Quantitative
Counts, timings, rates, paths, errors, and changes over time.
- Analytics and conversion paths
- Performance and accessibility measurements
- Search behaviour and content use
- Retention, completion, and failure rates
DATA-DRIVEN interaction DESIGN / Summer Season ’26
What is data?
Data is evidence left behind by reality. It helps us see what is happening, understand why, and check whether a design actually made things better.
A product is always producing signals
- 4.2 sA page loads too slowly
- Where is it?A person cannot find the next step
- 404A journey ends in a broken link
- 72%People return after the first visit
- Tab stops hereA keyboard path is incomplete
- 3 ticketsThe same support question repeats
Not just numbers
Numbers show what happened. People help explain why.
Useful product evidence usually combines quantitative and qualitative data. One without the other can be precise, but still misleading.
Qualitative
Language, expectations, hesitation, context, and observed behaviour.
- Interviews and usability testing
- Support questions and complaints
- Observation of real tasks
- The gap between a promise and the experience
Evidence needs judgement
Data does not make decisions.
A metric can improve while the product becomes more confusing, manipulative, or inaccessible. Being data-driven does not mean obeying every number. It means making the evidence, interpretation, and intended outcome visible.
Observe
Interpret
Design
Measure
Then repeat. A useful design process is a learning loop, not a one-time reveal.
Measure less, learn more
The goal is not to collect everything.
Good measurement starts with a decision. What are we trying to understand? What is the minimum evidence needed? What should remain private? More tracking can create more noise, more risk, and less trust.
What HAAM looks at
Signals across the whole experience.
The strongest evidence often sits between design, technology, content, and operations. HAAM connects those signals instead of treating each one as a separate problem.
Behaviour
Where people go, stop, repeat, and return.
Performance
What is slow, unstable, heavy, or failing.
Accessibility
Who is blocked and which states are incomplete.
Language
What people ask, misunderstand, or describe differently.
Operations
What teams repeat manually and where work gets stuck.
Outcomes
Whether the product changed anything that matters.
From data to design
Evidence becomes useful when it changes what happens next.
- A form is repeatedly abandonedRemove unnecessary steps and test the new path
- People keep asking the same questionMake the answer part of the product
- A page is technically live but rarely foundChange its structure, language, and entry points
- A metric rises but trust fallsReconsider what is being optimised
This is what HAAM means by data-driven interaction design: observe reality, interpret it carefully, design a response, and measure what changed.
Have a product full of signals?
Let’s work out what they mean.
Bring the analytics, support messages, broken journeys, assumptions, and unanswered questions. We can turn them into a clearer next move.
Start a project ↗