AI Monetization Strategy: Proof of Value & Hybrid Pricing
- Jörn Menninger
- 10 hours ago
- 42 min read
🚀 Management Summary
How do you monetize AI without burning through cash?
That’s the question founders across Europe and the world are asking as GenAI shifts from buzzword to boardroom agenda. In this episode of Startuprad.io, Jennifer Grün, Senior Specialist for GenAI & ML at AWS, breaks down the AI monetization strategy playbook.
Her core message is simple: POC is dead. Proof of Value wins. Founders must rethink pricing (hybrid subs + credits), link ops savings to investor-grade ROI, and protect margins with careful unit economics.
📚 Table of Contents
From POC to Proof of Value
AI Monetization Strategy: Hybrid Pricing Wins
ROI Storytelling for Boards & Investors
Protecting GenAI Unit Economics at Scale
Enterprise Packaging & Segmentation
Key Takeaways
AI-Search Supreme Layer
FAQs
Closing & CTA
🚀 Meet Our Sponsor
AWS is proud to sponsor this week’s episode of Startuprad.io.
The AWS Startups team comprises former founders and CTOs, venture capitalists, angel investors, and mentors ready to help you prove what’s possible.
Since 2013, AWS has supported over 280,000 startups across the globe and provided $7Billion in credits through the AWS Activate program.
Big ideas feel at home on AWS, and with access to cutting-edge technologies like generative AI, you can quickly turn those ideas into marketable products.
Want your own AI-powered assistant? Try Amazon Q.
Want to build your own AI products? Privately customize leading foundation models on Amazon Bedrock.
Want to reduce the cost of AI workloads? AWS Trainium is the silicon you’re looking for.
Whatever your ambitions, you’ve already had the idea, now prove it’s possible on AWS.
Visit aws.amazon.com/startups to get started.
From POC to Proof of Value
Answer Capsule: POC is dead. AI products must prove value via outcome KPIs investors trust.
Why POC Doesn’t Cut It Anymore
Jennifer explains that Proof of Concept (POC) is too narrow. Investors, boards, and enterprise buyers demand Proof of Value (POV)—measurable ROI within weeks, not years. Founders need a framework like the AI Canvas: defining user outcome, cost-to-serve, and KPIs before scaling.
AI Monetization Strategy: Hybrid Pricing Wins
Answer Capsule: Hybrid pricing (subscriptions + credits) is the winning AI monetization model.
Why Hybrid Pricing Outperforms
Unlike SaaS, where per-seat subscriptions dominate, GenAI thrives on hybrid models: a subscription base plus usage-based credits. Canva, Notion, and ChatGPT have all proven the model. Jennifer highlights:
Align pricing with customer-perceived value (per task, outcome, or API call).
Prevent margin erosion by mapping credits to COGS.
Upsell with premium features: agents, analytics, compliance, enterprise integrations.
ROI Storytelling for Boards & Investors
Answer Capsule: Translate operational savings into revenue metrics boards believe: CLV, CAC, churn.
The Investor Lens
Boards don’t care if inference latency dropped 20ms—they care about churn, CLV, CAC, and ARR growth. Jennifer urges founders to translate operational efficiency into top-line outcomes. Example:
“20% faster response time” → “5% higher retention” → “$2M ARR gain.”
“Lower infra cost per query” → “30% margin lift.”
Protecting GenAI Unit Economics at Scale
Answer Capsule: Optimize infra levers (batch vs provisioned) to defend AI margins.
Scale Challenges
At 10 users, infra cost is trivial. At 10,000? It’s your margin. Jennifer shares AWS practices:
Provisioned throughput for predictable workloads.
Batch processing for non-realtime jobs.
Model right-sizing (don’t default to GPT-4 when a smaller model suffices).
Pro Tip: Instrument COGS dashboards early. Unit economics don’t fix themselves.
Enterprise Packaging & Segmentation
Answer Capsule: Enterprise buyers pay for compliance: SSO, audit logs, privacy tiers.
Segmentation Strategy
Startups win when they package for willingness-to-pay:
Free + credits → entry.
Pro tier → analytics, team features.
Enterprise → compliance (SSO, audit trails, data isolation).
Stat Spotlight: Gartner predicts 70% of enterprises will require AI compliance add-ons by 2026.
Key Takeaways
POC is dead; Proof of Value is the new standard.
Hybrid pricing beats SaaS models in AI.
Translate ops savings into investor-grade ROI.
Unit economics matter more at scale than at seed.
Enterprise compliance packaging unlocks revenue.
Founder Quote
“POC is dead. If you can’t show Proof of Value, you’re not monetizing AI—you’re running an experiment.” — Jennifer Grün, AWS
Commentary: This mindset shift is critical. Founders that align pricing and KPIs early will dominate in 2025.
Market Lens
AI adoption is entering value-extraction phase. Hype is fading, CFOs are asking: where’s the ROI? Jennifer’s frameworks signal that pricing innovation, not model choice, defines winners.
Pro Tip
When testing pricing, always A/B value metrics (per-seat vs per task). Founders are often surprised which metric resonates.
🧵 Further Reading
🎥 The Video Podcast
🎧 The Audio Podcast
Subscribe here: https://linktr.ee/startupradio
🚪 Connect with Us
Partner with us: partnerships@startuprad.io
Subscribe: https://linktr.ee/startupradio
Feedback: https://forms.gle/SrcGUpycu26fvMFE9
Follow Joe on LinkedIn: Jörn Menninger
📝 About the Author
Jörn “Joe” Menninger is the founder and host of Startuprad.io — one of Europe’s top startup podcasts. Joe's work is featured in Forbes, Tech.eu, and more. He brings 15+ years of expertise in consulting, strategy, and startup scouting.
✅ FAQs
What is Proof of Value (POV) in AI?
POV shows measurable business outcomes—revenue, retention, margin—within weeks. It replaces demo-style POCs with investor-grade KPIs and a rollout plan.
How is POV different from a POC?
POCs validate feasibility; POV validates commercial value. POV ties usage to KPIs (CLV, CAC, churn) and defines pricing, packaging, and adoption milestones.
What’s the best pricing model for AI products?
