iThink Big
A live assembly program translated into a site that moves like the show and books like a business.
Capability · UX Research
Research that changesthe decision.
Every piece of bespoke software starts with how work actually happens. Research finds the workflow, the workaround and the decision the software has to support — and reduces the odds you build the wrong thing.
There's a version of UX research that produces beautiful reports nobody uses. The team runs a study because the process says to, presents a deck full of findings, everyone nods, and the product ships exactly as it would have anyway. That's not research. That's theater with a methodology.
Real research starts from a decision you're about to make and a risk you're trying to reduce. If a study can't change what you do next, we'd rather not run it. That single filter — will this change the decision? — cuts most research waste out of the room before it starts.
One conversation, done right, can be worth more than a survey of a thousand people. What matters isn't sample size for its own sake — it's whether you talked to the right person, watched them do the real work, and actually heard them.
The right method for the question
Research isn't one thing. Understanding whether a concept resonates is a different question from whether people can complete a task, which is different again from how a workflow really unfolds on the floor. Each calls for a different method — generative interviews, usability testing, contextual inquiry, diary studies, a movement study in physical space.
We pick the method that fits the question and your constraints, not the one that's fashionable. Sometimes that's a rigorous protocol. Sometimes it's watching five people try to do the thing and taking honest notes. Rigor is about matching the method to the stakes, not about maximizing ceremony.
Findings that travel
The hardest part of research isn't gathering insight — it's making it stick. Insight that lives in a researcher's head, or a PDF in a shared drive, doesn't change anything. So we design the output for how your team actually works: the two or three findings that matter, framed as decisions, with the evidence attached.
We're also careful to separate what we observed from what we're recommending. You deserve to see the raw signal and judge it yourself, not just inherit our conclusions. That's how research builds trust instead of just winning arguments.
Where it fits in the method
Every Artifact engagement runs on one method: Understand → Design → Build → Launch → Grow. These are the stages where UX Research does the most work. See how we work.
- 01Understand
Interviews and observation show how work actually happens — the workarounds, the handoffs, and the decisions the software will have to support.
- 02Design
We put rough, working software in front of the right people early, while their reactions can still change the direction cheaply.
- 04Launch
Usability testing and behavioral data explain what analytics can't: why people stall, where trust breaks, and what they expected instead.
What you get
- A sharp research plan tied to a specific decision, not a fishing expedition
- Well-recruited sessions with the people who actually matter
- Synthesis that names the few findings worth acting on
- Clear recommendations, kept separate from the raw observations
- Artifacts your team will actually reuse — journey maps, top-task lists, personas only when they earn their keep
Who it's for
- Teams about to invest in new software who want to de-risk the direction first
- Products with a nagging usability problem no one has pinned down
- Organizations rich in analytics but poor in the 'why' behind the numbers
- Leaders who want evidence to settle an internal debate honestly
Proof, not promises
Left Coast Printing Pros
A print shop site where customers place their own artwork on the garment — and the order arrives ready to produce.
IT Training & Certification School
A leading online IT training and certification platform helping people build technology careers through certifications and guided learning pathways.
Common questions
- How many people do we need to talk to?
- Fewer than you'd think, for most questions. Qualitative studies often reach the point of diminishing returns after five to eight well-chosen participants per audience. If you need statistical confidence — sizing a market, quantifying a preference — that's a different, larger study, and we'll say so.
- We already have analytics. Do we still need research?
- Analytics tell you what is happening; they rarely tell you why. The two are complementary. We often start from your analytics to find where behavior surprises you, then use qualitative research to explain it. Together they're far stronger than either alone.
- Can you research something that doesn't exist yet?
- Yes — that's generative and evaluative research with concepts and prototypes. We can test a direction with a rough working prototype long before it's built, which is exactly when the learning is cheapest to act on.
Get in touch
About to make a big bet? Let's reduce the risk first.
Tell us the problem. We'll tell you honestly whether software is the answer — and bring the strategy, design, engineering and AI to build it.