No One Knows How to Price AI Tools?

This morning, I ran into this fascinating article (behind paywall) in The Wall Street Journal with the following heading: No One Knows How to Price AI Tools. Furthermore, the article states that "with businesses unwilling to drop top dollar on unproven AI tools, AI software vendors are getting creative on new strategies to convince them to buy. "It is a wild west," said one CTO.

For over two years, I have written multiple articles on this topic, emphasizing that software vendors must pay attention to the new characteristics of software pricing and packaging of AI-driven solutions. Not all the standard SaaS-based pricing rules will apply, and software vendors must take a different approach when evaluating and building pricing models for their AI solutions.

I have focused on truly understanding what AI means for organizations building AI-powered software solutions. My focus has been on?business models, especially how to price solutions with embedded AI. Here are a few of the articles I have written, not only to share information but also to learn where the market is going and identify potential best practices.

??Are AI Agents taking over our jobs?

??Will AI kill my wife's cruise business?

??Why consumption-based monetization model could benefit AI SaaS software companies?

??Software vendors are facing a new reality in software pricing when using GenAI.

??Exploring characteristics of Generative AI Solutions from a pricing perspective.

??AI Pricing Strategies for SaaS companies Offering Copilots including Microsoft.

??Are you leveled up for the AI Economy?

??What are the real upskill needs for software engineers in the AI Era?

??Scaling coaching and development to an entire organization powered by AI and LLM.

??How AI apps make money - an interesting study of 40 companies.

??How do you monetize your AI features and examples from known incumbent companies - A study of 44 companies (Part 1)?

??What frameworks can an incumbent solution vendor use to monetize AI features (Part 2)?

??Is the software industry moving away from user-based pricing to agent-based pricing?

Based on the article from the Wall Street Journal, it is obvious that there isn't a standardized model for pricing AI solutions, and companies are testing different models. I have studied what they are (see articles above). I believe it is up to each software vendor to adjust the pricing based on the characteristics of the solution they are building and the use case that the solution is deploying from an AI workload perspective.

I thought I would summarize some of the findings in the WSJ article on how different well-known software vendors are using their AI solutions. It is also clear that the recent announcement from the Chinese AI vendor DeepSeek with its R1 model will change the market dynamics regardless of what the Western solution vendor claims. It will drive down the?AI costs?over time, which benefits both software vendors building AI solutions and customers buying and consuming the solution.

According to the article, Google is moving with the following model:

Google in January said its Business Standard plan would shift from charging $12 per person per month for its Workspace productivity?suite,?plus another $20 for access to its Gemini AI business?tools,?to a $14 package with Gemini AI features baked into Workspace.

Microsoft has announced the following:

Microsoft introduced consumption-based pricing with its new Microsoft 365 Copilot Chat that gives users access to use AI agents. Depending on the interaction, customers might pay a few cents for each “use.”

Microsoft said it is seeing accelerated customer adoption of the $30 Copilot offering. However, the new Copilot Chat aims to lower the barrier to entry for new enterprises using Copilot and build a broader user base that will ultimately use the $30 per month version.

Amazon is taking an interesting approach. Enterprises can access the same underlying models they do to build similar tools as the software vendors (like replicating Copilot functionality. Amazon Web Services is betting on that strategy and doing the following:

Its Bedrock platform allows users to access models from companies like Anthropic, Meta Platforms and Mistral AI with either a no-commitment, pay-as-you-go pricing model starting at less than one cent per interaction or a time-based term commitment, starting at $25 per hour of commitment to use the Bedrock service. Although Amazon also provides its work assistant, Amazon Q, for $3 to $20 per user per month, depending on the tier.

Salesforce is taking the approach of providing flexibility when it comes to its pricing options.

Last September Salesforce rolled out a pricing plan that allowed enterprises to toggle their spend minimums between per-month licenses for human employees and consumption-based agents.?A lot of customers are still trying to make sure they have the right value equation, said Bill Patterson, executive vice president of Corporate Strategy at Salesforce, and for some of the AI investments companies have made over the last two years, the jury is still out.

Pricing an AI solution is challenging as the software vendors have a hard time estimating the underlying cost model of the use. According to the WSJ article, the OpenAI has concluded the following:

OpenAI CEO Sam Altman posted on X that the $200 per month ChatGPT Pro plan was losing money because people were using it more than anticipated. (The ChatGPT Enterprise plan is separate and typically comes in at about $30 to $45 per seat, OpenAI said).

Software companies are also facing pressure to adapt their pricing to account for the fact that the actual cost of using the underlying models is going down. As that happens, CIOs don’t want to feel like their vendors are simply taking a bigger share of the profits. All the major software cloud providers, such as Microsoft, Amazon, and Google, want their customers to commit to long-term cloud commitments, and this causes an issue with CIOs. They know that the AI cost will go down, and they have a hard time committing to something that they know will only benefit the software vendor in terms of margin.?

Every organization contemplates defining the assumed value that AI will bring when different AI technologies are applied. However, I am seeing a considerable roadblock for many software vendors purely on the educational side. The development of features and functionality has accelerated to the point that it is impossible for humans to consume all of it.

I firmly believe that the software vendors that can gradually introduce clear benefits to "regular users" will benefit the most as the acceptance will spread in a viral manner. If tools are only targeted to specialists who are techies in the first place, the broader adoption will take a long time. I work with end customer organizations, and I see what day-to-day with busy professionals is. They don't have the time to spend lots of time learning new things unless they see the real benefits in a concrete manner.

If you are a business that creates AI solutions, I would love to hear how you manage pricing and if you have identified a pricing model that works well for you.

Yours,

Dr. Petri I. Salonen

PS. If you would like to get my business model in the AI Era newsletters to your inbox on a weekly or bi-weekly basis, you can subscribe to them here on LinkedIn https://www.dhirubhai.net/newsletters/business-models-in-the-ai-era-7165724425013673985/