Hybrid: a subscription base for predictability plus usage-based credits for value alignment. It maps better to COGS and incentivizes efficient usage.
When should I avoid pure usage-based pricing?
Avoid when costs are volatile or value is unpredictable. Add a base subscription to stabilize revenue and protect margins during spikes.
How do I calculate credits for AI usage?
Start from cost-per-unit (inference/time/task), add target margin, then bucket into simple credit bundles (e.g., 1 credit = 1 task) with volume discounts.
Which KPIs prove AI value to boards/VCs?
Revenue lift, retention/churn deltas, CLV/CAC improvements, gross margin, and payback period. Include a before/after baseline and confidence interval.
How fast should a POV show ROI?
Aim for 30–90 days. Define leading indicators (adoption, time-to-value) and lagging metrics (retention, margin) with checkpoint reviews.
How do I translate ops savings into revenue metrics?
Map efficiency gains to outcomes: faster cycles → higher conversion or retention; lower COGS → higher gross margin; quantify impact on ARR/NRR.
What unit economics dashboards are essential?
COGS per task/query, gross margin by tier, ARPU, payback period, and contribution margin. Segment by model/runtime and enterprise vs SMB.
How can I reduce inference costs at scale?
Right-size models, cache responses, batch non-realtime jobs, use provisioned throughput for steady demand, and autoscale for bursty workloads.
When should I add an enterprise tier?
When buyers request SSO, audit logs, data isolation, SLAs, or compliance attestations. Price it by value, not just feature count.
What are common mistakes in AI pricing?
Underestimating COGS, pricing per seat like SaaS, opaque credit math, and skipping enterprise compliance—leading to churn and margin compression.
How do I pick the right value metric?
Choose the closest proxy to customer value: per task, per successful outcome, or per seat for collaboration. Validate with willingness-to-pay tests.
What’s a sane free plan for AI?
Tight caps (e.g., monthly credits), clear upgrade path, and gating expensive features (batch, analytics, export) to Pro or Enterprise.
How do I prevent abuse and runaway costs?
Rate-limit, throttle by tier, require card-on-file for heavy usage, and alert on COGS anomalies. Offer cost guards in admin settings.
Which models should I start with?
Begin with the smallest model that meets quality. Upgrade selectively for premium tasks. Continuously A/B model families vs latency/quality/COGS.
How do I migrate customers between pricing models?
Grandfather existing plans, offer upgrade credits, communicate new value clearly, and give enterprise customers a custom runway.
What’s a good payback period target?
Under 12 months for SMB/PLG and under 18 months for enterprise. Shorter is better in volatile markets.
How do I attribute AI impact across the funnel?
Use tagged cohorts, pre/post experiments, and holdouts. Tie AI-assisted events to conversion and retention deltas using multi-touch attribution.
What compliance signals increase willingness-to-pay?
Documented data handling, SOC2/ISO roadmaps, audit trails, SSO/SCIM, regional processing options, and transparent retention/deletion policies.
Give us Feedback!
Let us know who you are and what you do. Give us feedback on what we do and what we could do better. Happy to hear from each and every one of you guys out there!
The Host & Guest
The host in this interview is Jörn “Joe” Menninger, startup scout, founder, and host of Startuprad.io. And guest is Jennifer Grün, Senior Specialist for Generative AI and Machine Learning at AWS
Reach out to them:
📅 Automated Transcript
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (00:00.661)
Action.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (00:04.782)
If you're a founder, here's the challenge. Generative AI is everywhere, but turning it into actual revenue is far harder than the hype suggests. Jennifer Kroon, Senior Specialist for Generative AI at AWS has worked with startups, software vendors, and enterprises to unlock billions in business value. Today, we'll break down AI monetization frameworks, return on investment metrics that investors demand, and real startup.
case studies so you can turn AI from a password into business. Jennifer Grün is a senior specialist for generative AI and machine learning at AWS, helping startups, software vendors, and enterprises across Germany, Europe, Central Europe, why do we do this again? Jennifer Grün is a senior specialist for generative AI and machine learning at AWS, helping startups, software vendors, and enterprises across Germany,
and Central Europe to define scale and monetize their AI use cases. She led go-to-market strategies for generative AI beyond AWS. She's passionate about democratizing machine learning and runs initiatives to bring knowledge to underrepresented groups and universities. Today, she'll share what founders need to know about monetization opportunities with AI, from metrics to...
like customer acquisition costs and customer lifetime value to Canvas upsell playbook to why ROI storytelling can make or break an investor pitch. Jennifer, welcome.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (01:43.414)
Thank you, Jern. Really excited to be here today.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (01:47.407)
Totally my pleasure. We had some issues with the sub app, but now we're totally good and everything should work out. Welcome to our episode. You often describe the trillion dollar question as not what to build with AI, but how to monetize it. Why is this gap so persistent?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (02:10.294)
Thanks, Joe. So that's a really great question. So for me, the main reason for this gap comes down to three things. So first, you have the economics of Gen.ai. So unlike traditional software, where adding a new user has a very low marginal cost, generative AI is different. So every time a user interacts with the model, so every query, every word that is generated, it has a real tangible operational cost. And this also creates a paradox.
So if you have a user who gets a ton of value from your service and uses it frequently, that might actually become less profitable for you. And this makes it really difficult to achieve the high gross profit margins that investors have come to expect, especially for instance, for SaaS startups. And the second issue that I see founders are facing is often defaulting to familiar pricing models that just don't fit the technology.
So we're seeing a lot of startups try to use a per seed or per user model, but that can also backfire. So imagine if your AI product makes the team more efficient, they might actually need fewer people. And that means also fewer seeds. And this means that it would actually shrink your addressable market. And also there's a market perception issue that I can observe. So when a lot of the AI features are given away for free to drive adoption,
customers can actually start to see them as stable stakes. So this makes it really tough to convince them to pay for what they believe should be a standard feature, where startups can't afford to race to the bottom because every dollar counts in a startup. And then thirdly, many founders are so focused on the technical what building the next great model or creating a really interesting feature that they fail to properly consider the business how so.
