UX Brighton 2018 – Advancing Research
Last week was the annual UX Brighton conference, with a theme of ‘Advancing research’. Two of our Lead UX practitioners, Jake Kitching and Craig Williams, jumped at the chance at making the long journey down to attend. The conference is set in the impressive Brighton Dome Concert Hall and was a fitting venue for the calibre of speakers.
The event continues to grow with last week’s boasting just under 600 attendees with global representation as far as India and Canada. The event was a melting pot of freelancers, agency and in-house UXers which made for some great conversations.
There is plenty of talk about the UX industry maturing, but what about UX research? Danny Hope, the conference curator did a great job gathering a list of great speakers that shed light on this exact question under the collective subject matter of ‘Advancing research’.
Kate Towsey: ResearchOps
No one was more fitting to kick off the day’s proceedings than Kate Towsey who dove into the emerging community of ‘ReserachOps’. Kate began by introducing ResearchOps for those who haven’t been privy to the aggressively expanding online community.
Kate tells us that ResearchOps “focuses on the craft, but is not the craft”, it is not the methodology or strategy itself but aids the implementation. In short, it allows research and research leadership to focus on the day job and not be weighed down by the operational elements. This is common practice in neighbouring industries with the likes of: DevOps, DesignOps, DataOps etc.
But why now in research? Kate summed it up perfectly:
|Research is accepted as an integral part of design and delivery||+||In-house teams are growing and research leadership is being formed||+||More execs and management are looking at researchers for guidance and holding them responsible|
In keeping with the theme of the day, as UX design grows so does research, meaning bigger research teams and bigger to-do lists, however the resources and infrastructure often remain the same. ResearchOps sets out to change this by providing fluid working guidelines for those who need help organising infrastructure for research teams.
The online community has spawned a working group which has set the foundations of what ResearchOps could be. Kate closed the talk by encouraging the industry to take it forward and tweak it where necessary. An exciting opening talk which highlighted the equally exciting state of the UX research industry.
Read more here https://medium.com/researchops-community
Daniel Pidcock: The Power of Atomic UX Research
With the first talk exploring the operational side of research, Daniel Pidcock followed by exploring the business end and how to ensure research lives on in a “Searchable and Sharable” format. Daniel enlightened us by explaining his Atomic research structure.
The concept of atomic research was influenced by Brad Frost’s Atomic design model, whereby finished products can be broken down into pages, templates, features and then individual components (atoms). Atomic research is similar, where all experiments are broken down into their facts, the insights gathered from those facts and then the resulting conclusions.
Research is expensively found data and so it is important to find a way to organise and understand the outcomes of each project holistically, to minimise losing what you’ve learned or repeating research unnecessarily.
By viewing research data “atomically”, findings from across all experiments can be analysed side by side allowing us to identify the most valid insights that are related to but not reliant on their source. One fact can inform several insights, but if several facts from different experiments support that insight then you’re on the right track to more valid and stronger conclusions. This approach encourages evidence-based decisions whilst identifying the most important areas to re-test.
Dan highlighted how this atomic research approach allows for continual customisation by adding, amending and removing parts, as well as providing an easily shareable format; he even drew attention to a product of his, that’s still in development but provides a platform to store and distribute research.
Katy Arnold: Building a user research practice in the Home Office
Katy’s talk bridged the gap between the first two talks with explanation of how she built the Home Office research department from the ground up; exploring both the business and operational sides of user research.
The Home Office now has a 100+ strong team of UXers, but this wasn’t quite the case less than two years ago when they had a team of just two. Katy shared some key tips on how to get the most when resources are squeezed (something we highlighted as an issue earlier), to which the two stand out points for us were:
1. Go under the radar
Do things that are needed. Katy shared a great quote from Grace Hopper:
“If it is a good idea, go ahead and do it. It’s much easier to apologize than it is to get permission.”
2. Do great work
Katy presented a great case study of two graduate researchers being tasked to ensure that those with low digital skill or limited online access are included in the product design. They carried out a three month ethnographic study of both paper and digital touchpoints of the visa application process.
The insights from this sparked interest amongst a range of stakeholders and departments which worked as catalyst for further investment in research.
Georgia Rakusen: Internal UX
Georgia Rakusens’ talk diverted from the speakers so far to explore how the methodologies we use in research can be harnessed and used for other experiences in the workplace.
Georgia started by sharing three key benefits to deploying ‘internal UX research’ such as:
1. Improving business efficiencies
For example, internal UX research can be used to learn about and improve: internal systems and how colleagues communicate with one another. In other words it can be used for any colleague-colleague, colleague-system and colleague-customer interaction as well as anything in-between.
2. Shedding light on the under-represented UX research department
Educating those who are maybe less familiar with UX
3. Breaking down silos
Introducing new cross-team working
But how would this be implemented?
Georgia went on to explain that much of the internal research she worked on at Moo arose from a separate initiative called “Research hours”. Whilst at Moo, Georgia had specific times where she displayed a “Research Hours” sign on her desk to encourage designers and the wider UX team to come and tap her on the shoulder to discuss anything related to research. This backfired in an interesting way with other departments and teams coming for advice. Georgia shared three case studies, but our personal favourite was:
Exploring data requests that the data controller receives
The data controller approached Georgia as he was receiving lots of data requests from different teams with little knowledge of the exact information they would need and to what extent.
