Dropping "recommendations" in favour of "How Might We" challenges to facilitate better outcomes


In the past, it’s been pretty standard practice for researchers to include "recommendations" as part of a debrief – often being included as a placeholder slide in a template debrief deck. I’ve been lucky to work with many practitioners who specialise in research, and I always appreciate (and encourage) researchers not to include "recommendations". I find that, for many reasons, they just don’t work.

Why they don’t work

1. They tend to not address root causes of problems

I think the most important job of a research specialist is to help teams and stakeholders gain a deep understanding of the people that use and are affected by their products and services. It is imperative to robustly establish the root cause of a problem and why something is happening. Focusing on how a problem should be addressed detracts from understanding the root cause of a problem, and often resulting in a symptom of a problem or the wrong problem being addressed.

2. They discourage collaboration

I think it’s important to collaborate with people to create solutions that are technically feasible, meet user needs and achieve business goals. This should include all representatives from all disciplines that are available within the team – for example, research, design, product, content, development – along with stakeholders who have specific interests in what is delivered, along with a deep understanding of problems and contexts. Making "recommendations" potentially skips the collaborative process entirely, often resulting in a solution that isn’t as good as it could be.

3. They can create unnecessary tension

Not only do "recommendations" miss the opportunity to utilise insight from other team members and stakeholders, it’s also not very respectful to them. This can create unnecessary tension and frustrate other team members and stakeholders who feel they are not getting opportunities to express themselves.

Present "How Might We" challenges instead

I think that good research promotes understanding and encourages debate, establishing the root cause of problems within the wider context of the end-to-end journey.

From there, every problem can become an opportunity for design by being framed as a "How Might We" challenge. For example, "a thing we have learned" becomes "How Might We address a thing we have learned".

The "How Might We" challenge centres on a problem and suggests that a solution exists for that problem. It also gives a clear frame to encourage collaboration and innovative ideas without suggesting a particular solution.

While working on "Book a Secure Move" I worked closely with a research specialist to develop "How Might We" challenges based on insights from our research. We prioritised these "How Might We" challenges with stakeholders across the criminal justice system, promoting stakeholder alignment and clear, agreed focus for the delivery team for the next phase of work. I then used these prioritised "How Might We" challenges to design and facilitate a series of design workshops with the team to generate creative yet feasible solutions, that focused on solving the right problems for our users.

Read more of my thoughts

Fancy a chin wag?

Whether you are just looking to discuss opportunities, looking for help starting out, or want more information - get in touch and I’ll be sure to get back to you

Get in touch at andrew.r.millar@gmail.com