In many of the conversations that I have with startups, I mentioned that you can't just build technology and hope for the best. So you really have to start with an outcome oriented mindset from day one. So this means understanding the specific customer pain point you're solving and how you will measure success. And a great example of this is a company that has navigated this beautifully is Lovable. So we're also going to have them soon in a hackathon here in Munich.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (04:32.67)
and they're building a platform that allows anyone to create full stack web applications just by typing in a simple prompt. So they're solving a huge problem for non-technical founders who want to build an MVP quickly without the high cost of a development team. And what they did here is that Lovable has aligned their monetization with the value that they provide. So the user credit based pricing model where users pay for what they use
which directly aligns with the operational cost and the value of the application they are generating. And this really shows again an example of a company that has successfully moved beyond the what of the technology, but also figured out the how.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (05:14.094)
I've seen many startups getting stuck in the phase for proof of concept. What do park technology choices fail to scale? Again, Madhu, many startups get stuck in proof of concept. Why do park technology choices fail to scale? And how does this kill monetization potential?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (05:41.846)
And that's an excellent question. And it's something that we see actually all the time. So a significant reason for failure is really the fundamental difference in purpose between a POC and a production ready system. So many AI pilots that I have observed are driven by a technology first mindset, where solution is developed in search of a problem to solve. So think about a cool technology that people want to build. But to really transform the organization, the product requires
buy-in from the entire organization. So we've seen startups where a lack of internal alignment on business priorities, funding or success metrics completely stalls a project. And this is why it's so important to use a structured approach. So I've worked here on an AI canvas that is forcing a shift in thinking beyond technology. So this canvas helps you to clearly articulate the customer needs, the value proposition and the metrics for success upfront.
to make sure that the solution is tied to a real business problem with a measurable outcome. And the shift from a proof of concept to a proof of value, how we call it, is really crucial in addressing this issue that I just mentioned. So the proof of value really focuses on delivering and measuring a tangible business impact, rather than really simply demonstrating technical feasibility. And what we've seen here is without a clear quantifiable metric for success,
Imagine cost savings or revenue growth, even a technically sound POC that wow is your team, we struggle to secure the continued investment required for full scale development. So this requires strong cross-functional collaboration among business, engineering and product teams to ensure alignment on a unified strategy. And also what I like to outline as well is that the technology choice is made for POC
are often not suitable for a production application. So for instance, a POC might use a popular general purpose model, but a production application requires choosing the right model that balances capabilities, performance and cost. And also it might need to be fine tuned on proprietary data to create a truly differentiated product. Also, what I observe a lot is that POCs are often designed for a few users only.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (08:04.851)
and they are really not built to handle thousands or millions of queries. So this means that they lack the necessary architecture for resilience, security and global scalabilities. And that brings me a little bit to the challenges and the startups monetization potential that is quite direct and critical. So if we think about the generative VIs space, a unique characteristic is that the cost of goods sold is often
tied directly to user consumption. So if the underlying technology is not designed for efficiency and cost optimization, every user interaction can become an expense that quickly erodes your profitability. So imagine an inefficient, unoptimized POC may function really on the small scale that we discussed, but its per unit costs can make it financially unsustainable at a larger volume. So when a startup fails to address these issues, they end up with a product that is slow,
expensive and unreliable. So their monetization potential is killed because you simply cannot charge for a service that doesn't work consistently and that can't handle a growing user base. So as a result, the startup is unable to leverage its technology for monetization through methods like premium features, tiered subscription or revenue sharing models.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (09:26.925)
I see. So what do you see? What's a common mistake that startups make when defining monetization models for AI features?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (09:38.068)
Really the most common mistake that I see startups make is creating a pricing model that doesn't align with the underlying cost and the value of AI. So many startups that I've spoken with try to use a traditional perceived model to start with, which works well for conventional software. But if we take for instance, agentic AI models, especially an autonomous one where we're developing towards, this may be completely ineffective.
So an AI agent doesn't really need a user to be active, so you can't charge per seat. So unlike traditional software with its low marginal cost, every interaction with a generative AI model has a real tangible cost. So as a result, startups that charge a flat fee for unlimited usage find themselves in a dangerous paradox. So their most engaged and arguably most valuable customers are actually making them less profitable.
because their high usage drives up operational expenses and eats into the bottom line. So this is, of course, the exact opposite of what you want in a business model. Beyond the cost paradox, another significant mistake is a failure to properly package offerings for different customer segments. So good pricing strategy isn't just about the price itself. It's really about how you differentiate your product.
For instance, you can create a basic package that offers a generic AI agent with broad expertise, and then a premium package that gives customers access to specialized agents trained on specific high value topics. Let's imagine legal or finance. And you can also package in enterprise grade features that are crucial for larger clients, but not necessarily for every user. So this could mean
single sign on audit logs and data privacy assurances in your top tier packages. So by doing this, you ensure that the customers with the highest willingness to pay are getting the most expensive product, which is a key principle of effective monetization. Another big mistake is choosing the wrong price metric. So the unit of measurement you use to charge for your product. Charging per token can be confusing and meaningless to a customer.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (11:48.35)
Instead, you can use a more value-based metric, such as charging per completed task, per generated image, or per video minute. And these metrics are more tangible and directly relate to the value the customer is receiving. So ultimately, the goal is to tie your price to the actual business outcome you deliver. Imagine the percentage of cost savings or fixed fee per support ticket resolved. However,
In the practical space, it is often too complex to attribute a specific outcome to a single AI agent. So especially if you think about multi-agent environments, so you might need to find a simpler proxy for value. And also, lastly, many startups make the mistake of not building a go-to-market strategy that can sell a new business model. So without a clear and compelling story that justifies a new pricing structure and quantifies its value, startups will struggle to get adoption and generate revenue.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (12:45.101)
I have to tell my audience you shared a lot of data, a lot of information with me for the preparation of this interview. In one of the monetization decks, you've outlined metrics like customer acquisition costs, monthly returning revenue, churn, customer lifetime value, and so on and so forth. Which are the most convincing for investors and customers?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (13:10.826)
I think what is key about this question is to remember that investors and customers are looking for fundamentally different things. So if we take an investor, the most compelling metrics are those that prove your business is a sustainable, scalable and efficient growth machine really. So they are looking for signals that the investment will lead to a predictable and high return. So if we take monthly recurring revenue or MRR for short,
That shows investors that your business has a predictable and repeatable revenue stream. So not just one of projects. And it's really the cornerstone for me of a healthy business. It demonstrates that your customers are continually willing to pay for the value you provide. Then customer lifetime value or CLV demonstrates the long term value of a customer to your business. So when an investor sees a high customer lifetime value,
they see a company with a durable competitive advantage and a clear path to long-term profitability. Also, even better, if you have a strong ratio between the lifetime value of a customer and the cost to acquire them. So this shows that your business model is fundamentally sound and that your go-to-market strategy is working efficiently. It tells an investor that you can pour money into a customer acquisition and get more back over the lifetime of these customers.
