I cringe every time I see product teams use a spreadsheet to rank the ideas in their backlog based on some made-up math formula usually consisting of things like business value, user value, and technical difficulty.
While this exercise is pervasive, it misses the point entirely. Our job is not to prioritize solutions. A product team’s job is to create value for the customer in a way that creates value for the business. This is rarely done by fixating on a ranked idea list.
A product team’s job is to create value for the customer in a way that creates value for the business. This is rarely done by fixating on a ranked idea list. – Tweet This
You aren’t one or two or three features away from a better product. This is the fallacy of the feature factory or as Melissa Perri calls it “The Build Trap.”
Prioritizing solutions is a left-over side effect of being output focused. When we are judged by what we deliver, the key decisions are focused on what to build when. But when we are judged by what outcomes we drive, it’s less about what solutions we deliver and more about what problems we solve for our customers.
Product teams should stop prioritizing solutions and instead prioritize opportunities.
Collect Stories in Customer Interviews to Generate Opportunities
Opportunities are customer needs, pain points, desires, wants—they are chances for us to intervene in a way that makes our customers’ lives better. You can’t prioritize opportunities if you don’t know what they are.
Thanks to tools like Ash Maurya’s Lean Canvas and Jeff Patton’s Opportunity Canvas, more teams are talking about opportunities. However, we are talking about an opportunity in the context of a solution. We are starting with a solution and asking, “What problem does this solve?” This is backwards.
Instead, we want to start with our desired outcome and ask which customer needs, pain points, desires, and wants—if addressed—would drive our desired outcome. The best way to answer this question is through customer interviews.
For example, if I work at Netflix and my desired outcome is to increase the number of viewing hours per viewer, I might interview Netflix users and ask, “Tell me about the last time you watched Netflix.” In their story, I’m listening for needs, pain points, desires, wants—opportunities that if we addressed them may lead to more viewing hours.
I like to frame opportunities as something a customer might say. This helps to ensure that I’m solving a real customer need and not a business need disguised as a customer need. For example, a customer might say, “I couldn’t find anything to watch,” but they aren’t likely to say, “I wish I binge-watched Netflix more.” The former is an opportunity that if addressed would lead to more viewing hours. The latter is something we wish our customers might say, but isn’t very likely.
I like to frame opportunities as something a customer might say. This helps to ensure that I’m solving a real customer need and not a business need disguised as a customer need. – Tweet This
Even though we might want our customers to spend more money with us, no customer would say, “I wish I spent more money on groceries.” However, they might say, “I’m looking for new ideas for dinner,” which if we did a good job addressing this need, may very well lead to the customer spending more money.
If we focus on collecting stories in our interviews, opportunities will emerge from those stories.
If we focus on collecting stories in our customer interviews, opportunities will emerge from those stories. – Tweet This
Map out the Opportunity Space Using an Opportunity Solution Tree
It can be hard to prioritize a flat list of opportunities, because odds are your opportunities are of different shapes and sizes, some are interrelated, others are subsets of others.
For example, if I have the following list of opportunities:
- I can’t find something to watch.
- I’m out of episodes of my favorite show.
- I can’t figure out how to search for a specific show.
- I don’t know when a new season is available.
- The show I was watching is no longer available.
- I fell asleep and several episodes kept playing.
- I want to watch my shows on my flight.
- I want to skip the show intro.
- Is this show any good?
- I want to know what my friends are watching.
- Who is that actor?
- I want to watch on my train commute.
I don’t know how to compare “I can’t find something to watch” with “I’m out of episodes of my favorite show.” These opportunities are not distinct. Running out of episodes of your favorite show is a reason why you might not have something to watch. But it’s not the only reason, so these aren’t exactly the same either.
To simplify this process, start by grouping related opportunities using an opportunity solution tree. Reframe similar opportunities to make sure they are distinct from each other. That might look like this:
This isn’t a perfect science. There is more than one way to group the same set of opportunities.
Two ideas that you want to keep in mind are similarity and distinctness. You want to group similar opportunities together, but you also want to make sure they are distinct.
“I want to skip the intro,” “I fell asleep while watching,” and “Who is that actor?” are all similar in that they all happen while watching an episode. They are also all distinct. There’s no overlap between one opportunity and the next.
As an example of how subjective this exercise can be, take a look at “I can’t find something to watch” and “I don’t want to miss out on something good.” Are these distinct? They are similar as they both involve deciding what to watch, but they have different connotations. I could make an argument that “I want to know what my friends are watching” could belong under either one of them. So what’s the right answer?
