Focus on: Opportunity space product discovery
What to do when the unknowns are plenty and the expectations are high.
A few weeks ago I kicked off a series of articles on the topic of Product Discovery by proposing that we categorise our discovery challenges from the start as Opportunity or Solution Discovery ones.
Catch up on this article here and let’s move on to talking about Opportunity Discovery in more detail.
🔎 Opportunity discovery overview
We’re doing Opportunity Discovery when we’re looking at a very unknown and, as such, unstructured space.
This is where we are discovering which user problem, or opportunity, is worth pursuing.
The really tricky bit of discovery in this space is that we’re dealing with an equation with so many unknown variables that an advanced calculus degree won’t help you make a start (I am now stuck with this metaphor which is unfortunate given that I don’t know basic maths and my terminology is all over the place, sorry).
The product risk focus in Opportunity Discovery is on customer value, business viability and technical feasibility and the three are in a constant state of flux — de-risking a top assumption in the value bucket can change the viability entirely, de-risking a top feasibility assumption may turn an aspect of the assumed value into a distant dream, and so on.
The outcome of Opportunity Discovery is having an opportunity, i.e. a user problem, that is worth us pursuing instead of all the other ones we could be going after because we got solid evidence that it has strong customer value, can be built within our technical landscape and fits our business goals accordingly.
👉 An important note on killing opportunities
The alternative but just as important outcome of Opportunity Discovery is the decision to scrap an area of opportunity altogether before investing any more in it because the sweet spot simply does not exist. Yes, finding out there’s no sweet spot is also a success because it spares you a time-consuming failure down the line.
Importantly, you shouldn’t get sidetracked by opportunity areas that do well on some types of risk but not at all on others. Oftentimes, the riskiest risk is unsurprisingly customer value — without customer value, arguments of the type “but it’s so quick and easy to build technically” can derail you massively. Keep your eye on the sweet spot. That’s your decision-making anchor.
Other times an opportunity space can have something in it but it’s not the best idea for right now. This can also distract you. Perhaps a sweet spot exists but where you found it doesn’t necessarily fit with the business’ overarching mission at this point in time. Perhaps you can build it with existing technology but the organisation is in the middle of taking a different technical direction. That’s OK and it’s OK to put that opportunity on ice. In a previous product team we used to call it Han Solo-ing an opportunity, referencing the time Solo got frozen in carbonite because George Lucas didn’t know if Harrison Ford would return for the next movie. We all find our way 🤷♀
So how do you go about conducting research and experimentation in the Opportunity Discovery space while remaining a relatively sane human being?
📘 Continuous discovery habits
Three years ago, in my days as a Senior User Researcher, my cross-functional product team and I had the honour of getting Continuous Discovery Habits coaching under the watchful eye of Teresa Torres herself.
Perhaps many of you have already read Teresa’s best-selling book “Continuous Discovery Habits” and I recommend familiarising yourself with the framework to those who haven’t.
I’m not going to rehash its contents but what I will do is highlight the key techniques I’ve found immensely valuable in my Opportunity Discovery practice and share some lessons I’ve learned along the way. At the end, I will explain why I’m veering away from the framework when I split Opportunity and Solution Discovery and how that’s been influenced by challenges “on the ground”.
Remember, folks — finding your discovery groove is a non-linear expedition with twists and turns, and many, many hurdles.
🌳 The Opportunity Tree: structuring the unstructured space
When my team was first faced with an Opportunity Discovery type challenge, we didn’t think twice before jumping on it in the same way we had successfully delivered results in Solution spaces before. If it ain’t broke, right?
We were looking at the possibility of attracting a potentially desirable customer audience (first unknown: is this a good customer for us at all?) we weren’t currently serving (second unknown: how do they behave and what do they need?) and trying to understand how and if we could create a proposition for them (third unknown: would what we learn about their needs translate into a viable direction for our business?) within our capabilities as a business (fourth unknown: would building anything in this space be technically feasible for us?).