Lastly, and I find this also very important metric to consider, a low churn rate tells investors that customers are finding so much value in your solution that they are staying with you, which really reinforces a high customer lifetime value and proves that your business model is working. If we take customers, however, so they don't really care about your MRR or CLV, they really care about their own business.
So the number one thing that will convince a customer is a clear, compelling demonstration of value and a positive return on their investment. So your pitch and your decks must really focus on the tangible business benefits you deliver. So customers want to know if you're reducing their manual processes, saving them money or increasing their revenue. And these factors directly impact their bottom line and are what they truly care about. Secondly,
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (15:26.495)
Customers are convinced by the results that they can see with their own eyes. So they need to know that your product can deliver on its promises and that it is accurate, scalable and cost effective. And these factors then ultimately build their confidence in your solution and give them the evidence they need to justify the investment. Lastly, also the price metric has to be tied to the value you provide. So charging per completed task
or a direct outcome rather than per user makes the value proposition very clear to the customer. So it shows them that they are paying for solution that delivers a measurable result, but not just a tool.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (16:06.093)
I'm not sure where I've seen this, maybe even in your decks, Canva's AI feature upsells our well-known case studies. What can SAS founders learn from, for example, how Canva monetizes Gen.AI?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (16:25.301)
So the core lesson from Canva is really to avoid building a separate AI product. So instead, they embedded GenAI capabilities directly into the core offering as a value booster. So to be more specific here, Canva doesn't put all its AI features behind a paywall from the start. So they give users a limited number of credits to use features like magic eraser or magic edit for free for their presentations.
And this freemium approach allows users to experience what one source calls the AI managed jig without any upfront commitment. So this strategy is highly effective because it demonstrates the value proposition firsthand. So if you take a designer or a small business owner, if they are able to remove a background or generate an image that can save 15 to 20 minutes on a single task. And that's also what I call often the 10 times value.
So it means that for a customer to justify the cost, the risk and the effort of adopting a new solution, the value it provides has to be at least 10 times greater than the price they pay. So in this case, once the user has experienced that level of value and hits their credit limit, the decision to upgrade is easy. The cost of a paid subscription, perhaps take a few dollars a month, becomes insignificant compared to the hours of work that they have saved.
And so they're not just paying for tool, they're paying to unlock the value they've already experienced. And another critical lesson from Canva is the need to align your pricing model with the underlying costs of Gen.EI. So unlike traditional model, traditional software, every interaction with the Gen.EI model, as we said earlier, has a real tangible compute cost. So Canva's credit based system is a perfect example of a user usage based pricing.
which is then again crucial for monetizing AI and SaaS. It really directly matches the model's cost with the value delivered. And it really ensures that the compute heavy features remain a profitable part of the business as it scales on. So by charging for usage after a certain free limit, companies can manage their unit economics and ensure that the most engaged high usage customers are also their most profitable ones. So this strategy prevents the cost paradox that
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (18:47.796)
I discussed earlier where a flat fee unlimited usage model can become financially unsustainable as customer engagement increases. So this approach really offers flexibility and a lower barrier to entry for customers who want to test the product without a large upfront commitment, while also still at the same time providing a clear path to monetization for high volume users. So really a great use case.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (19:15.103)
And when listening to this, I was wondering how do you help startups connect operational savings, for example, like faster development to hard revenue metrics that resonate with, that resonates, for example, with boards or VCs.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (19:33.271)
And that's a challenge really that I see a lot of the times when talking to startups, and it's a major hurdle generally speaking on the path to production. So you can't just go into a board meeting and say, we are building things faster. So you really have to speak the language of the board, which is revenue, valuation, and return on investment. So the key here is to build a bridge between your operational metrics and your business financial health. So let's explore a three-step framework for how you can do just that.
So first, you have to get granular and define the current reality and the future reality, as you can't improve what you can't measure, obviously. So the first step here is to define the exact time or resource you're saving. For example, a startup might be able to shorten its delivery timeline from 12 months to just two months by using AI, or reduce the time for a machine learning model to go live from six months to under two weeks. So you have to define these metrics first.
whether it's lines of code written per day, bucks found per sprint, or the number of support tickets that are handled, for instance, by an AI agent. So this establishes your baseline and provides a clear point of comparison. And once you have gathered all this operational data, you also need to connect it to your finances. So this is really where you translate the what into the so what for investors. So for instance, faster development means fewer engineering hours on a project, which really translates
to reduce personal cost. And this improves your burn rate and extends your cash runway, both of which are critical metrics for VCs. And when you can develop and launch new features faster, you're also accelerating your time to market. And this means you can start generating revenue from those features sooner. And it also means you can respond to competitive threats and customer demands more quickly, which can lead to a higher customer satisfaction, reduced churn, and ultimately what we discussed earlier,
a higher customer lifetime value. And finally, you have to package the story in a way that resonates with VCs and board members. So this is where you bring the operational and financial metrics together in a single powerful statement. For example, you can say, by reducing our development cycle by 50%, we were able to launch feature X, for instance, a quarter ahead of schedule, which is already contributing, let's say,
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (21:58.423)
200K in new monthly recurring revenue. And this is a repeatable process that will allow us to continuously improve our product and our go-to-market efficiency, which then in turn will reduce our customer acquisition costs and increase our customer lifetime value. And if you follow this framework, you're not just selling a product or a technology, you're really selling a quantifiable financial improvement that directly impacts your company's valuation and growth potential.
