The Idea · Roadmap & Action Plan
Build
Your Startup
2026Part 01 / 03Lean → Validate
Start with
Who,
not What
not What
A startup isn’t a product — it’s a solution to someone’s real problem.
// The mantra
The linear path
Why build
Who for
Which idea
Validate
01 / 16
// First principle
You never start from a product. You start from an audience you love, then a real problem, then the simplest solution.
A startup is a solution to someone’s real problem — not the other way around
Step A
Audience
A specific group of people you genuinely care about serving.
Step B
Problem
A real, recurring pain they already feel and talk about.
Step C
Solution
The most convenient fix — and you charge for it.
02 / 16
Part 01
01
The
Roadmap
Roadmap
// The path is linear
01
Why build now
Mindset & model
02
Who it’s for
The audience
03
Which idea
Problem → solution
04
Validate
The MVP
Four steps · in order
03 / 16
Step 01 — Mindset & model
Why Build
Now
Start lean. You can validate an idea in days, not years — then charge for the solution.
// The winning model
Subscription &
recurring revenue
recurring revenue
Build once, sell access. Revenue you can actually predict.
The shift
“The milk,
not the cow”
not the cow”
People want access without owning — and it now reaches far beyond software.
A startup can take many forms
→ An app
→ A marketplace
→ A subscription service
→ A physical product, with a smart twist
The best businesses all do the same thing: solve a real problem for specific people — and get paid for it.
04 / 16
Step 02 — Before the idea!
Find Your
Audience
Counter-intuitive but central: start with who, not what. Every product needs an audience — so start there.
The real fuel
Passion for an audience you love serving
It’s what sustains the endless cycle: talk → uncover pains → build → acquire.
Above all
Narrow it
as much as
possible
as much as
possible
The narrower the audience, the faster & smarter you test.
Case
Amazon
started with book buyers only.
Become THE product for a niche — not just another product.
05 / 16
Step 03 — Problem → solution
A Winning
Idea
You don’t start from an “idea” — you start from a problem.
The key ratio
Problem
Solution
Keep them in balance
// For each idea, ask yourself 3 questions
01
What REAL problem am I solving?
02
How do I solve it?
03
Are people willing to pay?
There are several “idea ponds” to surface problems worth solving → Part 03
06 / 16
Step 04 — The MVP
Validate.
Don’t Build
Yet.
Building too early is the mistake 99% of founders make. Replace “I think” with “I know”.
// Minimum Viable Product
Often just a
landing page + a sign-up form
landing page + a sign-up form
Dropbox’s demo video → 75,000 sign-ups before a product existed.
No-code proof
Zappos
Photographed shoes in stores; bought them only once ordered online.
Manual proof
Groupon
Started on plain WordPress, hand-emailing PDF vouchers.
Go where they are
Talk to people
Distribute, then 1-on-1 conversations to understand their pains.
Real validation
$75k+
Paying customers from day one — Linkdrip’s pre-sales.
07 / 16
Part 02
02
The Action
Plan
Plan
Explainable to anyone — recite it in order.
// From audience to a paying customer
10
Steps
Steps
A simple, sequential version you can say out loud — no jargon required.
08 / 16
The Action Plan — first half
Steps 01 → 05
01
Pick an audience you love — not a product. Gardeners, accountants, podcasters…
02
Narrow it down until ultra-precise. “THE planning tool for gardeners” > “a planning tool”.
03
Go listen to that audience where it hangs out — Facebook groups, Reddit, X, forums, IndieHackers…
04
Spot a real, recurring problem — note their exact words: “I wish there was something that…”, “I’m struggling with…”.
05
Imagine the simplest solution — the most convenient fix to that exact problem.
Where to listen
FACEBOOK GROUPSREDDITTWITTER / X
FORUMSINDIEHACKERSTHEIR EXACT WORDS
09 / 16
From idea to paying customer
Steps 06 → 10
06
Filter the idea — 3 questions What REAL problem? How do I solve it? Are people willing to pay?