We quickly started to struggle — we were slow, we went around in circles, we were overthinking because the more we learned the more questions it prompted, and whatever we learned was hard to place in the bigger picture. At the end of months — months! — of discovery work we had come to the conclusion that this is not a good opportunity for us right now.
Learning that something isn’t a good opportunity is a type of success but we were deflated because of the amount of time and effort we had put into this One Thing. We were reflecting back on whether we had actually reached this conclusion earlier but didn’t have the skill — or courage! — to recognize and call it. Moreover, we were stumped as to what to look at next because we hadn’t entertained any other opportunities while we were focusing on this one. This was not discovery at pace and if it was going to take us 3–4 months to find out that something isn’t a good problem to solve, we weren’t going to get far.
Cue the Opportunity Solution Tree.
The Opportunity Solution Tree or, as we started calling it for short, the OpportuniTree (I hope Teresa allows it…) provided us with a way of visualising and keeping track of all of our assumptions and learnings, and having a continuous shared view of the big picture and how it’s changing and evolving.
Because everything we learned was added to the tree in the form of more opportunity leaves, finding out that one opportunity wasn’t good enough meant that we had the next best candidate ready for us. This enabled us to move quicker because killing one opportunity didn’t mean that we didn’t know what to do next anymore.
💡 Critical shifts in mindset
- Solutionizing as a way to test an opportunity
In our established discovery practice, we had it in our heads not to solutionize too early. In contrast, this framework encourages it — except solutions here are merely testing tools to find out as quickly as possible whether there’s meat on an opportunity or not.
This is solutionizing that is not about the solutions at all. It’s a hard one to get your head around at first but it makes all the difference. The pace at which we found out which opportunities had legs and which didn’t increased immeasurably after applying this technique and getting more comfortable with it.
- Compare and contrast
A technique which underpins the solutions-as-ways-to-test-the-opportunity philosophy is comparing and contrasting three solutions against each other at all times.
The more different the solutions, the more directions you’re attacking the opportunity from — the more good evidence you’re getting whether the opportunity is one that’s worth investing in solving for or not.
- Evidence vs proof
Compare and contrast also shifted our thinking from trying to definitively prove or disprove something via our research and experimentation to trying to collect just enough evidence to tell us whether to bother collecting any more data in this space later on at all.
A particularly enjoyable example was faking the start of a live chat experience by hacking our website survey tool. It took three days to put together, run and analyse and know whether it’s worth looking into this space any more. We didn’t have to spend weeks building a full live chat experience to see if our customers would buy via live chat even though this is what we were ultimately after. We only needed to know if they’d engage with it because if they didn’t do that as a bare minimum then expecting them to buy through this channel was a moot point. And we tested this against two other solutions to see if our assumption about the core user problem/opportunity was strong or not.
- Two-way door decisions
The concept of two-way door decisions first made an appearance in a shareholder letter by Jeff Bezos (suspend your feelings towards him for a second) and refers to the type of decision that is easy to reverse. You walk through the door, you see what’s on the other side, you make a call, you go back if it’s not what you need. You have to treat opportunity assessment as a two-way door decision, and a very quick and emotionally detached one as well.
The combination of solutionising, compare-and-contrasting, evidencing and having a two-way door mindset means that we are not spending too much time on an opportunity, or worse on a solution, and are protecting ourselves from that pesky escalation of commitment bias. No one wants to feel like they’re working on the next Fyre festival, right?
Moreover, we’re keeping it real when it comes to Marty Cagan’s first inconvenient truth about product — namely, that more than half of your ideas are going to fail. Fail away with me, honey, but do it in the research and experimentation phase — not after launch.
- The Product Discovery Unit FKA Product Trio
In the original framework, the “product trio” consists of a product manager, a designer and an engineer.
My team started calling ourselves a Product Discovery Unit or PDU (not to be confused with Public Displays of Underpants) for short and who doesn’t like a bit of personal branding amidst the chaos? Take it, use it.