And this is really the language that boards and investors understand. And it's the key also in terms of fundraising based on what I've seen so far.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (22:38.733)
Guys, we'll be back after a ad break talking about monetization models here.
Guys, welcome back to our interview with Jennifer from AWS. AWS works with lot of very complex customers. That's why we decided to pick a very big example. Jennifer, can you, example, share how Pfizer used Gen.AI for scientific applications to unlock something around one billion annual savings and what that teaches early stage founders about
return and investment storytelling.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (23:18.784)
That's really a fantastic example because it perfectly illustrates the power of moving beyond a simple proof of concept and into a production mindset with a clear focus on tangible business value. So the core of the Pfizer story is that they apply Geni-i to their most complex and time consuming internal processes. So let's think about scientific research and development. So if you imagine developing a single drug can generate over 20,000 documents and working through this information manually,
is an enormous drain on a scientist's time. So in the Pfizer Amazon collaboration team, or PACT for short, we implemented rapid prototyping with a fail fast cycle. So this typically takes no more than six weeks for project that would have taken months to do internally. So their first high ROI use case was data discovery. They built an internal platform called Box using generative AI
which allowed scientists to search a massive repository of documents using natural language queries. So Pfizer estimates this has the potential to save its scientists up to 16,000 hours of search time annually and reduce infrastructure costs by 55%. However, they didn't stop there with us. So they also applied machine learning and Gen.i to other critical processes like manufacturing. So by using services like Amazon SageMaker,
they developed a prototype to detect anomalies in their continuous manufacturing processes, which also helps predict maintenance needs and reduce equipment downtime. So this is really a perfect example of applying AI to a valuable and unique business process. So in this case, ensuring the quality and consistency of drug production. And especially if we tie it back to the learnings for early stage founders,
it isn't the specific technology or the scale of the savings. It's really the story that they told. So the 70, 50 million to 1 billion in annual savings, that wasn't just a number for them. So it was the result of a clear, cohesive narrative that allowed them to redirect resources, shorten development cycles, and ultimately get their products to patients and the market sooner. And to tie it back to what we said earlier, Pfizer didn't just tell its stakeholders,
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (25:41.015)
our research and development is faster. They said, because our research and development is faster, we can bring new medicine to market sooner, which will unlock this much annual revenue and cost savings. So they connected really an internal operational gain to a clear external business outcome. So for founder, this is the blueprint for a compelling pitch. So you need to show your board investors how an operational improvement
Imagine what we said earlier, the past development cycles translates into a higher customer lifetime value or lower customer acquisition costs. So by focusing your generative AI application, your most valuable and unique business processes, you create actually a snowball effect of value that becomes central to how you operate. Also, you should apply Gen.E.I. to your most valuable and unique business processes. For Pfizer, in that case, that was research and development.
For a SaaS company, it might be your product development, your go-to-market strategy, or your customer support. So really, the production mindset is about focusing on these areas to create a snowball effect of value that becomes really the fabric of how you operate.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (26:53.165)
For our audience, you've seen startups crash because they scale Gen.ai features without a monetization plan. Could you share a raw story of what went wrong? The full story is in the founders vault.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (27:08.586)
Unfortunately, due to the confidential nature of this question, can't really answer to that, but I've definitely observed a number of startups that initially started without a clear monetization plan, but we actually helped them to develop that. So for instance, we've been doing...
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (27:24.791)
Jennifer, don't need that. It's just for our audience a hook. Jennifer, let's talk a little bit about monetization models here. Do you see startups better off charging for AI as a feature, add it to an existing SaaS, or a standalone enabler or like a new product line?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (27:28.072)
Okay. Okay.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (27:47.713)
That's a really great question and a very critical one for any founder, especially as they start off in scale. So it really depends on your business, the market and the problem you're solving. So let's start with Gen.ai as a feature. That is often the safest and most effective strategy. So you should use it when you have a strong existing user base and a clear use case for AI to improve a current workflow. So instead of building a new product from scratch,
you embed AI directly into your existing offering to enhance its value. So you typically can't charge a premium for a single feature, but instead you can use tier pricing, offering AI features in higher price plans, or a freemium model where basic AI functions are free, but advanced capabilities, imagine more queries or larger outputs are meted with a credit system. So a good example is Canvas Magic Studio that we discussed before.
So it's not a new separate product. It's a suite of tools that makes the existing design process faster and better. It's also a great way to get customer feedback for iterative experimentation and continuous improvement. So the main advantage here are that it reduces the friction of adoption because customers are already in your product and the value is immediately clear and it can be a powerful upsell drive.
On the other hand, you can't charge a premium for a single feature. So your monetization has to be tied to a different value metric. Imagine a freemium model with credits or a tiered pricing structure. If you move to the second part of your question, so generative AI is a standalone enabler. It is the path often I see with startups that are building foundational technology for other developers. So you're not selling a finished product. You're selling, let's call it the picks and shovels for others.
to build their own AI applications. So this is when your core intellectual property is in the AI model or a specific repeatable AI process. So you're giving other companies the ability to add product innovation to their own platforms by integrating your API. The main advantage I see here is that this model is highly scalable. So your market could be potentially every developer or company that needs your specific AI capabilities.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (30:04.564)
And it can lead to rapid adoption and significant revenue without the overhead of building a complete end user product. The most common approach here is usage based pricing. So you can charge here per API call, per token or per unit of data processed. It's also known as the bait as you go model. So it aligns to cost directly with the value a developer gets from your technology. And this is also where you might see a hybrid model.
with a fixed subscription fee for a certain level of usage and an additional usage fee once a customer surpasses that usage. So while this model is very scalable, it faces also direct competition from big companies like OpenAI and they can have less predictable revenue streams than a flat subscription. And for me, this is a model for companies with a very unique or highly specialized AI capability.
To give you an example of a startup that I worked with over here is DeepSet. So they don't create the end user facing product, but they build the tools, infrastructure, and platforms that other companies use to create their own Gen.AI applications. So DeepSet's business model is based on an open source of freemium model, combining their open source framework haystack with a commercial SaaS platform called DeepSet AI platform.
