My idea discovery framework
From raw idea to a go, pivot or kill decision
Five steps I use with every founder. It turns a hunch into a sharp problem, a testable bet and, finally, a decision you can defend to yourself and your investors.
-
01
Discover
Map the problem space and the people who feel the pain most.
What we look at: who feels the pain most, how they cope today and what it costs them. Real conversations, not surveys.
-
02
Define
Frame one sharp problem statement and the riskiest assumption behind it.
The output is one problem statement and the single assumption that would kill the idea if it were wrong.
-
03
Design
Choose the cheapest experiment that can prove that assumption wrong.
Interviews, a landing page, a concierge test or a prototype: whatever is cheapest and fastest to learn from.
-
04
Build
Ship the MVP or prototype, AI-assisted, in days rather than months.
With my team at Obii Kriationz, AI-assisted, so you get something real into users’ hands in days rather than months.
-
05
Decide
Read the evidence and commit: scale it, pivot it or kill it.
Scale it, pivot it or kill it. Either way you decide on evidence, not on hope.
For you: a go, pivot or kill decision backed by evidence, before you spend your savings.
Lean and agile philosophy
Every plan is a hypothesis
Small shipments, short cycles
Weekly sprints with something real to show at the end of each one.
Validated learning over opinions
Every feature starts as an assumption and earns its place with evidence.
Customers at the centre
Founders talk to users every sprint, not once before launch.
Inspect and adapt
Retrospectives turn every miss into the next experiment.
FAQ
Questions about validating an idea
Still unsure? Send a note with your idea and I'll tell you honestly whether I can help.
Ask me directlyAn idea discovery framework is a structured way to test whether a startup idea solves a real problem before building it. Abhiram’s framework runs in five steps: Discover, Define, Design, Build and Decide, ending in a go, pivot or kill decision.
Frame one sharp problem statement, identify the riskiest assumption behind it and design the cheapest experiment that could prove it wrong, such as customer interviews, a landing page test or a clickable prototype, before writing production code.
That is a win. Killing or pivoting an idea after a few weeks of evidence costs far less than discovering the same thing after a year of building. The Decide step makes that call explicit.
AI speeds up research in Discover and Define and lets the Build step ship an MVP or prototype in days rather than months, so founders get to real evidence faster.
Let's talk
Run your idea through the framework.
Tell me about the idea, the customer and where you are stuck. I will tell you honestly whether I can help, and what the first ninety days should look like.
- A validated problem, not a guess
- A launched MVP, faster and cheaper
- Evidence for customers and investors