There isn’t one. The goal is to find the grouping that best reflects how your customers think about the opportunities. If your customers want to know what their friends are watching when they don’t have anything to watch, then it belongs under that branch. But if they also want to know what their friends are watching, because of a fear of missing out, then it might appear under both branches. If too many sub-opportunities belong to two of the same parents, we might need to question whether those parents are truly distinct.
The key when mapping your opportunities using the opportunity solution tree is to play with different groupings. Different groupings will help you understand the opportunity space differently. Some will make more sense than others. But each will give you new insight into what you know about your customers.
This exercise can be deceptively challenging. It’s really a critical thinking exercise. It forces you to question what you know about your customers and you’ll uncover a lot of gaps. That can be uncomfortable, but it will help inform what you need to learn next.
Prioritize Opportunities Using the Tree Structure to Guide Your Decisions
While grouping and framing your opportunities can be challenging, it’s worth the effort because we can now use the tree structure to help simplify our prioritization process.
We don’t need to assess every opportunity on our tree. We can simply assess the first row. We can start by assessing our four top-level opportunities:
- I don’t know what to watch.
- I don’t want to miss out on something good.
- I want a better viewing experience.
- I want to watch where I want.
Once we’ve identified our top priority in the first row, we can ignore the other branches of the tree, and move to prioritizing just that opportunity’s sub-opportunities.
For example, if we decide “I don’t know what to watch” is our top priority, we can ignore the other three branches, and simply assess and prioritize “I can’t figure out how to search for a specific show,” “I’m out of episodes of my favorite show,” “The show I was watching is no longer available,” and “Is this show any good?”
Assess Opportunities Using Fast Two-Way Door Decisions
I like to assess opportunities by sizing each opportunity (i.e. how many customers are impacted by it, how often), evaluating how addressing that opportunity might affect our position in the market, how well an opportunity fits with our company vision, mission, and strategic objectives, and how important it is to our customers. Ultimately, we want to estimate how much impact addressing each opportunity would have on our desired outcome.
This too is not a perfect science. These factors will not all agree. One opportunity might win on one dimension and lose on another. You might be tempted to develop another made-up math formula to help you make these decisions. But I don’t recommend that.
These are subjective decisions. The more we quantify them, the more likely we are to treat them as truth, and we don’t want to do that. We want to stay open to the idea that we might have made the wrong decisions. These are messy, subjective decisions and that’s okay.
When assessing opportunities, we want to stay open to the idea that we might have made the wrong decisions. These are messy, subjective decisions and that’s okay. – Tweet This
Jeff Bezos in his 2016 Amazon shareholder letter, introduced the idea of Level 1 and Level 2 decisions. He defines a Level 1 decision as a decision that is hard to reverse. I like to think about Level 1 decisions as one-way door decisions, because once you’ve walked through the door, you can’t turn around and come back through. With Level 1 or one-way door decisions, you want to take your time, collect data, and make a cautious decision.
Level 2 decisions, on the other hand, are easy to reverse. Think about them as two-way door decisions. Once you walk through the door and see what’s on the other side, if you don’t like it, you can always turn around and come back through. For Level 2 or two-way door decisions, you want to move fast, you don’t want to wait for perfect data, because you’ll learn more by acting than by waiting.
Opportunity assessment and prioritization decisions are “two-way door” decisions: You can always turn around and walk back through the door. You want to move fast because you’ll learn more by acting than waiting. – Tweet This
Your opportunity assessment and prioritization decisions are two-way door decisions. Once you choose a target opportunity, you’ll test whether or not you made the right decisions by prototyping and experimenting with solutions that address those opportunities. If you learn through experimentation that you didn’t choose the best opportunity, you can always walk back up the tree and choose another opportunity.
Product Strategy Happens in the Opportunity Space
Product strategy is the result of these series of assessment and prioritization decisions. The opportunities we choose to go after are what distinguish us from competitors. It’s what allows us to carve out a unique space in the market.
The opportunity solution tree helps to ensure that you are considering enough opportunities. It helps you to group similar but distinct opportunities, simplifying your prioritization decisions.
Give it a try and let me know how it goes.
If you’d like more articles like this, subscribe to the Product Talk mailing list.
David Street says
Interesting but maybe this depends on the type of products (services) you are managing. I can assure you in some businesses, things like regulatory compliance or correction of possible safety issues will prioritize on a list. Second are value add to customer and profitability to the business. You’re just picking out the “fun stuff”. In the real world, many of have to deal with the reality of non-negotiable priorities before we get to the fun stuff.
