Watch the full episode on YouTube: https://youtu.be/3osOPHJCtjo
Value proposition design is not a writing exercise — and that, says Alexander Osterwalder, is exactly why so many companies misuse AI. In Part 2 of our Strategyzer series on The Innovation Show, the co-author of Value Proposition Design joins Aidan McCullen to explain why generative AI can formulate a brilliant value proposition, but only after you’ve done the homework it cannot do for you. Skip the homework and you get the same lofty, buzzword-laden message as every competitor.
The trap is speed. Because AI lets teams build near-perfect software astonishingly fast, they build before they understand. Alex calls it “running up the wrong hill” — and without validated customer insight, you’ll run up a hundred wrong hills and still have no map.
The four foundations of value proposition design
Alex updates the foundations from his book for the AI era. First, address highly relevant customer jobs, pains and gains — and validate them quantitatively. “Waiting is a pain” isn’t evidence; is it five minutes or an hour? Even his world-class clients, he admits, cannot produce that evidence.
Second, prove the benefits. If your solution saves five minutes, show what those five minutes are worth — and show that a majority of customers agree. Third, differentiate through the customer’s eyes. Most companies benchmark against competitors and never ask customers whether the differentiation matters to them at all. Finally, formulate a clear, simple value proposition — then test that too.
Develop, formulate, then let AI do the typing
Alex splits the work into two phases: developing (product management) and formulating (turning it into a marketing message), with a double loop of testing and iterating around both. For a large industrial tech company, Strategyzer built this into a playbook now rolling out to a thousand product owners: map the business model canvas for context, validate the customer profile with evidence, then use the new Value Stack tool to prioritise the benefits customers care about most, followed by differentiation and willingness to pay.
Those are the ingredients. Once you have them, you feed a neutral product description, the validated customer profile and the value proposition into your favourite AI — and out comes a distinctive brochure, one-pager or landing page in minutes. As Alex puts it, marketing teams today “shouldn’t be writing a word.” Because the ingredients are unique, the output is unique; without them, AI produces the same high-level message for every competitor.
Should you feed your secret formula into AI?
Aidan raises a dilemma from a leader at a major confectionery company: do you put proprietary data into enterprise AI? Alex offers two answers. Practically, open-source models running on your own servers keep sensitive work in-house, and value proposition work rarely needs cutting-edge models anyway. More fundamentally, the fear is largely irrational — value only exists for a specific customer segment, so even a model trained on your solution couldn’t hand competitors a usable value proposition. Today, and for the foreseeable future, AI cannot generate a great value proposition on its own.
The bigger risk, Alex argues, runs the other way: teams that under-use AI, still wordsmithing websites by hand instead of iterating at lightning speed with validated ingredients.
Overthink, under-experiment — and the Federer playbook
For the solo inventor — like listener Corey Rollins and his prototype pillow — the advice is concrete. Validate jobs, pains and gains before perfecting the product. Then run ABC tests: three landing pages, each emphasising a different differentiator or price point, and let conversion rates supply the evidence. “We tend to overthink and under-experiment,” Alex says. “Corporations even worse.”
However, content alone won’t change behaviour. Alex’s favourite analogy: e-learning could show you how Roger Federer holds a racket, but only a coach’s playbook — watch, then do, then repeat — builds a champion. Growth, he insists, is a profession you learn the same way: learn, do, learn, do.
Alex also joins friend of the show Tendayi Viki for a webinar on evidence-based business idea evaluation — details at strategyzer.com.
About the guest — Alexander Osterwalder
Alexander Osterwalder is the founder of Strategyzer and co-author of Business Model Generation, Value Proposition Design and The Invincible Company. Ranked among the world’s leading management thinkers by Thinkers50, he created the Business Model Canvas used by millions worldwide. Explore the Strategyzer playbook library at strategyzer.com — Innovation Show listeners get 20% off with the code INNOVATIONSHOW20.
About the host — Aidan McCullen
Aidan McCullen is the host of The Innovation Show, a Thinkers50-recognised podcast, the 2025 Thinkers50 Innovation Award recipient and the author of Undisruptable: A Mindset of Permanent Reinvention. A former professional rugby player turned change and innovation specialist, Aidan works as a keynote speaker on AI, change, innovation and transformation for organisations in Ireland and worldwide. Connect with him on LinkedIn or at theinnovationshow.io.
About The Innovation Show
The Innovation Show is the Thinkers50-recognised podcast where square pegs find their place in a world of round holes. Each week, Aidan McCullen hosts world-class authors, scientists and practitioners on disruption, innovation, change, transformation, leadership and creativity. This series is brought to you by Kyndryl, who run and reimagine the technology systems that drive advantage for the world’s leading businesses — learn more at kyndryl.com. Sign up for the Kyndryl Institute and send us a screenshot for a chance to win a copy of Value Proposition Design, with thanks to Kyndryl.
Listen or watch now: Apple Podcasts | Spotify | YouTube
Frequently asked questions
Can AI write a value proposition?
Yes — but only the formulation. According to Alexander Osterwalder, generative AI can produce brochures, one-pagers and landing pages in minutes, provided you first supply validated ingredients: evidence-based customer jobs, pains and gains, proven benefits and tested differentiation. Without that homework, AI produces the same generic message for every competitor.
What are the four foundations of an irresistible value proposition?
Osterwalder’s four foundations are: highly relevant, quantitatively validated customer jobs, pains and gains; benefits backed by evidence that customers care; differentiation tested through the customer’s eyes rather than against competitors; and a clear, simple formulation that is itself tested and iterated.
What is the Value Stack?
The Value Stack is a new Strategyzer tool that analytically prioritises which benefits matter most to customers. It sits between the validated customer profile and differentiation in Strategyzer’s playbook process, and helps teams build value propositions on evidence rather than opinion.
Is it safe to put company data into AI tools?
Osterwalder gives two answers: practically, open-source models running on your own servers keep sensitive work in-house; and strategically, the fear is largely irrational, because a value proposition only creates value for a specific customer segment — AI cannot hand your competitors a usable version of it today.
What is an ABC test for a value proposition?
An ABC test compares three versions of a landing page or message, each emphasising a different differentiator or price point. Conversion rates then provide the evidence for which message resonates most — replacing logical debate with market data.
Who is Alexander Osterwalder?
Alexander Osterwalder is the founder of Strategyzer, creator of the Business Model Canvas and co-author of Business Model Generation, Value Proposition Design and The Invincible Company. Thinkers50 ranks him among the world’s most influential management thinkers.