Product roadmap: what it is, and how to build your first one
Ask three people on a small software team what the roadmap is and you may get three answers: a spreadsheet of tickets, a slide from the last investor update, and "whatever we're building this month". That confusion costs time, and customers keep asking the same questions.
Below: what a product roadmap is and isn't, then how to build a first one from the requests you already have, with a small made-up example.
What is a product roadmap?
A product roadmap is a short, shared statement of the customer problems you plan to solve, in the rough order you plan to solve them, and why. It has three ingredients:
- Outcomes. Each item says what a customer will be able to do, not which component changes.
- Order, not dates. Items sit in broad buckets such as Now, Next and Later, or in quarters if your readers plan by the calendar.
- A reason. Readers can see why each item is there: the customer problem, the theme it belongs to, or the requests behind it.
Product management roadmap work is mostly editing: choosing the dozen items worth showing and leaving the rest off.
What a product roadmap is not
Three other planning documents get mistaken for a roadmap. Each one is useful; none of them replaces it.
| Product roadmap | Backlog | Release plan | |
|---|---|---|---|
| Question it answers | What problems are we solving, and in what order? | What could we work on? | What ships in the next release, and when? |
| What's in it | A handful of outcomes | Every idea, bug and task | Specific changes with target dates |
| Main readers | Customers, sales, support, leadership | The people building it | The team and anyone coordinating a launch |
| How far ahead | Months, loosely | No fixed horizon | Weeks |
A roadmap is also not a promise of dates. Once a public card carries a day, readers treat it as a contract. Keep dates on the release plan.
How to create a product roadmap from real requests
- Decide who will read it. A customer-facing roadmap leaves out costs, staffing and anything under contract. An internal one can hold all of that.
- Gather what people have asked for. Pull requests from your feedback board, support inbox and sales notes into one list. Keep the requester's own words.
- Group requests by the problem behind them. Different wording often hides the same need. Merge those into one request so the votes count together.
- Name three to five themes. A theme is an outcome customers care about, such as "get paid faster". Every item should fit one.
- Sort into Now, Next and Later. Now is work that has started. Next is work you've decided on. Later is what you're weighing.
- Write each card as an outcome. One short title in the customer's words and one sentence on what gets easier.
- Publish it and point people to it. Link the roadmap from your help menu, and link each card to its request so voters can follow along.
A worked example
Hollowpine is a fictional timesheet tool for small design studios. Say its team has gathered the requests below. The vote counts are made up to show the method; they aren't a rule for when to build.
| Requests, in customers' words | Problem behind them | Roadmap card | Theme | Column | Votes (illustrative) |
|---|---|---|---|---|---|
| "Remind people on Friday", "timesheets are always late" | Studios chase staff for hours every week | Friday reminder to fill in hours | Hours in on time | Now | 34 |
| "Bill a client from my hours", "export hours for invoicing" | Turning hours into an invoice is manual | Client invoice from logged hours | Get paid faster | Next | 27 |
| "Show which projects run over" | Owners learn about overruns too late | Budget warning on each project | See where time goes | Next | 12 |
| "Log time from my calendar" | Typing hours twice | Fill hours from calendar events | Hours in on time | Later | 19 |
| "Dark theme" | Comfort, not a blocker | Left on the board, status Open | None yet | Not shown | 22 |
Two choices stand out. The calendar card has more votes than the budget warning but sits in Later, because the team hasn't worked out how to do it well yet. And the dark theme stays on the board but off the roadmap: it has votes, but it doesn't serve any of the three themes. Votes tell you what's common. They don't make the decision for you.
Managing a product roadmap after the first version
A roadmap goes stale faster than any other document a small team keeps. A few habits keep it honest:
- Change it when the work changes. When something starts, move it to Now. When it ships, take it off and say so in your changelog.
- Explain moves. If a Next card drops back to Later, add a sentence saying why. Readers forgive a change of plan far more readily than silence.
- Keep it short. A Later column that keeps growing has become a backlog.
A new product roadmap, before you have many requests
If the product is new, the board will be thin. Fill Now and Next with your own bets from early conversations, keep Later short, and open a feedback board on day one so the next version has real requests behind it.
Start from a layout
Two layouts suit a first product roadmap. Both open in a free Votelantern account with fictional example cards you can replace.
By theme. Each theme is a row running across Now, Next and Later, like the Hollowpine table above. You can see it filled in on the product roadmap template page.
Now, Next, Later. The same three columns with no theme rows, for when you want to start with the simplest thing that works. It's previewed on the roadmap template page.
In Votelantern, both layouts read the statuses on your feedback board: a request marked In progress shows under Now, Planned under Next, and Under review under Later. You can rename the columns. When the work ships, you publish a changelog entry linked to the request, which moves it to Shipped and emails the voters who chose "Email me updates". For finished roadmaps from four kinds of product, see the product roadmap examples.
These prices are drafts, not final. Free is $0 for one public board with its roadmap and changelog, one admin and a "Powered by Votelantern" badge. Team (price not set yet) covers up to three admins and up to five boards. Business (price not set yet) adds private boards, for a roadmap only your staff should see.
FAQ
Who should own the product roadmap? One person, usually whoever runs product. Others suggest and challenge, but a single owner decides what goes on it, so it doesn't become a list of everyone's favorite idea.
How far ahead should a product roadmap look? Far enough that customers see your direction, close enough that you believe it. For a small team, six to twelve months of loosely held intent is plenty.
What should I do when an item comes off the roadmap? Say so where people will see it. Set the request's status to Closed with a sentence of why, or move it back to Under review if it may return.