So this platform provides enterprise-grade features and support that goes beyond the three open source version, including managed infrastructure, advanced tools, and enterprise-grade security, as well as dedicated support. And then closing the loop to the new product line. For me, this is really the most ambitious strategy. So you're building brand new end-to-end products with Gen.AI at its core. So when the AI provides a funden-
fundamentally a new experience or solves a problem that couldn't be solved before, the AI really isn't an add-on. It's the product itself. So this approach offers the highest potential for revenue and market disruption. So if you succeed, you can create a completely new market category and build a defensible model around your business. Nevertheless, it carries the highest risk as well, because you have to solve a real problem, build a product from scratch,
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (32:22.142)
and also deal with the high cost per inference and unit economics of running complex AI models. You also need to get to market before your end of runway, which is a key metric for startups. An example of a new product line in this context would be our partner Entropic. So as you might have used their cloud model as well. So rather than being an enabler for others to build applications, Entropic's business model is really centered on creating and selling the core
let's call it, going back to the analogy, picks and shovels themselves. So they create and continuously improve their co-product, in this case, the cloud family of large language models. So Antropic monetizes its models through various tiers, often different levels of intelligence, speed and cost to meet a range of user needs. So these models are offered on a tiered pricing model, often based on token usage with different plans for individuals, teams and enterprises.
As you might know, Entropic also created specialized products like Cloud Code, which is specifically designed for developers and their workflows. So this is a form of new product features that are aimed at a specific customer segment. So Entropic's strategy involves making their models available through multiple channels, including their own web interface, direct API access and partnerships with cloud providers like us.
So this ensures that their product can reach a wide range of customers from individual developers to large enterprises. So this approach has the highest potential for market disruption revenue, but it also carries the highest risk and requires significant investment in building a new product from the ground up, as well as, as I mentioned, managing that high cost per inference of running complex model. So really to answer the question, the best approach depends on your specific circumstances.
So for most startups, especially those that already have an existing customer base, adding AI as a feature is a great way to start because it offers a clear path to value realization. But it really depends where you're standing.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (34:33.741)
Do usage-based versus subscription monetization models play in AI and how do you get founders in choosing between them?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (34:45.866)
That's really for me a fundamental question that I like to take very early because it's one of the most critical ones that a founder can make. So it's not really just about a pricing model. It's a statement about your company values and how you see your relationship with your customers. So in the world of AI, this decision is even more critical because it directly ties to the unique unit economics of your product.
So what we're seeing in the market is that successful markets are moving beyond the traditional flat fee subscription and even the pure usage based model to embrace a hybrid approach. So this gives you the best of both worlds. So you can start with a predictable stable subscription that includes a certain amount of the eye credits or usage as we've seen in the Canva example. And then once a customer becomes a powerful user and exceeds that included amount,
you automatically move them to a usage-based model. And this really ensures that you're compensated for the value you provide. And it also helps manage the high variable cost of your AI workload, as we've just seen in what I shared earlier. And this hybrid approach also opens up a number of specific monetization avenues for founders that have Gen.EI as the end product, where you can, for instance, charge per image generated or per query answered.
And on the other hand side, Gen.ai is a super tier means that for an existing product, you can create a higher price subscription tier that offers exclusive access to your advanced AI capabilities. And this really is helpful for your most engaged users to give them a clear path to get more value. And you also get to charge a premium for it, just like we discussed in the Canva example. Another possibility is Gen.ai as a value booster. So in this model, the AI functionality enhances an existing product.
but it's not the central feature. So the eye adds value to the core service, which makes it more efficient or powerful, which can justify a higher overall price point. A good example, which many startups are using is Slack, for instance. And lastly, Gen.ai is an add-on, means that the eye feature is an optional component that customers can pay for separately. So this lowers the barrier to entry for your core product, while giving you still an opportunity to upsell and monetize
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (37:07.254)
those who want to use your AI capabilities. But for me, it's also important apart from the monetization strategy to look at the infrastructure supporting it. So really it's about managing the variable cost of the AI workloads. So when a founder is building their platform, they have to choose a lot of different things. So for instance, the right inference consumption strategy, which directly impacts their ability to scale profitably.
And there many options here. So for instance, there's the pay as you go model with no commitment. It's great for prototyping proof of concepts and small workloads because it's so flexible. But it can mean that you have higher latency because you're using shared resources. If you want to go for consistent production workloads with a predictable need for large scale processing, go for provision throughput. It's redesigned to process big workloads at scale.
but it can also be expensive if your usage has low utilization during off peak hours. And then another one that I see a lot of startups use is batch inference. So it's really great for asynchronous workloads, like imagine processing large documents or conducting offline experiments. It's a lot more cost effective, up to 50 % cheaper than on demand, but it's not of course suitable for every application that need real time responses. So really it depends.
As always, for me, the successful founders are really the ones who think about this from both a business and a technical perspective. So they don't just pick a pricing model. So they align it with a smart, cost-effective infrastructure strategy, which allows their business to scale profitably as their customers find more value in their product.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (38:55.469)
We know startups work a lot with examples, with tests, with their own setup. So how important is pricing experimentation? Can Gen.ai itself help optimize pricing points for AI products based on promotion data?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (39:18.058)
Yeah, you hit it here on a really great concept that I quite often discuss now with founders. So really as a founder, your job is never truly done when it comes to pricing. So there's always a need for iterative experimentation, how we call it. So it's about being outcome oriented and not being afraid to act and test potential solutions, even if they are not perfect. So this is particularly true for generative AI because the value proposition
and the customer willingness to pay can be unclear at the beginning. So you have to constantly learn and adapt based on how people actually use your product, what your customers truly value, and how they're willing to pay for it. And for me, this is where Gen.Eye can be a good sparing partner and a game changer for pricing itself. So we just discussed how to price an E.I. product, but Gen.Eye can also be used as a powerful tool to determine the optimal pricing.
