The State of the Discovery Phase — Takeaways from UKGovCamp18 and OneTeamGov
We work across sectors, doing product discovery work, but also helping teams adopt best practice through coaching and training.
We are particularly in interested in the public sector and the way that the Discovery and Alpha phases are working in the development of digitally enabled government services in the UK.
Here’s what we learned:
Discovery remains undervalued and misunderstood by many stakeholders and sponsors
Again and again, we heard about situations where senior people had pre-conceived ideas and just wanted to get on with building them.
Discovery in this context is an exercise in validation rather than exploration.
It means asking closed or evaluative questions rather than open, generative ones. It means skimping on time and people. It makes the whole process a box ticking exercise. And it means you may miss out on the opportunity to make a big difference.
If you set out to validate, you won’t learn what you don’t know. What you don’t know is the thing that will ultimately make your project fail.
Would this happen if Discovery phases were subject to a GDS service assessment? That’s worth exploring.
It’s not always clear when Discovery ends and Alpha begins
There was some uncertainty on what the outcomes and outputs from the Discovery phase should be.
In a culture that values stuff getting built, there appears to be a tendency to start prototyping too early and to narrow focus too soon.
Discovery should be about clearly articulating the problem (user needs, organisational goals and the hard constraints on any solution) with supporting evidence. And then defining and scoping the nature of a potential solution, and identifying important assumptions and ideas to be tested in Alpha.
As Matt Knight pointed out at OneTeamGov Wales — service assessors want to see teams look at a range of solutions in Alpha before settling on one.
But as someone (if it was you let me know) said in our UKGovCamp session
“It’s hard to put a value on the procurement of ‘understand the problem.”
People, time, focus = success factors
When we talked about what went well, the same ingredients kept cropping up:
- Get the right people involved — everyone agreed that we need a multi-functional team to do discovery well, including policy, technology, subject matter experts, user research and delivery management. The range of perspectives creates a ‘collective intelligence’ that’s greater than the sum of its parts. Not everyone needs to be involved in every activity, but the team needs to work as a unit, with everyone playing their role and communicating effectively.
- Allow enough time to consider options — there needs to be room in the schedule to relax for a moment, take a step back and look at what we’ve learned with fresh eyes. One post-it said “Allow time and brain space for serendipity”, this is true. Decisions in discovery have a disproportionate impact on the end product or service later. It’s worth sleeping on it.
- Focus — we need to stay open during discovery. But we need research questions to answer. And the answers to those questions should help us envisage services. And those services should bring about the outcomes we seek. So what are those outcomes and how are they measured? Is it reduced wait times? More eligible people getting a benefit? Better patient care? It’s the ‘why we’re here’ bit. And it needs to be defined upfront to give the team a focus for everything they do.
Exposure to users builds empathy, which helps a lot
We are always enthusing about this, but it was good to hear stories reflecting how bringing stakeholders and team members from different disciplines into direct contact with users helps to change their mindset.
Using quotes and videos in show & tells, and inviting folks along to user research sessions are obvious ways to do this. It’s Discovery’s equivalent of ‘Show the thing’.
So there we have it. A summary of all we heard. Lots to chew on. We will be thinking and blogging ideas for solving the bad bits and for amplifying the good bits as they emerge. We’d love to hear further thoughts from the session participants and any interested parties.
Useful discovery links
- Our 10 tips for discovery success
- Sarah Prag’s Discovering Discovery
- Will Myddelton’s Three ways to run better discoveries
We are also running a training course on How to do Discovery on Feb 28th in London.