Any article about roadmaps that has a lead in image of a Gantt chart is guaranteed to be missing the actual art of roadmapping. Any sufficiently agile/customer centric product company should never get near Gantt charts, with time estimates and dependency sequences, for roadmapping. No one can predict the future like that.
Gantt charts are not my preferred way of mapping out a vision and direction. But like some others stated, many orgs just like to work with that. It's also why I chose it as the header image, because it's familiar.
Kinda bummed out that would keep people from reading the article, but not I can't cater to everyone.
I'm not an expert in this matter, but it seems that the aspect of the chart that folks are questioning is the timeline. Could it be as easy as removing that aspect to dissuade the perception of a pseudo-commitment when presenting the roadmap?
That could be a way to go. Another option to represent time in a less fixed manner is to use a now-next-later roadmap schema.
It's more to the point of what's important vs. time as a measurement.
In the theory you might be right, but any company with a defined budget will simply ask for timelines and see it nicely displayed (often as gantt). Why do you think most Product Manager still use excel? Simply: Management wants this and you are not the boss... just my experience.
I agree, but also any real life product manager will likely have major stakeholders who can only comprehend what's happening or where the budget/time is being spent if they see it in the form of a gantt chart.
In the end, it's just a tool in a box of various others.
Gant charts can be used as a "living document" presenting the latest estimates. While individual task can deviate from the estimations the bigger picture view can be a nice rough estimate. I dont get the hate of the "YOU CANT PREDICT ANYTHING!1!1!" agile crowd subset. Getting good at predictions and dealing with deviations is what project management is all about, using tools that help you in this process is much advised.
