Choosing use cases and redesigning processes
Step 1 / 3·Prioritising use cases: value × feasibility
0/3 steps done
Learning objectives — after this module you will:
- Prioritise use cases with a value × feasibility matrix
- Redesign an HR process end-to-end around AI capability
- Define what humans keep and what AI assists with
Prioritising use cases: value × feasibility
Pick the right starting point instead of spreading thin.
There are dozens of places AI 'could' help in HR. Resources are finite. A two-axis matrix decides quickly: VALUE (time saved, quality improved, business impact) and FEASIBILITY (data availability, complexity, risk level, user readiness).
Scoring questions for each use case
| Axis | Questions to answer |
|---|---|
| Value | How many hours a month does this take? Is current quality a problem? Who benefits? |
| Value | If it improved, which business metric would move? |
| Feasibility | Is the data available and clean enough? |
| Feasibility | What's the risk if AI gets it wrong here? Does it directly affect people? |
| Feasibility | Are users ready, or will they resist? |
Start from the visible 'do now' quadrant
Within high-value, easy-to-do, prioritise a use case where MANY PEOPLE SEE the result. An early tangible win builds the trust and budget for harder steps. Conversely, opening with a complex 18-month project is the fastest way to get the programme cut halfway.
Key takeaway: Score use cases on value × feasibility; start with high-value, easy work whose results are visible.