07
Build an MVP — without building it all Landing page (Carrd, Webflow) + form (Beehiiv), no-code (Softr, Bubble), or by hand.
08
Distribute & talk 1-on-1 — where your audience lives. DMs, a genuine non-automated email, a chat widget like Crisp.
09
Iterate until it clicks — change the message, the page, the prototype.
10
Chase real validation — a customer who pays.
Best of all
Charge upfront in pre-sale
Let the customer fund the build itself.
Build-an-MVP toolkit
CARRDWEBFLOWSQUARESPACE
BEEHIIVEMAILOCTOPUSSOFTR
BUBBLECRISP
10 / 16
Part 03 — The method
03
Find At Least
5 Ideas
5 Ideas
// The technique
The idea comes from the problem. For each of the 8 “idea ponds”, write down at least 1 problem + 1 solution.
8
Ponds → 8 leads → keep the best 5+
11 / 16
The 8 Idea Ponds
For each pond: 1 problem + 1 solution
#
Idea pond
The question to ask
Examples
1
Scratch your own itch
What annoys me or eats my time daily?
Slack · ProductHunt
2
Friends & family
What painful manual work would people around me ditch?
Internal tool → boss
3
Research the internet
Who’s complaining online: “I can’t find anything that…”?
Jasper AI · Intercom
4
Tap into an ecosystem
What need arises around Shopify / Notion / Stripe?
Baremetrics · Save to Notion
5
Disrupt an outdated market
Which manual-heavy market can be reinvented?
Uber · Airbnb · Loom
6
Big market, bad products
Which incumbent is too costly / complex / poorly supported?
Crisp · Plausible
7
Platform shifts
Which emerging platform offers an opening? (never just one!)
Hypefury · ChatGPT plugins
8
Copy what works
Which small, profitable business (10–50k MRR) could I rebuild?
Acquire.com · startups.fyi
12 / 16
The 3 Filters of a Winning Idea
Run every idea through all three
01
The 3 base questions
What REAL problem am I solving? How? Are people willing to pay?
// The non-negotiables
02
Better / Cheaper / Easier
Beat what exists on at least one axis.
Crisp = simpler + cheaper than Intercom. Instantly = simpler + cheaper → $8M ARR.
03
Sticky
The more data & config a customer accumulates, the less they leave.
Crisp = conversation history. Instantly = funnels already built.
13 / 16
Your exercise
5 Ideas
in 30
Minutes
in 30
Minutes
One row per pond = 8 leads already. Keep the best 5.
// Copy this template & fill it in
Pond chosen ______________
Target audience (be precise!) ______________
Real problem (their words) ______________
My solution ______________
Filter 1 — Willing to pay? yes / no / why
Filter 2 — Better / cheaper / easier? ______________
Filter 3 — Why is it sticky? ______________
Repeat × 8 ponds
14 / 16
Action Recap
Do these this week
→ List your passions — what you love and hate about each.
→ List 3–5 audiences you’d love to serve.
→ Go where they’re active and collect complaints & requests.
→ Explore ecosystems of products you love to spot a gap.
→ Identify a familiar market that could be reinvented.
→ Read negative reviews of big incumbents → a simpler/cheaper alt.
→ Pick 3 ways to build an MVP and the platforms to launch on.
→ Then go talk to people. Real validation = a paying customer.
📘 To go further
The Lean Startup
Building an MVP the disciplined way.
📘 To go further
The Mom Test
Asking the right questions of future users.
15 / 16
// Remember
Audience.
Problem.
Solution.
Problem.
Solution.
A startup isn’t a product — it’s a solution to someone’s real problem. Start lean, validate fast, and let paying customers tell you the truth.
Now go
Pick an audience. Find the pain. Pre-sell the fix.
Build Your Startup
Roadmap · Action Plan · 8 Idea Ponds
16 / 16