Teresa Torres says
Thanks for your comment. I’m not sure I understand your concern. I’ve worked with many teams in regulated contexts and / or have to account for safety concerns using this model. Nowhere does this model preclude prioritizing those types of activities.
Nothing wrong here, but there’s a strawman in the thesis on trashing ranked lists.
1) A ranked list is just one of several communication tools to help other orgs understand your prioritization approach.
2) Assessing a ranked list of things versus a row of opportunities is basically the same thing if the team has done the tablestakes homework of customer interviews, market analysis, JTBD, etc. that links opportunity with broad solution.
It all boils down to how to prioritize and then articulate the prioritization so people trust you.
Ultimately you need a sequenced plan and it needs to cover the problem/solution combo to get an actual outcome.
Teresa Torres says
I’m not trashing ranked lists. I’m trashing ranking solutions across a set of opportunities. When we rank solutions across a set of opportunities, we miss the more important decision of which opportunities matter most. And yes, you may rank your opportunities when making this decision. I’m not suggesting there is anything wrong with that.
David Martinez says
This is a really nice distinction, prioritizing by opportunity rather than feature makes sense…
John S says
what happens after opportunity has been identified? eventually you’re assessing different solutions that best address the opportunities.
I like the article makes it explicit to pay attention to opportunity, challenge the initial assumptions (I find team often think they know what the opportunity they need to address, because validating the need takes time and is uncomfortable because you’re not producing anything) and not just solutions, but having a ranked list of potential solutions is also needed
Teresa Torres says
Generating a good set of potential solutions is needed. However, I don’t recommend ranking them the way that most teams do. Ranking solutions based on criteria like business value, user value, and feasibility is rife with untested assumptions. The best way to evaluate solutions is to experiment and prototype. Generally, you should be working with 3+ solutions at once, comparing and contrasting which looks best at any moment in time. As you learn which looks most promising, you deliver it. As you learn how well it performed, you either work with your next set of solutions or move on to your next target opportunity. If you have a long list of ranked solutions, it means you have too much work in progress and are getting ahead of what you actually know.
Guido Lonetti says
Might not be related to the essence of this article.. but by reading all of your posts, I was wondering: how do you usually organize Product teams towards Product Discovery (on a scrum basis)?
I assume that Dual track Agile might be a good practice, by having parallel discovery and delivery tracks.
Teresa Torres says
I like to see the same team doing discovery and delivery, so that you don’t have hand-offs between the two. The trio typically leads discovery. See this article / video for more details: https://www.producttalk.org/2018/10/continuous-discovery-mindsets/
Julien PAFFUMI says
In your article you decide (as an example) that your top priority is “I don’t know what to watch”, so you address its different sub-opportunites, ignoring the other branches.
Should you cover ALL sub-opportunities before moving on to another opportunity?
Probably not, or at least not always.
Then how do you know how far you should go in this specific opportunity, before considering again the other branches?
Teresa Torres says
Great question. The answer depends on the situation. However, here’s the guiding principle to keep in mind. You always want to work on the opportunity that will have the highest impact on your desired outcome. If you make enough progress on one sub-opportunity, that may address the parent opportunity enough, that you may have a higher impact by moving to another branch. Other times the next highest impact opportunity is another sub-opportunity. I recommend redoing the assessing and prioritizing exercise whenever you think you’ve made enough progress on a given opportunity. And “enough progress” should be defined through testing and iteration.
Greg Bragg says
Yes, Yes, Yes – but how to convince your engineer-led ‘product’ organization to get away from their feature-function, stack-rank, process?
Teresa Torres says
Instrument your product. Track how often you are building the wrong stuff.
Nate Archer says
The desired Outcome you mention in the Netflix example is very much in the view point of the company and not the user. Is this just the way you see it framed most often? How do you ensure that you are driving toward an outcome that actually resonates with users?
Randall Gibson says
I’m a big fan of this reframe from outcomes and outputs, then using the opportunity tree to prioritize opportunities. It helps orient ourselves towards the user and customer needs instead of talking about ideas, which is what we all love to do. This strategy is a solution to the problem of how we are solving problems (my 1-min read on this problem t.ly/z7OeP) and I’m going to see how it works.
As always, I sincerely appreciate your knowledge sharing.
What kind of prioritisation framework do you use to prioritise “opportunities” and “solutions”? Obviously you can’t use RICE model to prioritise opportunities because there is no solution yet. Are you able to share some examples with me?
Teresa Torres says
Check out the first paragraph under “Assess Opportunities Using Two-Way Door Decisions.” That’s the best public summary I have of how I assess opportunities and then I use that assessment to prioritize.