Case study 04 · Station 02 · Discovery and the decision
Killing my own product’s positioning
Asked to take our AI experience builder standalone, I ran the analysis — and recommended against my own product’s positioning.
01 ·The problem.
The ask: the builder — the system that constructs simulations, AI roleplays, AI coaches, and micro-assessments from machine-readable specs — looked like an obvious standalone product. “No-code AI experience platform” writes its own landing page.
02 ·Discovery and the decision.
I wrote the go/no-go the way I’d want to receive it: who exactly is the end user (undefined); what does “standalone” mean when the product’s magic depends on upstream tailoring intelligence (a contradiction); what does the market comparison say (comparables in the no-code creation space win on outcome-led positioning, not feature lists).
The verdict: not viable as positioned — but not dead. Conditional GO via a sequenced wedge: prove value with internal designers, extend to delivery partners, and only then open client self-serve — the pattern the strongest creation platforms followed.
03 ·The build.
What we did not build: the standalone launch as positioned. What did get built was the sequence — internal designers first, then delivery partners, then client self-serve — and a messaging rewrite around outcomes such as “ship a client-ready simulation in days” rather than capabilities.
04 ·Trust.
Every claim traced to evidence; every risk priced. A go/no-go that cannot show its evidence is an opinion.
05 ·Launch and learn.
The narrowed strategy was adopted. The standalone launch that would likely have failed didn’t happen; the sequenced one is in motion.
The most valuable strategy document is sometimes the one that says “not like this.”