result, per-seat or per-user pricing systematically under-captures value as autonomy grows. AI-native companies are increasingly anchoring pricing on consumption and outcomes rather than seats — pricing has to scale with what the software actually does, not how many people are logged in.
Usage and consumption-based pricing charge customers based on what they use — API calls, tokens, transactions, compute time — while outcome-based pricing charges based on a result the software delivers, such as a resolved support ticket or a closed deal. Outcome models are harder because they require clear, measurable, attributable value, which is why early movers started with binary outcomes like "ticket resolved: yes/no." Most AI-native companies blend consumption pricing as the foundation and add outcome-based components where attribution is strong enough to defend.
These are three distinct categories with overlapping vocabulary but very different jobs. Billing platforms calculate what is owed and produce accurate invoices, including against rating plans for usage. Usage data management platforms ingest, normalize, enrich, and route raw usage data across systems — they make data usable but don't typically bill. Entitlement management solutions define and enforce what each customer is allowed to access in real time, and only they can grant or block access in the moment. Most buying mistakes in the quote-to-cash "messy middle" come from confusing these categories.
Use usage-based pricing when consumption is measurable but the outcome is hard to attribute to your product alone — this fits most AI features today. Move toward outcome-based pricing when value attribution becomes clean enough to defend commercially: a specific, measurable result your software demonstrably delivered. Outcome-based pricing is increasingly positioned as the "best" tier in a good-better-best hybrid model, not a replacement for usage or subscription — it's an additional tier for customers who want to buy on results.
Start with pricing strategy, not pricing models — define what value your product delivers, how observable and attributable that value is today, and which customer segments are ready to buy on consumption or outcomes. Then instrument the product to measure that value. Only after measurement is in place should you select billing, usage data, and entitlement systems to execute the strategy. The clearest practical principle: pricing flexibility follows measurement capability — the reverse order leads to expensive rework.
Assistive AI features (Copilot-style suggestions that augment a human) typically price as part of the subscription bundle or a small per-seat uplift, because the human seat is still the unit of value. Agentic AI changes the equation: when software takes action independently, one seat can trigger far more output, and consumption or outcome-based pricing becomes the natural anchor. The further a product moves from assistive toward agentic — and from fuzzy attribution toward clear, measurable results — the more pricing flexibility opens up. Many software companies run both models in parallel: subscription for assistive features and a consumption layer for agentic ones.
Entitlement management is the layer that defines and enforces what each customer is allowed to consume in real time — making it the most direct lever for protecting margin on AI products with variable infrastructure costs. Billing tells customers what they owe after the fact; entitlements decide whether usage happens in the first place. For AI features specifically, real-time enforcement lets product teams set guardrails — per-customer token caps, tier-based throttling, feature toggles tied to plan changes — that surface runaway consumption before it shows up on the P&L. That visibility is what makes consumption pricing financially safe to scale.
Resources
In-person Event
Revenera Connect 2026: Dublin
Wednesday, June 17
Register for Revenera Connect 2026 in Dublin! The event will offer new insights and expertise into the latest monetization and pricing trends, practices and technologies and provide you with new ideas for how to make your business more successful in 2026.
Webinar
6 Steps to Launching Usage-Based Pricing for SaaS and AI
In this webinar, Nicole Segerer, General Manager at Revenera, will share this practical 6-step framework for launching usage-based pricing in SaaS, AI, and even on-premises products.
Report
AI Monetization, Done Right: A Practical Guide for Product and Pricing Leaders
This guide outlines how product and pricing leaders can achieve sustainable AI profitability by aligning pricing, usage, and cost structures.
Want to learn more?
See how Revenera's Software Monetization platform can help you take products to market fast, unlock the value of your IP and accelerate revenue growth.