Why do we still create roadmaps like it's 1995? (PWW #12)

It's time to start thinking differently about your roadmap (or backlog). I have already written previously about the liberating feeling of deleting your backlog, and why you should get rid of feature roadmaps.

In my experience, while a roadmap is invaluable and serves as an excellent conversation piece, it can also become a source of stress and anxiety. Why is that so?

The way you present your roadmap or backlog significantly impacts subsequent discussions and questions. Remember, there's no single best way to present a roadmap. It's worth exploring different methods to find what suits you best.

The Dreaded Feature Roadmap

Feature roadmaps are common, and you're likely working with one. They typically outline specific features to be developed over time (weeks, months, quarters, etc.).

This format is familiar to most in the organization, reminiscent of old-school waterfall project timelines. That we're still using this in today's world seems a bit odd right?

The Agile movement has been advocating for 2 decades to replace long-term plans with short iterations. In order to quickly react and adapt to new findings.

So, why do we still create roadmaps like it's 1995?

This familiarity, especially to senior leadership, partly explains the persistence. However, this approach has some major issues:

1. It forces you to do precise planning far into the future. That's proven to be virtually impossible, especially when building a digital product. No one knows what the future holds. Why pretend?
2. Everyone knows a plan will never work out exactly how you envisioned it. But we still grasp on to timelines like these for dear life, against our better judgement. It's all one big mirage.
3. Feature roadmaps give a false sense of certainty to stakeholders. You can bet that sales will start selling features on the roadmap to customers today. Now the plan has lost all flexibility. And the only possible outcome is disappointment.

It's no secret how I feel about this type of traditional feature roadmap... Now, let's have a look at some alternatives that are a much better fit for Agile product development.

Alternative 1: The GIST Framework

The GIST framework by Itamar Gilad is a compelling alternative. GIST stands for Goals, Ideas, Step-projects, and Tasks. These represent the various levels of building blocks used in the roadmap.

I has a clear hierarchy of moving from business Goals to concrete Tasks. Everything that is being worked on can easily be traced back to either a larger project or idea, and the corresponding business goal.

In addition the GIST framework embraces the fact that there might be Ideas but not necessarily 100% certainty on how to achieve a Goal. So not too much time is spent planning out all the little steps for each idea. The focus is put on the near-term work. Everything else can change.

Each Idea is broken up into small step-projects. Each step-project can be seen as an experiment to test (part of) the idea it relates to. So no 6-month long step-projects. Instead a first step-project might be to build a visual mock-up and test it with 10 users. Or to interview 10 potential customers and get their input.

Finally this results in specific Tasks for teams to work on. These are usually broken down so they can be finished in a 1-2 week period, similar to a Scrum approach.

Be sure to read the introduction post by Itamar Gilad to learn more about the GIST framework.

Alternative 2: Now-Next-Later

I've written previously about the Now-Next-Later format. It was created by Janna Bastow at ProdPad. It is a very simple way of visualizing which initiatives or goals have been prioritized.

The Now-Next-Later roadmap embraces the fact that there are varying levels of uncertainty. It visualizes that uncertainty in three columns:

Now?- Where are you now, and what are you currently working on in the short term? These are detailed initiatives for which you already have a reasonable idea of the scope and the approximate amount of time they will take from the team. In terms of the time horizon, these are your upcoming few sprints.

Next?- What will you pick up immediately afterward? These are the highest-priority initiatives you want to tackle as soon as the current work is completed. These initiatives are less detailed, and the scope is sometimes not entirely clear yet. They could also be initiatives dependent on what you are working on in the Now column.

Later?- What's on the idea list for later? These initiatives are still vague, perhaps only one-liners without detailed elaboration. The (technical) solutions are still unclear. You will further develop these initiatives as you progress in your Now and Next activities.

Alternative 3: Outcome Roadmap

Outcome roadmaps are a very powerful option if GIST or Now-Next-Later do not fit your needs. The premise of an Outcome roadmap is to plan outcomes instead of output. It can be seen as a light version of the GIST framework.

Working together with management to create an outcome-based roadmap is a great first step away from the traditional output-based roadmap. It combines the benefits of a timeline roadmap (providing insight into development time and planning resources) with the advantages of focusing on outcomes (providing clarity on goals, making it easier to link to the company's strategy, and being more business value-oriented).

For each business goal you would create these separate "phases" on the roadmap:

1. Research & Ideation (optional)
2. Product Discovery
3. Product Delivery
4. Effect Delay

These are pretty self-explanatory, but one stands out: the Effect Delay. Effect Delay acknowledges that even after launching a feature or a product,?some?time will elapse before data can prove if you have succeeded.

You need to take that delay into account. Because if the effect you were expecting does not happen, you might have to go back to the drawing board and find another way. So a Goal can and should be iterated on until the desired outcome has been achieved.

What is your preferred roadmap type?

I have highlighted three options here but of course there are many more.

I'm curious to hear from you! Leave your opinion below or send me a DM if you have a question.