Welcome to my session on monetizing AI for strategic growth. It is a huge honor to be part of the Women in Tech Conference.My name is Sarah Yamase, and I'm a partner at Simon Kucher and Partners and our global head of TMT. So a couple, of background points. Simon Kucher is very well known for our experience in driving growth strategy and unlocking better growth, for companies across multiple industry verticals, and across multiple sizes. I would say we're very well known for our expertise in pricing, and I will be double clicking into some of that in my session today. While I, again, am Sarah Yamase and the head of TMT here at Simon Kucher, I do have many colleagues that span different sectors, including consumer, financial services, health care, life sciences, and industrials as well. And this year, we're super excited to be celebrating our forty year, anniversary as a company.

So while, we do a lot of different topics from a growth and pricing strategy perspective, one of the very hot topics for a lot of my clients these days, is looking at generative and agentic AI solutions that they're building and figuring out how to go to market and monetizing it effectively.

We've had a decent amount of clients in this space and just showing a selection of them, but we're not just reacting to what our client specific needs, but also driving additional thought leadership, through reports, articles, partnering white papers, on our own, but with also with also with other thought leading organizations, and also driving web, webinars and podcasts, on these topics as well.

AI has been taking kind of a very, very rapid shift, in what it's able to offer us, and what it means for our own work. But especially in the tech industry, what solutions we can provide to our customers as well. So when we first started talking about, especially generative AI, I wanna say in 2022 time frame, most of what we were seeing was on the left with assistance based products, that were augmenting user productivity. Really good example of this is is, you know, Grammarly, where a a person is being helped drastically, with their writing and with their grammatical fact checking, as they are as they're developing content, themselves. But the reality is is that as you see more kind of generative, solutions come to the market and this year, AgenTic solutions come to the market, this has taken a quicker turn on what does this mean for how we go to market with our products, what do we go to market with our products, and how do we monetize these capabilities, as effectively as possible.

The the main difference I'll call out right now between generative and and agentic, is that one, generative, is very much what we've been seeing with, you know, things like chat bots or kind of workflow automations that work with us to help us do our jobs and tasks a lot faster.

Agentic takes it one step further and says, hey, you know, are there tasks that we can take away completely from, a human, from an individual user, and have that be automated through using generative technology to be able to to finish and complete a task. And I think the ones that have made the news, I guess, most rapidly, are in the customer service, customer care and customer, contact center as a service space like the Zendesk and the intercoms of the world, where chatbots are basically, interacting with customers directly rather than having them augment human capabilities.

The the interesting thing, I would say, is that the speed of with technology is being adopted, in the AI generative and agentic AI spaces. It's just increasing at an increasingly rapid pace, with Gartner predicting that 33% of, SaaS applications will include AgenTic AI in the next three years. I would imagine it's just as much even, if not even more, especially with the the rapid rate, in which customers or companies are innovating. The other thing that we were having to think through as we're setting up, you know, how executives should be thinking about this change is that it while, yes, generative and agentic AI especially are at the cutting edge of technology, from a pattern recognition standpoint, I don't think it's fundamentally different than what we've seen before.

So the shift I want to highlight, is the shift in kind of the 20 tens that we saw from on premise software to cloud based solutions. Yes, they're, the speed at which we're going from cloud to generative to agentic, I think is a lot faster. But I think the some of the key learnings and the things that you should keep in mind as you're making decisions on your AI solutions are very similar to, the transition from on premise to cloud. The main things I wanted to highlight are, you know, obviously, the amount of data control that your customers have, is fundamentally shifting. Right? The whole premise of on premise, was that they control the software and the data that's being put into it. As you get to cloud, you know, vendors managing the software solution and a lot of the data that's going into it with agentic, that's being once taken one step further where some of the data and capabilities, can actually feed back into the LLM if you don't have the appropriate guardrails in place.

