| name | design-research-integration |
| description | Use when integrating user research into the design workflow. Covers when to do research (always vs sometimes), the methods (interviews, surveys, usability tests, analytics), and how to actually act on research instead of filing it.
|
Design Research Integration
Research without action is decoration. The goal is decisions, not deliverables.
When research is critical
| Always research before designing |
|---|
| New product / new audience |
| Major redesign |
| Building for a vertical you don't know |
| Solving a problem that isn't validated |
| Skip research if |
|---|
| You're an active user of your product |
| The change is small / reversible |
| Speed matters more than precision |
Research methods quick reference
| Method | Best for | Time |
|---|
| 1:1 interviews | Deep understanding of motivations | 1 hr per session |
| Surveys | Validate at scale | Day |
| Usability tests | Watching how users use the product | 30 min |
| Card sorting / tree testing | Information architecture | Async |
| Analytics review | What users ACTUALLY do | Ongoing |
| Diary studies | Long-term behavior patterns | 1-4 weeks |
| Customer support review | What's hurting them today | Ongoing |
Continuous discovery
Don't do "a research project." Do continuous discovery:
- Weekly user interview (1 hour, recorded)
- Monthly analytics review (which features used, drop-offs)
- Quarterly customer support theme analysis (top 5 issues)
This builds intuition over time.
Synthesizing research
Common methods:
- Affinity mapping (cluster sticky notes by theme)
- Journey mapping (visualize the user's path)
- Personas (cautiously — easy to oversimplify)
- Jobs-to-be-done (focus on what user is hiring product to do)
From research to design
For each research finding:
- What did we learn?
- What does it mean for the design?
- What specifically will we change?
- How will we measure if the change worked?
If you skip the last two, the research is decoration.
Common mistakes
- Doing "research" by surveying employees instead of customers
- Talking to 50 users but synthesizing nothing
- Confirmation bias — research that validates pre-existing decisions
- Personas that nobody uses
- Beautiful research deliverables that don't change the product
Where this fits in X3 Compass
X3 Compass currently: founder-led research (Joshua knows the trucking industry). Future: monthly customer interview, quarterly support review, continuous analytics.
