Embracing Uncertainty: A Modern Take On Strategy, Goals, and Roadmaps

We all strive for certainty — if we do X, Y will happen. Certainty is comforting; it allows us to operate without fear, doubt, or having to think too much. Many of the processes your company employs, especially strategy, roadmap, and project planning, are meant to create certainty — Here’s what we need to do and by when. The plans flow nicely into each other to create a cascading planning process. 

The Planning Waterfall

But developing tech products is anything but certain. There are many moving parts in the market, technology, product, and org; many assumptions, dependencies, friction points, and potential snafus. Our well-crafted plans are often no more than a house of cards based on opinions, consensus, and sparse data. The time we waste on false-certainty planning is just the tip of the iceberg. Below the surface we waste a lot more in in unnecessary development, product bloat, and maintenance.

Dealing Honestly With Uncertainty

The alternative philosophy embraces uncertainty as a reality to deal with, openly and unashamedly. In fact, handling uncertainty well is a major competitive advantage as so few companies are able to do it.

There are a few principles that help reduce uncertainty and hedge risk:

  • Optimize for outcomes rather than output — your goals are less likely to be wrong
  • Embed evidence-based learning into strategy and tactics
  • Consider multiple options rather than just one

This is all good in theory, but I want to show concrete techniques — in strategy, goals, roadmaps, and idea prioritization. 

Upcoming Workshops

Practice hands-on the modern methods of product management, product discovery, and product strategy.  

Secure your ticket for the next public workshop
or book a private workshop or keynote for your team or company. 


Embracing Uncertainty in Product Strategy

Strategy is where we face the highest levels of uncertainty. As we’re projecting years into the future, any concrete plan of action is very likely to fail. Even coming up with a clear set of choices (the more modern definition of strategy) is very hard. Research, analysis, and experimentation therefore have to play a major part of strategy development, and the strategy has to keep evolving over time.

Here are some ways to do it.

The screenshot below is taken from a demo of Dotwork, a currently stealth startup whose CPO is John Cutler, the prolific author of The Beautiful Mess (highly recommend).
Update: the video seems to be down, but you can see the pertinent screenshot below.

Dotwork demo: 5 strategic choices with 5 potential options for each creates a 5×5 possibility space, from which we can create multiple combinations, each representing a different strategy. See the full demo video here

What I especially like here is the concept of Possibility Space, and how a set of strategic choices, in this case based on the writings of Roger L. Martin, creates one possible strategy. Comparing and scoring multiple possible strategies makes it hard for managers to fall in love with just one shiny object. Basing the scores on research forces the developers of the strategy to face reality rather than bask in fantasy. 

Ash Maurya, author of Running Lean and Scaling Lean (also highly recommended), gives similar advice to startups. He recommends creating a number of Lean Canvases (a derivative of Strategizer’s Business Model Canvas) to model various 0-to-1 strategies. Eadh canvas represents one possible combination of target market segment, problem, solution, value proposition, business model and other dimensions. Once you’ve created a few plausible lean canvases, you identify the key assumptions in each and validate them, which will eliminate most ideas, but (hopefully) leave you with at least one valid option.

Creating and validating multiple Lean Canvases. Source Ash Maurya/LeanFoundery.com

In the same vein, in this article I argue that some of the biggest strategic successes — the iPhone, the Netflix Streaming Service, and many of Google’s innovations — stem from a process of parallel strategic exploration which I call Multiple Strategic Tracks (MuST). Back then it was a lot of launch-and-iterate and fail-fast (projects that didn’t work were killed, as it should be). Today we can probably be much more efficient using pre-launch product discovery. 

How Apple used multiple strategic tracks to arrive at the iPhone (source

There’s a lot more to strategy development than these examples suggest, but they do show how it can be a dynamic, multi-track process that is based on continuous learning and iteration.

Embracing Uncertainty In Goal-Setting

Earlier I said that optimizing for outcomes (what we want to achieve) is better when you face uncertainty because your goals are more likely to be right than your ideas . But even setting the right goals can be hard, perhaps because there are many potential goals to choose from, or because we’re not sure what’s possible. Good strategy should inform the goals, but many goals are tactical. 

Objectives and Key Results, with its top-down-bottom-up alignment process, can facilitate goal discovery. For example a manager can create a draft OKR for her area of responsibility — some objectives and key results may be firm, while others TBD. She then shares these with her reports and the OKRs cascade down and are processed. The manager then gets back draft OKRs from the various teams that reflect what her org thinks the goals ought to be — things she specified as well as things she didn’t. That’s a good way to learn of other potential goals she may have missed, as well as key results and target ranges. Andy Grove, renowned CEO of Intel and inventor of OKR, argued that 60% of key results should be defined bottom-up.

It’s also good to think of goals as having a lifecycle. Instead of jumping directly into execution, it’s often better to start with research and discovery, which gives you some room to understand the goal better, refine it, or even (in rare cases) give up on it entirely.

Example of a lifecycle of a goal (source)

Embracing Uncertainty in Roadmapping 

The classic 12-month roadmap is the most obvious example of false predictability, which is why many orgs replaced it with shorter versions like Now, Next, Later roadmaps, or Shape Up cycles. These options still demand certainty about what we build in the next 6-12 weeks, and both offer at best a vague longer-term plan. 

An alternative to consider is the Outcome Roadmap which uses goals as an underlying structure, and gradually fills-in ideas as we gain more certainty. This gives you the benefit of planning for the entire year, while committing only to those ideas we have good confidence about. 

An example of an outcomes-based roadmap (free template here)

Embracing Uncertainty in Idea Prioritization

Growth guru Sean Ellis took classic impact/effort analysis and made it much more uncertainty-ready by adding a third element — Confidence — which indicates how sure we are that the idea will have the predicted impact and effort. Ellis codified the popular ICE scoring method. But how do you estimate confidence in an idea? The scientific method suggests you should base your confidence on evidence. This is why I created the Confidence Mete that lists common types of evidence you may find and assigns them confidence weights.

The Confidence Meter. Download it as a free calculator here

ICE (and other prioritization models) are only good if you combine them with idea validation (aka experimentation). These build-measure-learn loops offer a powerful method to reduce uncertainty, risk, and waste, while accelerating value delivery (here’s a full example). 

Sidenote: Combating uncertainty in product development is the key theme of my book Evidence-Guided — Creating High-Impact Products in the Face of Uncertainty.) 

Final Thoughts

The need to deal well with uncertainty isn’t new. In many senses, Agile Development, User-Centered Design, Lean Startup, and Growth Hacking aimed to do just that and changed the way we work. Still, in a company that demands certainty, these modern approaches are stripped of their value and are often subverted to support the plans. To deliver and capture high value we have to be honest and accept uncertainty across the board, in company culture and practices. Not an easy transformation, but one that will pay off generously.

Join my newsletter to get articles like this 
plus exclusive eBooks and templates

Share with a friend or colleague