Pricing Shastra reposted this
AI is accelerating innovation across the Tech industry and for us, at Pricing Shastra, many of our conversations with leaders of fast-growing tech companies focus around the best ways to monetize AI driven innovation. While the spotlight so far has been on the Foundational Models, AI is also powering the applications that are building on top of the Foundational Models and requires a revaluation on everything from how AI delivers value, how it fits in with pricing structure, most effective way to monetize etc. 𝗖𝗵𝗼𝗼𝘀𝗶𝗻𝗴 𝘁𝗵𝗲 𝗿𝗶𝗴𝗵𝘁 𝗣𝗿𝗶𝗰𝗶𝗻𝗴 𝗠𝗼𝗱𝗲𝗹 → Many Gen AI products seem to be choosing the consumption based pricing models because LLMs are priced on token-based models. You do not always have to choose a consumption based pricing model for your product just because LLMs are priced that way. It depends a lot on the value of your product. → We also see many providers who believe including AI in their highest priced plans is a good way to monetize. That in our view is also an extremely simplistic way to approach AI monetization. → Here are our thoughts on when a rethink of the existing monetization model may be needed vs. when a simpler approach is more appropriate. 𝗥𝗲𝘁𝗵𝗶𝗻𝗸 𝗽𝗿𝗶𝗰𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹 𝗼𝗿 𝗠𝗼𝗻𝗲𝘁𝗶𝘇𝗲 𝗶𝗻𝗰𝗿𝗲𝗺𝗲𝗻𝘁𝗮𝗹𝗹𝘆? → AI can significantly influence product usage patterns - mostly because of elimination/automation of activities. From a pricing perspective, a key question you need to ask is “How does it impact monetization based on my current model?”. Let’s look at it for the two broad monetization models in use - User based and Usage based. → 𝗨𝘀𝗲𝗿 𝗯𝗮𝘀𝗲𝗱 𝗽𝗿𝗶𝗰𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹: →→ In user based pricing models, user count can go down with the introduction of AI. Gen AI products generally result in increased productivity, which could result in workforce reduction and fewer addressable users. In this case, you need to rethink your pricing model and meters. Example: Agent based subscriptions for Service Management software such as Zendesk, Salesforce, ServiceNow etc. →→ If however, there isn’t any significant impact to User counts because of AI, you can continue the existing model and monetize AI incrementally. Example: Office 365 subscriptions and the introduction of Office CoPilot (priced as an add-on but not changing base model). → 𝗨𝘀𝗮𝗴𝗲 𝗯𝗮𝘀𝗲𝗱 𝗽𝗿𝗶𝗰𝗶𝗻𝗴 𝗺𝗼𝗱𝗲𝗹: →→ If you currently have a usage based pricing model, u͟s͟a͟g͟e͟ ͟m͟a͟y͟ ͟g͟o͟ ͟d͟o͟w͟n͟ ͟w͟i͟t͟h͟ ͟i͟n͟t͟r͟o͟d͟u͟c͟t͟i͟o͟n͟ ͟o͟f͟ ͟A͟I͟. Example: For case based pricing for Service Management, AI can deflect case creation and reduce the number of cases. In this case, you need to rethink if the usage meter is still the right one or if additional meters need to be introduced. →→ If usage is not expected to change drastically, you can introduce AI as an additional charge or might just monetize because of additional volume generated. Facing this challenge? Connect with us to discuss more.
