Most SaaS companies are missing the opportunity to monetize GenAI


Generative AI (GenAI) took us by storm in 2023, and it graduated from promising technology to mainstream innovation. OpenView Partners 2023 SaaS Benchmark Report shows 3 of 4 SaaS companies launched or are developing a GenAI offering. The interesting finding is that companies announce or launch their products before considering their long-term monetization strategy. That is surprising to me, especially as the ISV is paying for their consumption, and the consumption, for example, by using OpenAI APIs, can become expensive for the ISV if no controls are set in place.

According to a TechCrunch article, GenAI gross profit margins are, on average, 10 to 30 percent lower than those of regular SaaS solutions. If monetization is an afterthought, the solution will likely fail to achieve the company's profitability goals.

The Cost of Goods Sold (COGS) for GenAI can be significant. For example, OpenAI reportedly spends an estimated $700,000 daily to run ChatGPT. As a result, the operating costs for a GenAI solution can be high, and a monetization model needs to be built and evaluated.

According to Zuora's research, more than a third of the 70 GenAI offers assessed had a monetization model based on consumption, usage, or a hybrid model, a combination of recurring and usage charges. Linking actual customer usage with pricing logic ensures a better alignment of revenue and cost models.

In TELLUS monetization/pricing/packaging workshops, the approach we have taken for years is to focus on building the pricing based on building the value drivers for the solution to ensure that the customer feels getting the value of the solution is aligned with the value/price metric (such as price per user, price per gigabyte, etc.)

Our monetization process is based on three phases: Position, Design, and Validate. The steps within each phase are as follows:

🎯Phase 1 (Position)in the model establishes the solution's positioning and provides a solid market understanding.

🎯Phase 2 (Design)explains the solution's value drivers using the value proposition canvas framework. The framework allows us to start looking at potential solution bundles. The phase also dives into value/price metrics to understand what the user would value as a charge basis for the solution.

🎯Phase 3 (Validate) is all about building the validation process and supporting internal commercial processes to test the package/market fit by reaching out to potential customers.

Traditional SaaS pricing models are well known, and an ISV should ask whether a typical SaaS pricing model should be considered when embedding GenAI into its solution. This is what, for example, Microsoft has done with Githubt Copilot, Microsoft 365 Copilot, etc. According to some reports from the Wall Street Journal, the pricing model is not without its challenges. In the case of Microsoft, they are reportedly losing 4x as much per active user compared to normal users. As an active user of Microsoft 365 Copilot and working with customers in the Office 365 environment, I can easily see this as true.

Software vendors such as Box have decided to limit the number of GenAI queries a user can do monthly and have a "company-wide" pool of queries that can be used. This approach limits users from running roque in their usage model.

We at TELLUS have based our pricing models on a value-based pricing approach. Zuora also suggests that traditional SaaS user models are not "future-proof way for GenAI offers to cover the tech's unique cost dynamics." Furthermore, instead of "using purely cost-focused or adoption-focused strategies, companies should employ the same monetization best practices that drive recurring growth across industries: value-based monetization models."

I have worked on multiple software industry transformations, from client/server architectures to cloud transformation. I have also facilitated tens of business designs as software pricing/packaging workshops for large brand-name independent software vendors (ISVs) and startups. I claim that GenAI will have a fundamental impacton software companies, and I worry that organizations do not put enough emphasis on monetization models.

Building monetization models, in retrospect, could hurt the reputation of the ISV as they might have to go back to the pricing discussions with the customers after realizing that the company is losing money or that the profitability of the solution of the GenAI is not hitting the expected targets. I have also seen this in business model transformation initiatives when an on-premise software vendor has "cloudified" its solution only to realize that the architectural model does not support the expected price/value drivers.

How is your organization managing the monetization of GenAI if that is what you are embedding in your solution?


Dr. Petri I. Salonen

Leave a Comment