5. Micro: the touchpoint level

1. On a digital product:
The surface is every direct moment of interaction between a user and the system. The work is mostly perception and cognition, whether people see what is on screen, understand it, and can act on it right away. Examples: CTA button, iconography, color palettes, form, etc.

2. On a physical product sold and shipped online:
The listing and its photography, the packaging, the labels and care tags, the unboxing moment, and the returns flow when the fit is wrong, etc.

3. What micro-level work aims for:
Friction reduction on the critical paths users already take. Examples:

• Reducing search, so people find what they came for faster;
• Reducing mistakes, so the wrong size or variant does not slip through;
• Reducing annoyance on the way to what they came to do.

4. How micro contributes to the company outcome:
The gains here show up fast and compound quietly if the team stays disciplined. When I explain this to an IC, I tell them their work is the atom of the company’s outcome, meaning what the whole thing is built from, not how small their piece is.

5. Ownership:
ICs (each team member) maintain this level day to day. The manager’s role is to set up the measurement system so the signal is honest and continuous

• Explicit feedback: NPS surveys, star ratings and reviews, thumbs up/down on specific features, exit intercepts, etc.
• Implicit feedback: Heatmaps, session recordings, scroll depth, funnel drop-off, error rates, time on task, etc.

6. Methods: 
We lean on patterns and components already proven by larger studies and leading teams, keeping delivery fast, cost low, and user risk minimal. Primary research is rarely feasible at this scale, so judgment and borrowed evidence do the work. 

Examples: Heuristics, Laws of UX, retail design conventions, behavioral psychology.

I treat this level as “cheap by design”, so it needs shared rationale, not research budget. When the team disagrees, we settle it against a principle both sides accept, not with a new study.

7. AI at micro level:
Of course we all use it, and it makes micro even cheaper than it already was. The risk I watch is handing entire micro tasks to AI, because this is the ground where junior and mid-level ICs used to learn how to defend a decision. Without that practice, they will not have an answer when the question comes: Why should I pay you when I can get the same output from a 22-dollar subscription?