Photo by charlesdeluvio on Unsplash

The Risk Roadmap

John Zeratsky


All startups have a product roadmap, but at the earliest stages, most of them don’t need one. That’s because a product roadmap is a plan for what you’ll build—but as a startup, you probably don’t know quite what you need to build.

On the other hand, you probably know exactly what you need to learn.

Remember, startups are learning machines, and the work you do as a founding team—talking to customers, prototyping, running experiments, etc—makes the machine go.

That’s why having a clear Risk Roadmap is so important for every startup.

“Risk Roadmap” is how I often describe your plan for what you need to learn. It should cover:

  • What risks does your startup face? (Sometimes phrased as “what needs to be true?” or “what do we need to believe for this to work?”)
  • What questions can you answer to de-risk your startup? (Essentially translating risks into testable hypotheses)
  • What techniques will you use to answer these questions?

Let me share a few examples of how this looks and how you can use it.

Digit’s one big risk

Digit is an app that automatically budgets, saves, and invests for you, so you don’t have to think about it. We invested in Digit at GV and ran a Design Sprint with the team in 2014—right before they launched their initial savings product.

From talking with customers, the Digit team decided that trust was the major risk for the product. This was still early in the days of consumer fintech, and most people weren’t comfortable giving an app the ability to move money into and out of their bank account.

We decided the most important thing to learn was: Will users trust us enough to link their bank account? If not, nothing else on the roadmap would matter.

Digit’s key risk pre-launch: Will users trust us?

Experiments with copywriting, product, and marketing design helped us answer these questions, and the launch went really well. Digit eventually helped their customers save nearly $8 billion and was acquired by Oportun in 2021.

Building your Risk Roadmap in Design Sprints

Digit was focused on one major risk that would unblock everything else, but most times, a Risk Roadmap should contain a sequence of risks that need to be addressed.

Some teams are already doing this (we love to see it in a fundraising memo 😉) but many times, we build the first version of a startup’s Risk Roadmap together in a Design Sprint.

Here’s a rough example, with the risks phrased as “assumptions” here…

Building the first version of a Risk Roadmap in a Design Sprint. (P, U, L, C, and H indicate the type of experiment we’ll run to validate those assumptions.)

Here’s how we do this in Miro… (sorry it’s so heavily redacted!)

Capturing a startup’s Risk Roadmap in Miro

Marking progress with your Risk Roadmap

The best part of The Risk Roadmap — very much like a regular product roadmap—is that it’s so actionable. You can use your Risk Roadmap to plan what experiments you’ll run, and to keep track of progress toward de-risking your business.

For us, this often starts by connecting results from customer testing directly back to your risks.

Here’s how this looked in a Design Sprint with Phaidra

Each question is a risk, and each column represents our answers to that question based on a particular customer interview. The right-most column is our overall answer to the question.

And here’s an example of how you can summarize what’s been de-risked, and which risks are still open, as we did recently after a sequence of Design Sprints with a portfolio company…

High-level summary of what’s been de-risked and what remains open after multiple Design Sprints

The learn-to-build pipeline

Look, it’s definitely more fun to focus on building product than on answering these tough questions, and if you read enough VC tweets and blog posts 👀 it might seem like the only thing worth doing. (I’m still waiting on that Andreessen sequel, It’s Time to Run Experiments and De-Risk Our Startups.)

And product roadmaps are part of the fun—as a founder, there’s nothing more exhilarating than mapping out your vision for the future.

But if you don’t start by learning, you may end up building a future that doesn’t make sense for your customers or your company.

That’s why I think it’s so important to build a Risk Roadmap—to deeply understand the risks your startup is taking and make a plan for figuring out whether you’re on the right track.

There’s no perfect way to do this—it can be as simple as a list on a whiteboard—but it’s worth going slow now so you can go fast later.

After all, the only thing more fun than building your vision is going all-in building a vision that you know the world wants and needs.

Learn, then build.

And use The Risk Roadmap to find your way there.

— JZ

The best part of using a Risk Roadmap is at the end—going all-in on building stuff you know people want



John Zeratsky

Supporting startups with capital and sprints. Co-founder and general partner at Character. Author of Sprint and Make Time. Former partner at GV.