- Internal personas were created to help the data controller understand the end user when a request came in so he could serve their needs better.
- On top of the obvious benefit of making the data controller’s job easier and the end user being happy with their high quality, appropriate data, the study highlighted the value of qualitative insights in a quant-heavy department.
Katie Taylor: A true story of Usability Benchmarking
Benchmarking and tracking studies are a staple within market research but maybe less so in a qualitative dominated UX research landscape. Katie walked us through a case study of how to use usability benchmarking, highlighting some key things to consider for those less familiar with the approach:
Define the method which is to be used, this can be a mixed method approach:
- Remote vs in person usability sessions?
- Moderated vs unmoderated?
- Follow up data collection methods such as:
- Usability sessions followed by a post-task survey
- Survey followed by moderated sessions to deep-dive
The approach and question design needs to stand the test of time because the value of benchmarking is tracking year-on-year metrics so changing the approach in later years can impact the data being collected and make it difficult to compare like-for-like.
2. What metrics?
There are a number of usability metrics you can benchmark such as: task completion rates, journey or task abandon rates, average time taken etc. However, benchmarking can also be used for wider metrics outside of usability such as brand loyalty or trust; top task analysis can be used to understand what metrics need to be covered.
If we consider each round of benchmarking as a wave of research, tracking these wave-on-wave metrics helps to highlight impact to ROI and in turn, set goals for the future. For example, we want to lift x metric by y % over z amount of time.
3. Treat the first wave of research as an alpha
As you would with any usability study run a trial project, ensure the question asked and metrics measured work as required. If not, be prepared to change it as for the study to have value it needs to run 3-5+ years
Will Myddelton: Research Heresies
One of the most impactful sessions from the day was Will Myddeltons debunking some common research beliefs within the industry. To caveat, some of these beliefs are more of an issue within certain sectors than others, but are interesting none the less:
1. ‘User needs’ are really helpful?
Will explained how the phrase ‘User needs’ is not necessarily helpful and this is particularly an issue because of a lack of understanding amongst junior researchers (Will said this was more of an issue in The Home Office). User needs and user needs stories don’t need to be created for every function and button; understanding beliefs, wider contexts and emotions are equally as important as goals and tasks.
2. OMG! Let’s use research for everything!
As UX research has matured we have moved from not testing anything to testing everything. Will stated that we should only be testing things that matter and that areas such as payment pages and checkout pages have been researched enough so that we should know through best practice how to design these pages with little or no research. To help the audience get onboard, Will used the axis of a Wardly map to demonstrate when not to test.
‘Genesis’ is too early. It is blue sky thinking and doesn’t need research yet. A ‘Commodity’ is tried and tested or tried and researched and therefore requires no research.
3. User researchers must make recommendations
Will argued UX researchers in product teams (in-house) shouldn’t be making recommendations. He considers recommendations design solutions in disguise, and therefore step on the role of UX designer within the team, and restricts their design thinking.
Emma Boulton: Real World Discovery Research
Emma’s talk set out to discuss the art of discovery research with real world metaphors. One interesting take away from the talk and a theme that emerged throughout the day was taking inspiration from the market research industry to ‘advance’ the UX research industry.
Emma took us on a discovery journey using her scouts ‘Explorer Belt’ expedition as a metaphor for the journey you go on during formative research. For those of you who don’t know, the ‘Explorer Belt’ is a prestigious scouting award where you trek for a number of days in a foreign country and immerse yourself in the landscape and culture…sound familiar?
Emma wrapped up with 5 key takeaways which summarise the ‘journey’ excellently:
1. Discovery is an expedition
Get out the office and experience new things and meet new people. Immerse yourself in foreign territories and be prepared to get lost.
2. You are the expedition leader
The journey should be a team expedition, however you are the leader, so keep a log and report home throughout the journey.
3. Avoid the ‘Research Graveyard’
The journey is long and you can get lost amongst the data so bring along key stakeholders on the journey to increase buy-in.
4. Discover new methods
This is where Emma explained how we can learn from the market research industry.
Short term communities, recruiting from panels, using pre-tasks, projective techniques etc.
5. Discovery is a mindset
Emma’s final nugget involves a mind-shift; traditional models of discovery are heavily biased towards research being solely at the beginning of a project. However, she proposed discovery research should be deployed throughout the research cycle as needed sharing Jane Austin’s Double Diamond and her own Research funnel as a better working solution.
Double Dimond: Jane Austin – Moo.com
The Research Funnel: Emma Boulton
James Woudhuysen: Some ABCs of Forecasting
We unfortunately had to duck out of James’ talk to catch our train northwards. The talk was set to be an interesting discussion of everything forecasting in the UX industry, exploring some dos and don’ts when predicting how humans are likely to evolve.
Find out more about James and thinking about the future at: https://www.woudhuysen.com/
The day flew by and we left with a buzz and excitement to implement the day’s learnings into our research; we strongly recommend looking into these speakers further for all the useful knowledge they have to offer. Attending the UX Brighton conference was a first for SimpleUsability but will be a staple in the calendar moving forward.