So imagine if you have Gen.Eye models that can process and analyze vast amount of data from your promotions, user interactions, and even market trends. So it can define and identify patterns that a human analyst might miss. So let's think about a specific type of user response to a discount or how a certain feature usage correlates with a particular pricing tier. So if you learn from past promotion data and user behavior,
The AI can build predictive models. So you can ask it, for instance, to forecast if we want a 20 % off promotion for our super tier, what will be the impact on our monthly recurring revenue and customer churn? It can then provide data-driven insights to help you make a more informed decision. Lastly, GenAI can enable really a dynamic pricing model that automatically adjusts in real time based on demand, usage, and even individual user profiles.
So, for example, if a user is a heavy power user, which is who is perhaps on the bridge of exceeding their included credits, the AI could serve them a personalized offer to upgrade to the next tier. And this means that you can capture more value while providing the user with a predictable bill. So really, to sum this up, instead of just using spreadsheets and intuition, founders can leverage any AI to become more scientific and automated in their pricing.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (41:42.528)
So it's really a shift from a static pricing page to a living, breathing pricing engine that's constantly learning, adapting and optimizing for both customer value and business profitability.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (41:55.085)
Now that we talked so much about monetizing, let's go a little bit into risk and barriers. In the past, we had a lot of startups and subject matter experts talking about data quality that needs to go into models. What hurdles around data quality and use case alignment most often block the monetization?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (42:21.172)
And this is a problem I observe a lot of the times. And it's also a ties back to what we discussed earlier, why promising proof of concepts failed to scale. So a POC is often built for curated success, how we call it. So it really operating on a small, clean data set in an isolated environment. So, but when you really try to bring it to a production ready state, you of course also encountered the messy fragmented data of the real world.
And the challenge here is that Gen.ai models require high quality, up-to-date data to perform effectively. So if your data is incomplete, outdated, or unstructured, the garbage in, garbage out principle applies. So an AI solution built on a weak data foundation will produce inaccurate, unreliable, or nonsensical outputs. So this really directly kills monetization. So as we discussed, a customer will only pay for a product that delivers a tangible return.
So if your AI isn't producing consistent high quality results, it can't deliver on its promises. So no matter how elegant your pricing model is, a product that doesn't work is impossible to monetize. And secondly, and perhaps the most interesting hurdle is the lack of clear use case. So many startups fall into the trap of being a solution in search of a problem. So they develop a technically impressive AI model, but fail to connect it to real quantifiable business need.
This is why together with a group of colleagues, we launched an EMEA-wide Gen.EI Launchpad for about 200 customers in five cities in Europe, where we worked with each startup prior to the event to refine their use case, align architecture, and ensure data readiness. So as part of the event series, we also worked on the ICanvas I created, which gets customers to define measurable KPIs to track Gen.EI and data-driven impact.
So the program success came from pushing customers to dive deeper into the business metrics, which helped them select the right use case from the start. So really the AI canvas to elaborate starts by asking question around the customer's pain point and the negative consequence on the productivity or revenue. So the value added by your project and what's the impact of not doing it. And lastly, very important, how do you measure success?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (44:44.223)
And when a founder fails to clearly define these elements, their project becomes a costly experiment with no clear path to revenue. So really, this is a job number zero without a tangible business problem to solve. The solution can't be tied to a measurable outcome, and therefore it can't be monetized. And in the end, these hurdles that I just mentioned are very linked. So a great use case on its own is just an idea. So it really needs a foundation of good data to become a reality.
And GoodRater on its own is also just a resource. So it needs a clear use case to unlock its value. So for an AI solution to be monetized, it must consistently deliver on a clear, compelling value proposition. And it really can only do so if it's built on the right data and solves a real business problem.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (45:33.869)
You've warned about risks like cross-tenant attacks in SaaS AI products. How should founders communicate risks transparently without scaring away their investors?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (45:49.621)
And it's a really interesting one because first of all, when I talk to investors as well, they are not really looking for risk free company. They know also that those don't really exist. They're really looking for founders who are aware of the risks and have a clear, well thought out plan to manage them. So by presenting your security strategy with confidence and specific details, you're not scaring them off. And sent you really building trust and showing them that you have thought about this. And also you are thinking about
building a successful, resilient business. So the first step is to communicate that security isn't an afterthought. It's really a foundational principle to your business. So this means that you've baked in security measures from the very beginning, rather than patching them on at a later stage. And this is really key for gaining investor trust. So instead of talking about the risk of a cross-tenant attack as a scary possibility, frame it as a known challenge that you are actively solving.
So to also recap for people that don't know about cross-tenant attacks, that means that there's a malicious actor gaining unauthorized access to a single instance that it uses to breach other customers' data. And it is a well-known vulnerability in a multi-tenant SaaS system. And this is just my kind of recommendation, how you can communicate your solution in a confident, transparent way.
So this could be explaining how you've implemented strong guardrails to ensure that customer data is secure and private, or how each customer's data is stored in its own dedicated encrypted database or schema. So this is a crucial technical defense against cross-tenant attacks. So mention that you've implemented robust identity and access management policies to ensure that one customer's credentials can't be used to access another's.
Talk for instance also about your robust monitoring and alerting systems that don't just tell you when something has failed, but also provide early warnings of potential issues. So this really demonstrates to your investor a productive and proactive approach rather than a reactive way and thinking towards security. So you can show also that your applications are designed to gracefully degrade rather than completely fail. So if one service becomes unavailable,
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (48:10.485)
your system will continuously operate with reduced functionality instead of shutting down entirely. So by presenting your security strategy in this way, you're really not scaring investors. Instead, you're building trust and showing that you have the foresight and technical capabilities to build a successful, resilient business. So this moves the conversation from what if this happens to what makes your security plan better than the competition. So this is a lot more powerful.
when you are in the fundraising discussions, for instance.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (48:45.1)
Let's talk after so much input and we may add for our audience that we are already recording here for more than 45 minutes. Let's go a little bit into the outlook. Which industry verticals do you expect to see the fastest return on investment from Gen.AI, SaaS, FinTech Healthcare or actually others?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (49:09.367)
So based on what I've observed, so the quickest ROI tends to come from operational improvements rather than really high risk product innovations. So this is where companies can automate internal processes, reduce manual labor and drive immediate efficiency gains. So industries that have a high density of these types of tasks are more likely to win early. So SaaS and FinTech are at the top of the list for rapid GNI return on investment.
