Creating a product roadmap can be approached in numerous ways. You can, for instance, gather quantifiable data and calculate a project timeline with milestones. Or you can make stuff up. Facetiousness aside, both are valid approaches but serve different aims. The first is a projection that is derived from an estimated development backlog. Below is an example:
You can create this in your favorite spreadsheet application. I used Apple Numbers which replicates the functions I care about in Excel with reasonable fidelity. Excel will work too, obviously. I cobbled together this tool from various artifacts collected over the past few years with the goal of creating a full backlog burn-up. Other tools, like JIRA, can create a Sprint Burn-up which is just a Sprint Burn-down turned on its head. By contrast, this tool doesn't just produce a pretty chart—it can be used to predict the future! All it requires is a well-estimated backlog and a little manual input of "Story Points Completed", sprint to sprint, and the total "Backlog" points if and when that changes. The rest it will calculate.
The nice thing about timelines derived from data is that they adjust to current realities. An increase or decrease in project scope, or a change in team velocity, will alter the projection. If the team puts in the requisite estimation work, you can expect to get a six to nine month projection using this technique. In fact, the time horizon is unlimited, but you can’t expect any accuracy beyond a nine-month period. It’s just the nature of the product development universe.
The second type of roadmap is more familiar on the business and marketing side of things. These can be of any format, but tend to be highly visual and focused on a narrative about the company’s objectives over the longer term, typically projecting two to five years into the future. For obvious reasons, they tend to be aspirational and are always highly inaccurate. Which isn’t to say there isn’t a place for them! There's nothing wrong with having aspirations. I’ve created more than a few of these myself.
My role as a product manager has often been to reconcile marketing roadmaps with projections from the engineering team. This is a good use of time—what the business wants and what we can predict will actually happen is a gap that needs to be bridged (and communicated) again and again. And again. Good luck!
Interested in a copy of this burn-up tool? Just send me a note.