A structured 90-day sprint for founders who want to move the needle — without burning cash on channels that don't fit their stage.
Most growth advice is timeless and useless: "build a great product," "know your customer." This is not that. This is what a competent operator would actually do if you dropped them into your business tomorrow with 90 days and a mandate to move the numbers.
The sprint has three phases. Do not skip the first one. Founders who skip diagnosis and jump straight to tactics almost always spend the next 60 days undoing the wrong bets.
Days 1–14: Diagnose
You cannot fix what you cannot see. Two weeks of honest looking beats two months of enthusiastic doing.
- Pull the last 12 months of revenue by channel, product and customer cohort.
- Interview five customers who bought recently and five who almost did but didn't.
- Map every stage of the funnel and measure the drop-off between each.
- List every marketing activity currently running and the last time you looked at its numbers.
- Write a one-page "state of the business" — plain language, no jargon, no hedging.
The one-pager is the deliverable. If you cannot fit an honest snapshot of your business on one page, you do not yet understand it well enough to fix it.
Days 15–45: Design
Pick the two levers with the highest ROI at your current stage. Not five. Two. This is where founders lose the sprint — the temptation to try everything is enormous and every attempt to spread bets thinner is a small vote against the outcome you claim to want.
Common pairings we see work well by stage:
- Pre-product-market-fit: positioning + qualitative interviews.
- Early revenue, no repeat customers: onboarding + retention.
- Repeat customers, slow growth: pricing + conversion rate.
- Growing but unprofitable: channel mix + unit economics.
- Profitable but stuck: brand + partnerships.
Days 46–90: Deliver
Execution is the boring part and the whole point. Six weeks is enough time to run three fortnightly cycles — plan, ship, measure — on each of your two chosen levers.
- Execute weekly, not quarterly. Ship, measure, adjust, repeat.
- Kill the two worst-performing marketing activities to free up budget and attention.
- Instrument every experiment before you launch it, not after.
- Keep a written log of every decision and the reason for it — future-you will thank present-you.
- End the sprint with a written retro: what worked, what didn't, what's next.
What a good retro looks like
The retro is not a victory lap. It is a document you will read at the start of the next sprint, when you have forgotten every lesson from this one.
- What did we predict would happen? What actually happened? Where were we wrong?
- Which experiments earned the right to become permanent activities?
- Which activities failed cleanly, and which failed messily (and why)?
- What are the two levers for the next 90 days, and why those?
"The businesses that grow fastest are almost always doing fewer things, better — not more things, faster."
Ninety days is not long. It is exactly long enough to prove one hypothesis and disprove another. Use it that way and compounding does the rest.