The primary reason for this is that these industries have a massive volume of customer interactions and data, much of which can't be automated. For instance, startups can deploy AI agents to handle common customer inquiries. This reduces the need for human agents on the front line. This directly translates into lower operational cost and faster response time, which provides ultimately a clear and quantifiable ROI. FinTech in particular,
has many high volume repetitive tasks in thinking about verification, data mapping for B2B clients. And Gen.AI can automate these workflows, really improving the time to value for customers and freeing up human employees for more strategic work. SaaS companies can also use Gen.AI to create personalized brand complying content at scale, think about marketing emails to sales collateral. So this really speeds up go to market efforts and reduce the cost of content creation.
I also am a strong believer of healthcare and life science industry because they have a compelling case for rapid ROIs we just discussed in the Pfizer example, because they have the potential to automate the administrative and research related tasks. So Geni-i can automate the generation of medical reports, summarize patient records for doctors and even streamline billing and coding processes.
So these three sub-doctors and nurses to focus on patient care, which is a significant value driver. And also as we've seen before in drug discovery and clinical trials, GenAI can analyze vast amounts of data to identify new compounds or predict the yield of new medicine. So as we've seen in the Pfizer example, they have focused their GenAI efforts on these scientific applications and predicted annual cost savings of hundreds of millions of dollars.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (51:33.025)
So the overall takeaway here is that while the product innovation customer phasing features offer the promise of high returns, they often come with higher risk. So if we contrast this with internal operation improvements, they offer a lower risk path to a medium risk path, but they often also have a faster return.
So for founders, ultimately the key is to prioritize use cases with a clear link between a genuine solution and a measurable business outcome, as we said earlier, so this could be cost reduction or efficiency gains. So you can really demonstrate a clear ROI to stakeholders and build momentum for even more ambitious projects down the road.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (52:18.092)
Do you believe in the next five years we'll see AI first companies valued differently by investors from traditional SaaS companies? We've seen this, for example, with with remote first with cloud first and so on and so forth. Do you see that also playing out in the future?
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (52:42.935)
Absolutely, really, when looking at the current state of the market. So I believe that over the next five years, we are very certain to see AI-first companies that are valued differently by investors than traditional SaaS firms, for instance. So let's break that down a bit more. So for me, an AI-first company is a business whose core product, internal operations, and competitive advantage really depend on artificial intelligence. So this is often agentic AI,
which means software not just performing really narrow tasks, but autonomously making decisions, for instance, setting goals and refining actions across complex multi-step processes. So agentic AI is a step beyond classic automation. We're talking about systems that can analyze problems, coordinate with other AIs and adapt over time. So essentially managing entire workflows without constant human input. So this is also what the research shows around this.
So if you think about investors, major VCs and enterprise focused funds, they are already allocating significant capital to agentic AI startups. So according to DROOM, there more than 755 agentic AI startups as of mid of this year, with combined funding that is topping 10 billion for categories like coding agents, business process automation, health agents and agent building frameworks. Gardner predicts
that why the current environment is experimental with limited production examples, agent AI projects will both attract acquisitions from legacy automation manufacturers and also face challenges, including also high cancellation rates for projects that don't meet real autonomy needs. And we're really seeing a pivot here. So if we think about these deterministic domains like finance or logistics, enterprises chase a lot of these proven automation solutions
and my also acquire agentic AI startups as an addition. And we see this also, or what I see here happening is to have this for creative or knowledge based workflows. So agentic AI is opening doors for brand new players. So these are the companies really investors are keeping a close look on, because they're not just replacing old tech, they're really redefining what's possible. And I'm working actually with different startups that are using
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (55:07.839)
Gentic AI for workflow automation, data science and cybersecurity, where nearly half of them are AI first. So they're building solutions like autonomous AI, employees, decision making executives and multi agent platforms that handle everything from debt collection to healthcare automation. And if we think about SaaS businesses, they still deliver value by digitizing and streamlining manual processes.
So they're offering software plus a service. But agentic AI firms really promise scalable autonomy. So the ability to grow and adapt without requiring a customer to upsize their human team. So investors see a lot of growth potential here. So imagine a startup that is serving thousands of customers with only a handful of staff and a suite of autonomous AI's running the show. So it really changes a lot how value is created.
But also I want to outline a little bit the challenges. So reliability, hallucinations, context limitations, as we also mentioned earlier, these all pose a big hurdle. So it is really important to have evaluation oversized frameworks in place, which are really for me timeless principles that are still high in demand and will remain that way. So let's also acknowledge this. So I see a lot of agent washing also happening.
labeling really simple automations as agentic AI for hype, really savvy investors are probing deeper here to really find true autonomy and scalability. So if we think also about numbers that Gardner is publishing here, more than 40 % of agentic projects will be canceled because many use cases don't actually need genuine creative autonomy. So often partial autonomy, not really full independence is actually the way to go.
So on the flip side, those who get the tech at this stage, and that's our mostly startups with really true agentic capabilities, will actually receive good valuations and get a lot of interest. So there's a high probability that EI first companies, especially those that are generally agentic, will be valued differently by investors than traditional SaaS firms. So the premium relic will come from this autonomous scaling, new margin structures,
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (57:30.151)
And yeah, investors will really look for those where AI is not just a feature, but the engine powering growth and innovation. But also there will be experimentation insurance. So as mentioned, some projects and even startups will not survive this hype cycle. So for me, the winners will be those who combine deep agentic autonomy with robust evaluation, security and operational infrastructure.
Jörn ‘Joe’ Menninger | Founder and Editor in Chief | Startuprad.io (57:54.156)
Mm I have to admit I have to smile when you talk about mislabeling of gigantic AI. We've seen this with AI in general, we've seen this with green text, so it kind of a repeating pattern. Jennifer, thank you very much for being a guest. We will have you back for a second episode together with AWS. Thank you very much.
Jennifer Grün | Senior Specialist for Generative AI and Machine Learning | AWS (58:07.914)
Yeah
📝 Copyright: All rights reserved — Startuprad.io™