ao

Pros, cons, and favourites.

Originally posted on Ghost as 10 Versions of UX Research Documentation.  Written in January 2022, as a work reflection.

In the past 10-15 years, I’ve also probably tried around 10-15 different forms of UX research documentation and dissemination.  Here’s a rundown of what worked and didn’t.

In order of what-I-keep-coming-back-to:

1. The product-planning debrief

Made for: Small product-focused working teams — that value action and nuance over formal reporting.

In other words, this needs a research-mature team: that understands how to connect observed behavior, to needs, to priorities, to feasibility – and when to think of which.

Bad for: Future-proofing. 

What I’d do better: Detailed minutes.

• Notes + Miro/ Post-its + Wireframing -> only understandable to the people present

I can also call it: Top 3 takeaways

Since, it’s most often a debrief that I run after small usability tests (5-10 users; 5-10 tasks  each) – where the team gets together and contributes/ writes down their  observations. (Example of this process: Featured image of this post)

2. The customer journey mapping session

What I use most often as a failsafe for discovery research, particularly related to (digital) service improvement (which has been my work for the past decade).

This hits the key information you need ->

• which touchpoints when,
• what info is important,
• what decisions are being made,
• what sub-actions are being done.

It’s the fastest route to knowing what needs to be fixed.

Not fit for non-discovery phases. So, if the team’s in evaluative method mode, then don’t do this.

Also, this takes a bit of research (i.e. time).

3a and 3b. The full-blown usability results “table”

I have at least 2 forms of this table (depending on my teammates’/audience’s need for detail).

Primarily, it’s a summary of the key test differentiators.

A. One version I have is a high-level summary – that shows either aggregated task success rate, or average completion time for – various tasks, screen sizes, competing products tested, demographics or technographics – depending on the goal of the test.

B. This is more straightforward – just a summary table of the same (task success, completion time or pain points) per participant (with profile/description).

Obviously, I mainly use these for usability testing. Summaries this direct don’t apply to discovery research (unless forced to. Which I also experienced once in my life).

4. Monthly User-data-from-different-channels summary (triangulation)

This one I love, but keep forgetting – only routinely did once/at one job. I attempted to implement at another team, but failed at doing it collaboratively/scaling it so that others could apply.

I summarize key learnings from: Product analytics, Internal company information/imperative, Ad hoc/Qualitative research and social media/customer service feedback.

This is connected to product plans/feature changes, and sent out to the team. I love what this represents – connecting multiple streams of customer findings to org plans.

Cons:

• This is already quite processed, so has hidden pre-work/data clean-up, that isn’t visible or apparent in the final output.
• Monthly is actually not “frequent” enough for most cases, but this was the sustainable pace, for me (we were in a very lean team).

I think if I was able to systematize this, it would be awesome.

5. Key Drivers analysis

I realize that my brain defaults to this for qual – and for good reason.  I think it’s still a remnant from my market research days – where the foundational piece of information was choice drivers.

What this has: Hygiene factors VS Differentiators

I love making this analysis, because the analysis steps are so fun.  The final outcome is rewarding, too.

Obviously, only applies to discovery research – anything whose goal is determining preference “why would someone buy/choose/use”.

Cons:

• There needs to be “thick enough” data to make this analysis – and you can only do this using data from in-depth interviews/actual detailed conversations on why people do what they do.  It can’t be inferred.

6. Pure artifacts and images + stories = Story-sharing

This is actually my favorite – if I could do it all the time.

It takes a team that is open-minded, or at the very least aware of the value of qualitative information.

i.e. This does not work for teams and individuals hard-wired to value numbers.

Also in other words, you really can’t use it all the time – unless you are/have a good evangelist.

Back to the Pro’s:

• This is really it, for me.  The epitome of the value of observational, behavioral design research.  Seeing “the how” and the nuance.
• Great for “innovation” – finding the unspoken, intuitive considerations people have when they do something.  It isn’t something you can always explain.
• The other “information” you get from it is “three-dimensional awareness, or contextual awareness – it’s a different type of imagination boost.  Which is really the spark you need for solving problems creatively or sharply.

7.  The full report

I don’t particularly enjoy this, but every long-term job of mine has had this at some point.

Ingredients: background and methodology, rundown of answers to all the goals of the study: can be behavioral archetypes, drivers, a journey and conclusion.

Essentially, it’s a long document describing who the userbase is (based on the participants), motivations, options and decision factors, lifestyle, service or user experience and recommendations or implications.

Pro: Comprehensiveness

Con:

• I have never made one in faster than 10 working days – due to the data processing and clean-up work, analysis and actual writing and report-crafting (models, etc.) phase.
• Time + makes research seem like a static artifact.

8. The database: “Atomic” research repository

You know, I want this to work.  I really do.  I know this is supposed to be THE way.

9. Key quotes per theme/ finding

10. The “navigation bar” summary with participant photos

Quite involved; not sure how applicable this is to any discovery research project. Lends itself well to research goals around “scales” or gradients.

Additional entry: The one-pager format:  Created by my design research team member Vikki.

This was a handy summary (most would probably say it’s an executive summary). [I do not have an excerpt of this, as this was my friend’s output.]

Bonus round: Least favorite format ⬇️

11. Persona document

Among the ~5 times I’ve done “personas” as a research output (and I usually try to follow Indi Young’s “Thinking Styles”, as well as the original Cooper process for personas), the format I found most impactful was the first (very detailed, by-the-book). [Don’t have a copy of that anymore, I believe.]
This image below was, instead, from the latest research cycle where I used personas. The presentation favoured a more concise description, but I did go in-depth for the needs discussion. The brief descriptions were better fit for this project, because there were very few participants (even for a qualitative research cycle).