But with that also comes the opportunity to shift price models to be more and more customer friendly as well. So one of the big innovations slash, you know, investors love this, of the shift from on premise to cloud was the shift from perpetual to subscription pricing, with subscription pricing being, one, a lot more affordable because it doesn't require a huge upfront investment, and, two, much more aligned with customers' needs because if they weren't getting the value out of the software that they purchased, they could cancel their subscription over time or or at any given time.

Agentic and consumption based or outcomes based models takes that one step further to align not just with, you know, your usage, or your engagement with the software over time, but how much or how little you're engaging with the software as well. There are a lot of nuances obviously to this. And I think the other thing that I wanted to highlight is that consumption and outcomes based pricing. While we're shifting in that direction, necessarily the near term end all be all, to how, you wanna go to market with these models moving forward. Okay? So key objectives for this session that I wanna highlight are, one, you know, how do you actually go to market with your AI solutions? How do you productize it? How do you package it in a way that's going to be compelling for customers? What do you position your pricing as? You know, is it full on outcomes based? Do you take a more conservative approach and and stick with something like users that's more traditional, or do you do something in between?

And how does this fit in with the rest of your go to market functions, and your ability to actually provide this solution in the market, as effectively as possible? The main things that I wanna highlight is while, yes, I'll double click a little bit into what are some of the, like, tactical things that you should be considering. As executives, you really should be focusing on kinda what are the general frameworks that you should be using to pressure test your teams on how they're actually going to market with their with your generative AI solutions, because often there are clear pitfalls that we can fall into given how how new some of this technology is.

Okay. So double clicking into AI productization and how you go to market, it's very easy for companies, and I've seen a lot of companies starting to do this, to do kind of the extremes of either I'm gonna give away my AI capabilities for free, or I'm gonna charge it separately as a separate add on, and try to monetize it as as early as possible.

And I think the reality is that often, yes, there are a lot of features and capabilities that lend themselves well to these models, but there is a good set of questions that you can ask yourself and your teams as to whether or not that's the right solution for all of your AI capabilities.

I think the reality is is that it's super important to understand how much adoption will your AI capabilities have, how much adoption do you want it to have, and how much differentiated business value is there to your to your customers. A lot of what we were seeing and trying to sort through with companies, especially in, you know, 2023 time frame, was, yes, ironically, generative AI is both the cutting edge of technology, but because it is powered by, you know, a a handful of very similar, LLMs, it's also relatively commoditized.

So someone two capabilities who are powered by, you know, GPT as their their their background model, don't really look that different to different customers. And it's really important to make sure that you're understanding and communicating to your customers what is the differentiation you're able to provide on the LLM cape or the the agentic or generative capabilities that you're offering to your customers. And, typically, that manifests itself in either additional security capabilities, additional, integration into either data or workflows, that you're then able to say, hey. You know, it's it's an incremental on on the solution I'm providing you. But this has been a question that I've been grappling with with my clients, and my on my own for, how do you really have kind of differentiated business value when 94% of, software and tech companies are saying they're gonna roll out generative or urgent AI solutions over the next couple of years?

So I spent a lot of time on the axes. I think the reality is is that, the outcomes that we see with, many companies is that they're either doing kind of their roll in with no price change or offer as an add on. But being able to roll in capabilities with an incremental price increase that is either, just built in or has kind of its own independent metric is one option to consider. The other option, which we've seen from a few companies, UiPath, but also ServiceNow, is having a preview a premium tier where it's like, hey. If you want the additional, AI capabilities, you'll upsell yourself into a higher package, which is also really nice because it's a very kind of normal buying motion and price value trade off that is not fundamentally different with how a lot of companies, consume their their SaaS solutions today.

Okay. The other things I wanted to highlight, was obviously around pricing as well. There's a lot of buzz in the market these days with some of the revolutionary price models that, companies are coming out with, especially, the intercoms and the Zendesks and the Salesforces of the world, aligning more to usage based pricing or outcomes based pricing. But I think the reality is is that considering, you know, two dimensions. One, how much is your capability or your solution augmenting human, work versus automating it? And then how close is it to kind of end and tangible value delivery, is the other dimension. So I'm a consultant. I like two by twos. We made another two by two for based on how much human impact it has and what is the tangible value delivery, what type of price model should you be considering. And while, yes, there's a general move from kind of user based pricing models to usage based or outcomes based or hybrid pricing models, the reality is is that especially in the near term, we don't have to get there immediately.

