UX Researcher & CRO Specialist · UK / Remote

Research didn't change how I think. It gave a name to how I already did.

I was watching how people behaved in products, communities and games long before anyone paid me to. UX gave me the methods to investigate properly — but the habit came first, and it doesn't switch off at six o'clock.

Currently Open to research & CRO roles
Based UK · Remote or hybrid

Open questions I keep returning to

  • Why do people keep using products they complain about?
  • What turns a community transactional?
  • What actually makes someone feel like they belong?
Mark Jumbo
Mark Jumbo UX Researcher & CRO Specialist

ContextsSocial care · E-commerce · Media & advertising · Consumer tech

WHERE IT CLICKED

A Discord server taught me more about belonging than any framework.

I met Discord the way most people in media do — as another workplace chat tool. Then my partner's gaming server pulled me into a completely different version of it.

What I noticed

People arrived wanting to network. Most stayed because they wanted connection. Those aren't the same thing, and the gap between them explained almost everything going wrong.

The symptoms

Members promoting constantly and building no relationships. Conversations turning transactional. Channels going quiet.

The wrong question

"How do we grow this?" — growing it would only have added more people to the same dynamic.

The questions I asked instead

Why aren't people getting what they came here for? What behaviours create belonging? How do you let people with different intentions share one space?

What I changed

Rebuilt the server around motivations rather than topics — room to promote, room to just play, room for creative work, room for support, room for the film nights that mattered more than any growth tactic.

What changed

People came back more often. Conversations loosened up. Promotion became a by-product of the community rather than its purpose — and members started describing it as somewhere they belonged.

Nothing about that is different from product work. Same questions, same evidence, no research budget.

Community server structure, blurred for privacy
The rebuilt structure Channels grouped by why someone is there — promotion, play, creative work, support — rather than by topic. Member names and conversations are cropped out.

Philosophy

I don't optimise interfaces. I study environments where people interact.

Products, communities, games, creative spaces. The questions barely change: how do people behave here, what motivates them, where does confidence disappear, what makes them stay, what makes them leave. The interface changes constantly. Human behaviour rarely does.

AN UNDERRATED RESEARCH ENVIRONMENT

Gaming communities are the largest unmoderated usability study running.

Millions of players describing, in public and in detail, exactly what frustrates them, where they stopped playing, what felt unfair, what broke their trust and what they wish existed. No recruitment, no incentive, no observer effect.

What most people see

Complaints. Noise. Players who don't understand the constraints the team was working under.

What I see

Behavioural evidence at a scale no research budget could buy — with the drop-off point, the trigger and the emotional reasoning all volunteered unprompted.

I'd like to end up in those rooms eventually — not because I'd know better than the people making those calls, but because I want to understand how they get made.

What evidence was actually on the table What the constraints really were Where player expectation and business reality had to be traded

STILL UNANSWERED

Questions I keep coming back to.

Not things I've solved. Things I notice repeatedly and haven't fully worked out.

01

Why do people keep using products they complain about constantly?

02

What turns a community transactional — and can it come back from it?

03

What actually makes someone feel like they belong somewhere?

04

What happens inside a product team in the weeks before a controversial decision ships?

TOOLS

ResearchGA4 · Hotjar · Dovetail · Maze · Moderated interviews
DesignFigma · Miro · Prototyping · Design systems
OptimisationA/B testing · Hypothesis design · Funnel analysis

EDUCATION

MSc Digital Design

Southampton Solent University

2019 – 2020

Postgraduate

BSc Business Information Technology

Southampton Solent University

2014 – 2019

Undergraduate

EXPERIENCE

2024–
2026

GoToHealthcare

CURRENT

UX Consultant · Social Care · Embedded

Embedded research on a platform where a drop-off means someone didn't get help. Mixed-method: heuristic audit, behavioural analytics, two rounds of moderated testing.

−11% drop-off 68s → 22s comprehension
Platform research programme
Conversion optimisation audit
AI-augmented synthesis rollout
Stakeholder research scoping

2024–
2025

WiseLaundry

UX Analyst · E-Commerce

Conversion research on a high-consideration retail purchase — where the barrier is confidence, not price.

Under NDA NDA

2022–
2024

Kinesso

UX Analyst · Media & Advertising

Research into how campaign and media teams interpret performance data, and where that interpretation goes wrong.

Client work · not published

2018–
2022

Contract & Agency

musicMagpie · Volvero · Fareham IC

E-commerce, mobility and innovation support. Where I learned to run research without a research budget.

Early career

3+

Years building & running creative communities

Multi

Discord · Twitch · streaming ecosystems

UX→

Applied beyond products — in culture & community

GMT

UK-based, available remotely

"The most interesting user behaviour doesn't happen in a usability lab. It happens when people think no one is watching — and I've spent years in those spaces."

