AI and vibe coding tools make resources abundant. The MVP concept was based on a time when resources in engineering were scarce. I am recommending everyone focus on an MSP or Minimum Sellable Product instead, because that's the real test. Can your prototype generate revenue?
Yup. The shift needs to be to MVH — Minimum Viable Hypothesis. MVH should test the belief that MUST be true for the company to work. Everybody can ship now so the question has to shift from “can we ship?” to “are we right?”
I’d argue that all of these are consistent with the underlying principles of lean startup. They all test different hypotheses but the capability to experiment at speed and without compromising quality has evolved massively. The concept of MVP isn’t predicated on poor UX it’s about validating assumptions at speed with the tools and resources available to you.
Except, the answer to the question whether MVP is dead is the same as to all the previous ones:
"TL;DR: No. Thank you for reading."
I could take most of the paragraphs and respond to them either by "what's new again?" or "that's not what [a thing] is" or "yup, that's how people misuse the concept."
A few random comments:
- Competitors being quick to copy any remotely sensible idea? What's new again?
- Customers churning because an early version is clunky or confusing? What's new again?
- Companies building big MVPs? Yup, that's how people misuse the concept.
- MVP is all about the functional aspect? That's not what MVP is.
- Juicero spending $120M on an MVP? Seriously? MVP? For 9-digit?
One observation nails it, though. That's how people *perceive* MVP. In the common understanding, MVP is anything anyone wants to pack into the first version of their product. In its popular usage, the term has lost any meaning.
"I'm building an MVP" is the startup equivalent of "We're doing Agile." The only sensible answer to that is: "But what do you mean by MVP/Agile?" Without clarification, it can literally mean anything.
Depending on the context, MVP might, indeed, be about:
- the smallest experiment that mitigates the biggest unknown
- delighting the first customer
- validating (typically only Problem-Solution Fit, but sometimes Product-Market Fit, too)
It's never been about:
- delivering something ugly or non-functional on purpose
- going fast and getting there early just to be the first
AI and vibe coding tools make resources abundant. The MVP concept was based on a time when resources in engineering were scarce. I am recommending everyone focus on an MSP or Minimum Sellable Product instead, because that's the real test. Can your prototype generate revenue?
Yup. The shift needs to be to MVH — Minimum Viable Hypothesis. MVH should test the belief that MUST be true for the company to work. Everybody can ship now so the question has to shift from “can we ship?” to “are we right?”
100%
I’d argue that all of these are consistent with the underlying principles of lean startup. They all test different hypotheses but the capability to experiment at speed and without compromising quality has evolved massively. The concept of MVP isn’t predicated on poor UX it’s about validating assumptions at speed with the tools and resources available to you.
We live in the era of dead things. It's been "Lean Startup is dead," then "Product-Market Fit is dead." No wonder it's time for MVP... I need to add it to my list (https://pawelbrodzinski.substack.com/p/ai-killed-product-market-fit-or-did).
Except, the answer to the question whether MVP is dead is the same as to all the previous ones:
"TL;DR: No. Thank you for reading."
I could take most of the paragraphs and respond to them either by "what's new again?" or "that's not what [a thing] is" or "yup, that's how people misuse the concept."
A few random comments:
- Competitors being quick to copy any remotely sensible idea? What's new again?
- Customers churning because an early version is clunky or confusing? What's new again?
- Companies building big MVPs? Yup, that's how people misuse the concept.
- MVP is all about the functional aspect? That's not what MVP is.
- Juicero spending $120M on an MVP? Seriously? MVP? For 9-digit?
One observation nails it, though. That's how people *perceive* MVP. In the common understanding, MVP is anything anyone wants to pack into the first version of their product. In its popular usage, the term has lost any meaning.
"I'm building an MVP" is the startup equivalent of "We're doing Agile." The only sensible answer to that is: "But what do you mean by MVP/Agile?" Without clarification, it can literally mean anything.
Depending on the context, MVP might, indeed, be about:
- the smallest experiment that mitigates the biggest unknown
- delighting the first customer
- validating (typically only Problem-Solution Fit, but sometimes Product-Market Fit, too)
It's never been about:
- delivering something ugly or non-functional on purpose
- going fast and getting there early just to be the first