I have had multiple clients say, hey. Our customers are just so used to user based pricing. It's going to be a, a slow rev evolution to be able to change to something that includes usage or outcomes. But the reality is is that, as you get closer to Agenca capabilities, the more you'll have to consider, more either usage based or outcomes based metrics. Even within companies like Salesforce and Intercom, we see them using multiple metrics depending on the solution that it applies to. So, for example, for Salesforce, the agent force capabilities that is priced per conversation, is a is effectively helping to expand the capabilities that, an agent or a a, you know, a salesperson is doing, whereas Einstein is enhancing a user in their abilities to make decisions or drive analysis.

And so they have kind of fundamentally different price metrics. And main point here being, you know, different metrics have kind of different needs, or different companies have different needs for what their AI capabilities are doing and therefore should consider different metrics. The spectrum in which you can consider kind of usage based metrics, and I consider outcomes based pricing as a subset of usage based, to really be, you know, are do you want to make sure that your usage is aligned more with cost and therefore a proxy of value, Or do you want it to be more purely value driven where it is an outcome or, an output of the, work that is being done by the solution?

But really using these models to kinda pressure test your teams to, you know, hey. You know, we're doing something based on resource. Like, can we push ourselves a little bit more to consider what an output based metric would look like and whether or not that's acceptable in the market? The one that we've landed on with a lot of our companies, especially in kind of the medium term, is hybrid pricing models, where you're not just taking, you know, a fully usage based model, but you're pairing that with either a usage or something more stable, or traditional in terms of a pricing model to say, hey.

We may have a platform for access that is scaling by number of users, but you have kind of usage component that allows a company to, one, capture the upside of the the adoption, but, two, more so balance the shift over time, as users or kind of platform capabilities decrease and it moves over to the Agennta capabilities, you'll have a model that's able to flex with that over time.

The last thing that I wanted to highlight, because it's very clear, like, we've thought a lot about the pricing and the packaging, is that if you're going to consider a more transformational model like outcomes, or like even user based pricing, you know, this is not just a monetization strategy change.

This is an organizational change, and all of the different kind of executive stakeholders need to have, one be bought into this process, but more so helping to make sure that everything is kind of holistically aligned in order to bring, a more, I guess, innovative model to market.

This includes not just kind of the the leadership, but also their teams as well, making sure that product, the ability to deliver the models, finance, the ability to track, measure, and charge for those models, sales and marketing to be able to communicate the value, and the c suite team to be able to drive kind of the overall vision and leadership for these types of models.

It's usually important that all of these things work together because generative and agentic AI pricing is not just a pricing model change. It really is an overall business model change. So just to summarize, and to give you a couple of key questions and takeaways that you can take back with you, really when you think especially about Adjentic AI and pricing for it and go to going to market with these types of solutions, outcomes based pricing really is the North Star.

Because if you are taking, a capability and working and effectively driving that process for your customer, You know, they need to be able to see value on that. It may not be appropriate in all situations, but it is very important that if you are considering moving to those models, making sure you can define it effectively, that there's clear value to the customers, that you can work on, with your customers to automate those workflows and make sure that it's doing what it needs to be doing for customers, that sales teams can communicate both the value, and address concerns on both the product side, but also the, price model side is very important.

And that you're able to support that both with kind of the product technical capabilities on the back end, but also kind of the financial capabilities as well. Bill both billing on the back end, but forecasting upfront is critical to ensuring that this sort of model, is successful. And with that, I really appreciate you all joining my session. Hopefully, you learned a Hopefully, you learned a thing or two and, have a great time at the rest of the Women in Tech Conference. Thank you.