In my experience so far, depending on your resourcing privileges and the starting point of the outcome at hand, you may have a different mix of people. Some teams I’ve been in or have observed have a researcher and/or a data analyst but the biggest addition I’ve found very valuable, in some problem spaces particularly more than in others, is to have a dedicated Subject Matter Expert as part of this core unit.
Which leads me to…
🎿 Where and why I go off piste
- Separating Opportunity and Solution Discovery
Many paragraphs ago I said that finding your discovery groove is a non-linear expedition with twists and turns, and many, many hurdles. It’s been precisely the hurdles that led me to make the half-strategic, half-practical decision to separate Opportunity and Solution Discovery for our organisation.
I’ll raise my hand in admitting that I’ve made the mistake of getting way too idealistic and evangelical about the Continuous Discovery Habits framework. It’s a common pattern in an industry where a new framework is invented every hour to fall in love with an approach, especially one that makes so much sense, and use it as the hammer for every nail.
It hasn’t been my first line of attack but Splitting Opportunity and Solution Discovery has been the most digestible approach so far on a more organisational rather than team level. It required me to put my researcher hat on once again — understanding where we lack the common language and why, how unclear goals and misaligned expectations make communication ineffective and success criteria ever-changing, how aspects of the business I’d never had visibility on actually work, and ultimately how categorically different types of problems were being treated quite similarly and all under a very vaguely defined “discovery” umbrella.
Being able to bring to life how we’ve been going into Solution Discovery before having Opportunity evidence with tangible examples, how a real user problem doesn’t always translate into a good opportunity, how a variety of our current challenges can be related to the appropriate bucket and seen through a more actionable lens… all of it was in our zone of proximal development on an organisational level and proved to be the most effective way to move forward.
- Having a dedicated business Subject Matter Expert in the Product Discovery Unit
Having a business SME person in the unit has meant that the business viability risks and constraints are not only always represented but that we get a very deep understanding of them on-the-go and move accordingly. It’s also allowed for a better conversation in the constantly changing balance between customer, commercial and tech in the discovery process.
This is especially important if the industry you’re working in is very specialised, regulated and complicated — I’m coming from the experience of someone who’s been in financial services for years. Expecting the Product Manager, User Researcher or UX Designer to learn the ins and outs of it to the extent that an industry specialist does not only slows you down but also makes you bump your head against avoidable walls.
I’ve written before about the importance of user researchers understanding the business in order to do good work and working in a PDU is a fantastic way to elevate this knowledge and practice.
- Still doing in-depth generative research where appropriate
Before you accuse me of sticking to my user research guns and being in love with in-depth research studies, this is why I say where appropriate. Trust me, I spend more time in my day explaining to people why they don’t need in-depth research studies than the opposite.
However, in some cases we can’t not invest in a kick off round of exploratory research in order to set us on a reasonable path when it comes to our foundational understanding of the opportunities on the table. That’s not to say that we won’t continuously feed our understanding of the opportunities with lighter touch research and experimentation but conducting the primary research can tame and frame what could be a very, very wild Opportunity Tree starting point.
If you’re starting from knowing absolutely nothing about the customer in a given problem space, I will always see the value of in-depth qualitative research and recommend it wholeheartedly. (And I will talk more about the skills and toolkit of the user researcher working in modern product discovery in a future article.)
- Struggles in the optimisation space
I’m putting optimization in the Solution Discovery bucket for which I still think there’s space for double diamond-ing.
I’ve tried bringing the Continuous Discovery framework to the day-to-day optimisation space, to a variety of teams and problems, and I’ve found that stealing a few of its components and applying them to something more familiar like the double diamond is a more effective approach.
But this is already a very long article so more on the last topic next time!
In the next and final article from the Product Discovery series, I’ll focus on Solution Space Discovery.
If you don’t want to miss it, hit the follow button. In the meantime, let me know how you approach discovery in unknown, unstructured spaces in the comments!
Don’t forget to read “Continuous Discovery Habits” and subscribe to the Product Talk blog.
Edit with links to the rest of the articles:
Part 1— Get better at Product Discovery by defining your starting point
