Ninety-Day Launch Roadmap: Find Them, Keep Them, Repeat What Works
A ninety-day roadmap has three phases, each with a different job. Days one to thirty find the few people who care. Days thirty-one to sixty make them stay, which means onboarding, a first price and conversations. Days sixty-one to ninety repeat what worked and drop what did not. Each phase ends with a written milestone and a rule for what to do if you miss it.
Why plan ninety days, not thirty?
A month is long enough to learn whether strangers will try a product. It is not long enough to learn whether they will keep using it, whether they will pay, or whether any channel repeats. Those questions need a second and third month of the same product in front of new people. A roadmap holds that longer arc so that you do not mistake a good first week for an answer.
The phases below each ask one question: who cares, will they stay and what repeats. If you need the detail of the first phase, it is the thirty-day plan. This page spends its words on the two phases that plan stops before.
Days 1 to 30: who cares?
The aim is a short list of named people who finished the main task and told you something specific. Publish on a free board, post in one community, answer every reply and keep one sheet of numbers per tagged link. At day thirty you should be able to write the sentence “this is for [person] who [problem]” and name three such people.
If you cannot, do not move on. Spend the next two weeks reaching more of the right people, not adding anything to the product. The free launch board is a convenient first public page for this phase because it costs a form and one emailed link, and it holds nothing back for a paid tier.
Days 31 to 60: will they stay?
Phase two is quieter on the outside. You are not looking for new channels; you are watching the people you already have. The work is mostly conversation and small fixes.
- Talk to five users for twenty minutes each. Ask what they were doing before they found you, what they did the last time they used it and what they would miss if it vanished. Do not pitch.
- Fix the first five minutes. Where do new users stop? Shorten that path, even if it means removing a feature.
- Put a price in front of the people who returned. A founding price, a paid pilot or a trial that ends. See how to price a micro SaaS for models.
- Write the repeating answers down. Every question asked three times becomes a line on the page or a short help page.
- Measure return visits weekly, not total users. The retention rate entry explains the idea in plain terms.
Days 61 to 90: what repeats?
Now you can afford to scale one thing. Take the channel that produced the best conversations in phases one and two and repeat it with discipline: the same kind of post in a new community, a second page for a second search question, a second free directory. Drop everything that did not produce a reply or a return visit, even the places you liked.
- Build assets that keep working. A page that answers a question brings visitors without you. The launch with SEO from day one playbook shows how to pick the question.
- Collect proof. With permission, keep a quote or two from users, and note what changed for them. Real quotes beat invented ones every time.
- Decide whether a relaunch is earned. If you changed something substantial, the relaunch playbook covers how to announce it without repeating yourself.
What are the milestones, and what if you miss one?
| Day | Milestone | Evidence | If missed |
|---|---|---|---|
| 30 | Three named people finished the main task and said something specific | Notes with names and the link each came from | Spend two more weeks reaching the right people; do not add features |
| 45 | Five user conversations done; first-five-minutes fix shipped | Call notes and a before and after of the path | Pause new channels; fix the path, then talk to five more |
| 60 | A price shown to returning users; at least one reply to it | The message sent and what came back | Re-ask with a different model, such as a trial; if still nothing, rethink the user |
| 75 | One channel repeated with similar results | Two tagged links, side by side | Change the sentence, not the channel |
| 90 | A written decision: grow, change or stop | A one-page review | Stopping at this point is a result; keep the notes |
How does the roadmap change with fewer hours?
The phases stay; the number of things in each shrinks. These are planning figures for comparison, not measurements of anything.
| Hours a week | Keep | Drop |
|---|---|---|
| About 3 | One public page, replies, one conversation a fortnight | Any second channel; the search page until phase three |
| About 8 | The full plan above | Anything beyond three places in total |
| About 15 | The full plan, plus a second search page in phase two | Paid options; they still need a tested price |
At any level, stay honest about what the week cost. A plan you can follow beats a bigger one you drop in week five. The launch time budget template helps you count.
What can wait until after day ninety?
- A rebrand. Change the name only if users are confused by it.
- Paid acquisition, until you know what a retained user is worth.
- A second product. Finish the answer to this one first; product-market fit explains how to tell when you have it.
If you want ideas for what the next ninety days could build, the ideas board is open, and you can post one of your own on the submit page if you would rather someone else built it.
Frequently asked questions
Is ninety days too long for a launch plan?
For a small product it is the shortest span that answers the questions a launch raises: will people stay, will they pay and does a channel repeat. Thirty days answers whether strangers try it. Ninety days gives you a second and third month of data, with a written decision at the end.
What should I do in days 31 to 60?
Watch the people you already have. Talk to five users for twenty minutes each, fix the first five minutes of the product, show a price to returning users and write down repeating questions. Do not add channels yet. Measure return visits weekly and treat total users as secondary.
When should I stop a product at day ninety?
When nobody beyond your circle used it twice after you reached the right people with a clear sentence, and a different price or user did not change that. Stopping is a result, not a failure. Keep the notes and spend the next ninety days on what you learned.
Can I follow the roadmap with only three hours a week?
Yes, with a smaller plan: one public page, regular replies and one conversation every fortnight. Leave any second channel and the search page until phase three. The phases stay in order, and a plan you can keep going is better than a bigger one you abandon by week five.
Written by Post your build ideas with AI assistance and edited for this site. Where a page names another product, the facts came from that product’s own pages on the date shown above; terms change, so check them before you rely on them. About the publisher.