— ON RESEARCH METHODOLOGY

HOW I THINK

Four things I keep relearning.

I've spent time on both sides of the screen — studying how people behave, and building the spaces they behave in. Those two things teach you different lessons, and the second one is harder to fake.

I don't just ask what the data shows. I ask what it feels like to be the person generating it — the empathy isn't performed, it's earned.

Behaviour over self-report

People are honest about what they want and unreliable about what they'll actually do. The interesting bit is always the gap between the two.

Context is everything

The same person is patient in one place and impatient in another. I've stopped believing in personas that follow you between contexts.

Trust is the metric that matters

Product page or Discord server, people stay where they feel safe and leave quietly where they don't. Nobody files a ticket about it.

Evidence, not instinct

My instincts are usually a decent first guess and a bad final answer. That's roughly what research is for.

SKILLS & METHODShover to pause ·

RESEARCH & STRATEGY End-to-End Research Mixed-Methods Research Research Roadmapping & Planning Voice of Customer Advocacy PRODUCT & CRO Conversion Rate Optimisation A/B Testing & Experimentation Behavioural Analytics Funnel & Journey Optimisation INSIGHT & DECISIONS Insight Synthesis & Storytelling Journey Mapping & Personas Behavioural Pattern Analysis Hypothesis Development Data → Design Translation AI & DATA Human-AI Interaction Explainability Awareness Data Interpretation Insight Translation System Thinking RESEARCH & STRATEGY End-to-End Research Mixed-Methods Research Research Roadmapping & Planning Voice of Customer Advocacy PRODUCT & CRO Conversion Rate Optimisation A/B Testing & Experimentation Behavioural Analytics Funnel & Journey Optimisation INSIGHT & DECISIONS Insight Synthesis & Storytelling Journey Mapping & Personas Behavioural Pattern Analysis Hypothesis Development Data → Design Translation AI & DATA Human-AI Interaction Explainability Awareness Data Interpretation Insight Translation System Thinking

EXPERIENCE

Background

4+

Years in UX research & CRO

Embedded

In-house product teams & agency delivery

MSc

Digital Design · Southampton Solent

UK / Remote

GMT · available now

Previously in research, analytics and CRO roles across social care, e-commerce, media and marketplace products. Full background →

WORKING TOGETHER

What it's like to have me on the team.

I work best embedded rather than at arm's length — close enough to the product decisions that research arrives while there's still time to act on it.

I'd rather be useful than right

A finding that arrives after the decision is made is just an opinion with a chart. I'd rather give a team something directional on Tuesday than something perfect a month late — and be clear about which one they're getting.

I'll disagree, then commit

At GoToHealthcare the client wanted more homepage content; the audit said less. I argued it with the data, they changed their mind, and drop-off fell. If the decision goes the other way, I'll help make it work and we'll measure it anyway.

I say what the evidence can't support

Eight participants can tell you a mechanism exists. They can't tell you what it's worth. Being explicit about that is what makes the confident claims believable.

Ambiguity is the normal state

Most briefs start as "something's wrong and we're not sure what." That's the interesting part. I'm comfortable spending the first week establishing what the actual question is before anyone commits to answering it.

OFF THE CLOCK

I can't stop asking why.

When I use a product, I've never been able to stop at "do I like this?" The question that actually interests me is what the team was trying to influence, what evidence they had, and what they had to give up to ship it.

A game changes its progression systemEveryone argues about whether it's fair. I want to know which retention number prompted it, and whether the fix addressed the cause or the symptom.
A streaming service reshuffles its homepageSomebody decided discovery mattered more than resuming what you were watching. That's a real trade-off, made by real people, and I'd like to know how they argued it.
An AI product launches with a confident toneInteresting choice — confidence drives adoption and destroys trust the first time it's wrong. I want to see where that lands six months later.
A creative community starts to go quietThat's rarely a content problem. It's usually structural, and it's the failure mode I find hardest to look away from.

WHAT IT COMES DOWN TO

People deserve to feel heard.
PatientsShoppersPlayersCreatorsCommunity members

Every project I've worked on started from the same observation — somebody's needs weren't being represented well enough, usually because nobody had asked properly.

Research is how you find that out before the assumption becomes a product.

FORMAL WORK

The corporate context I bring the same thinking to.

Across my case studies I've applied mixed-method UX research and CRO thinking to high-stakes product decisions at scale — from second-hand e-commerce trust to premium device conversion journeys. The briefs are different. The human behaviour isn't.

CeX

Trust · Second-hand retail

GoToHealthcare

Social care · Drop-off reduction

Samsung

PDP optimisation · A/B experimentation

WiseLaundry

AI companion · In progress

SEE IT IN PRACTICE

Curious how this thinking
shows up in the work?

Four studies — one commissioned, three run on my own initiative. Each one shows the reasoning, not just the outcome.