AI Product 1.0–3.0 Adoption Ladder
Advance from curious tourists to early adopters and then the mass market.
- Difficulty
- Moderate
- Time to result
- ~months to results
- Steps
- 5
- Confidence
- 98%
Segment product evolution into three adoption stages. Version 1.0 serves AI tourists who arrive because the capability is novel; many will leave, but their behavior reveals the product’s most magnetic value. Version 2.0 serves genuine early adopters who tolerate limitations and actively articulate what they need, creating a concentrated product-learning loop. Version 3.0 is the mass-market test: the experience must become reliable, understandable, and familiar enough for users who dislike changing established habits. Teams should use a different success criterion at each stage—discovery in 1.0, customer-led iteration in 2.0, and durable mainstream adoption in 3.0.
Origin
Grant Lee used Gamma’s successive product generations to explain how AI products move from experimentation to mainstream adoption.
Core principles
- 01Different product stages serve different user tolerances.
- 02Early users reveal what is interesting before the product is polished.
- 03Early adopters pull missing capabilities out of the team.
- 04Mass-market readiness is tested by people who dislike change.
How to run it
- 1
Launch for tourists
Expose the novel capability to curious users and study what repeatedly captures their attention.
Pro tip Separate novelty-driven traffic from recurring usage.
Watch out Do not interpret every initial signup as durable demand.
- 2
Find the committed adopters
Identify users willing to keep working through shortcomings because the core outcome matters to them.
Pro tip Interview users who return despite friction.
Watch out Vocal enthusiasts may not represent the eventual mass market.
- 3
Let users pull the product
Collect repeated requests, observe workarounds, and build the capabilities that enable serious use.
Pro tip Prioritize requests tied to repeated workflows.
Watch out Avoid becoming a custom-development shop for isolated requests.
- 4
Prepare mainstream usability
Remove avoidable complexity, raise reliability, and make the experience recognizable to ordinary users.
Pro tip Test with people who are not AI enthusiasts.
Watch out A stronger model alone does not guarantee an accessible product.
- 5
Run the change-resistance test
Launch 3.0 broadly and assess whether change-averse users can adopt it without extraordinary support.
Pro tip Track activation and retention by user sophistication.
Watch out Do not mistake another burst of curiosity for mainstream acceptance.
In the wild
A reporting startup uses its first release to discover that users value automatic narrative summaries. Its second release works with analysts who request source citations, templates, and editing controls. Its third release embeds those capabilities into a familiar dashboard that nontechnical managers can use without learning prompting.
→ The product progresses from novelty to a repeatable mainstream reporting workflow.
Common mistakes
Selling 1.0 as mass-market ready
Tourists may tolerate defects that ordinary users will treat as immediate reasons to leave.
Ignoring early-adopter pull
Building from internal assumptions wastes the unusually detailed feedback supplied by committed users.
Is it for you?
Best for
It is best for novel AI products that attract experimentation before they are reliable enough for mainstream workflows.
Not ideal for
It is not ideal for products that must be complete and safety-certified before any user exposure.
From the transcript
“I oftentimes think about like 1.0 as this is for the AI tourists, right? They're gonna come, many are gonna leave.”
“And then you get to your like 2.0 lodge. And this is for your true sort of early adopters.”
“And then you finally get to the point where you feel like you can launch a 3.0, and 3.0 is the first time you're actually…”
From the episode
The AI Workflow That Lets 50 People Do the Work of 500 ($2B Founder Reveals)