This article offers practical guidance for software vendors looking to successfully embed AI into their core products. We explore how to price AI features based on the value they deliver, avoid common pitfalls that slow adoption, and ultimately turn innovation into predictable revenue without adding unnecessary complexity.
Key Highlights
- AI enhancements should more often than not be folded into the core product and monetized through a clear price uplift.
- Splitting AI into a separate tier or add-on can often backfire; the “non-AI” core offering may fade in competitiveness or become obsolete.
- Common missteps, including over-indexing on cost of goods sold (COGS) and defaulting to usage-based pricing, can create unnecessary complexity without driving value capture.
The introduction of AI capabilities into software solutions has sparked debate across the market around how they should be monetized. In recent years, we have seen vendors introduce new AI tiers, experiment with usage based metering and develop multi-faceted pricing mechanisms to protect margins. Increasingly, these efforts add undue complexity without improving the customer experience or revenue/margin performance.
While AI’s impact on software is undeniable, monetizing these new capabilities doesn’t require a radical departure from existing commercial playbooks – nor is the most complex option always the best. In our experience, the simplest approach often wins. AI developments should be treated like any other area of product development and monetized accordingly.
Teneo's POV
- AI is product development, not an exception: Monetization should follow the same approach applied to any other product development – grounded in customer value, competitive dynamics and margin goals.
- In most cases, embedding functionality in the core product is the best answer: Placing AI capability in the core product and raising the price keeps the core offering competitive. Developing separate AI packages risks leaving the baseline version behind rivals that embed AI as standard.
- Let the use case determine whether a separate product/tier is necessary: AI functionality should be sold separately only when the capability performs a distinct, standalone function or represents a transformational step-change in value. Incremental enhancements typically belong in the core.
- Outcome-based pricing requires a clear, consistent, predictable metric: Hybrid or outcome-based models make sense only when AI delivers a measurable and repeatable benefit. Without a clear metric, a straightforward subscription uplift is often the better option.
Deciding Whether to Embed the AI Capability in the Existing Product or Monetize it Separately
The first step is to decide whether the AI capability should be incorporated into the core product offering or monetized as a separate product / tier. Two questions can provide a quick test:
1: Is the new capability an enhancement of the existing product functionality, or does it serve a different use case
2: Does the new capability provide a modest improvement to the value of the offering, or is it completely transformational, representing a step-change? While AI’s impact on software is undeniable, monetizing these new capabilities doesn’t require a radical departure from existing commercial playbooks – nor is the most complex option always the best.
If the capability serves the same use case and customers, while adding an important but modest / incremental level of value, it typically belongs in the core product. Businesses that choose to offer AI and non-AI editions of the core offering in this situation risk devaluing the core, falling behind competitors and failing to truly integrate AI into the core product strategy.
In some cases, it may still be appropriate to monetize AI capabilities separately, even if they represent a logical enhancement of the core offering and use case. If the additional value is so premium that it justifies a substantial step-change in price, it may make sense to treat this as a separate offering to upsell customers. Alternatively, if it’s intended for a specific customer segment (e.g. enterprise), placing it in a higher tier may be appropriate.
Treating AI as an upsell can be useful as a short-term strategy, even when the value isn’t necessarily transformative or the target customer isn’t different, to allow time to convince the market of its value without forcing an immediate migration.
In situations where the AI capability serves a different use case, treating it as a separate product may be appropriate. The value proposition and ideal customer profile (ICP) are often different. Separating it allows for a distinct monetization story and may drive more rapid adoption.
Teneo’s recent survey of 300+ software vendors found that roughly half of AI capabilities are being embedded into core functionality (the majority of which are then monetized through a price increase), versus being sold as add-ons or standalone products (56% vs. 44%, respectively).
Selecting the Right Revenue Model
Productivity gains are best captured through a price uplift
When a new AI feature makes the existing user base meaningfully more productive, increasing the core product price after AI value has been injected remains the cleanest way to monetize. A fair-usage clause can protect margin risk by capping outliers without penalizing typical adoption. An invoice line specifying the AI-related portion of the uplift helps finance teams code spend to their AI budgets without requiring additional SKUs.
Hybrid or outcome pricing fits when the AI performs measurable, uniquely attributable work
If the capability is more agentic in nature – completing distinct units of activity, either replacing human effort or introducing new functions – then outcome-based pricing may be more suitable. Outcome-based pricing requires both the vendor and customer to agree on a consistent, easily understood and predictable value metric. These models often introduce revenue variability. Many companies are embedding outcome-based pricing into hybrid models with a recurring subscription component to protect revenue stability and reflect sources of value outside the defined measurable outcome.
Common Pitfalls Vendors Make
- Creating an “AI tier:” Splitting the catalog into AI and non-AI versions can create a confused product strategy over time. AI should be central to your product offering, or at least not gated only for the highest paying customers.
- Letting cost drive the pricing strategy: Customers expect to pay for business value, not vendor infrastructure costs. Cost of Goods Sold (COGS) should be covered to protect margins but should not dictate pricing strategy. For most AI capabilities, model costs are rapidly decreasing (flagship LLM model prices have fallen 97% over the last 18 months).
- Rushing to usage-based pricing: AI usage often has a weak correlation with value. For some capabilities, the better it is, the less it needs to be used. The value of a chatbot response can be highly variable and often poorly linked to token spend. If the outcome’s value can be measured, leverage outcome-based pricing. If not, be cautious with usage-based models.
Conclusion
AI brings powerful new capabilities, but it doesn’t change the fundamentals of software pricing strategy. For the vast majority of incremental enhancements, embed the feature, raise the list price and keep the billing process simple. Isolate pricing only when the AI performs work that is clearly separate, countable and valuable. By anchoring decisions in customer value and resisting unnecessary complexity, vendors turn AI investment into predictable, sustainable growth. For help designing or refining an AI-ready pricing strategy, Teneo’s experts are ready to assist.