Process Experiments
Part of Work process experiments
Measuring a process change before applying it to everyone
Define a trial’s result, implementation and safeguard measures before deciding whether a process change can be used more widely.
Before widening a process change, measure the result it should improve, confirm the new route was used and look for costs elsewhere. Set measures and decision conditions before the trial. A favourable number noticed afterwards warrants investigation, but is not enough on its own to justify rollout.
Define the rollout decision
State the change, the trial group and the wider group that may later use it. The trial must establish whether the route is workable, whether the intended result improved, and what another team needs to use it safely and effectively.
Choose an eligible work type and a completion point. For a hypothetical handover change, the intended result might be an item reaching its next owner with enough information to proceed. Record whether that happened, how long the recipient waited and whether missing details had to be chased. Keep definitions consistent before and during the trial.
Set conditions that rule out wider use even if the intended measure improves: a missed required check, uncovered live service or unreasonable burden moved to another role. Give workers a way to raise these problems during the trial.
Collect evidence for different questions
Question / Evidence to plan
- Was the change used as intended?
- Eligible cases, exceptions and the route each used
- Did the intended result change?
- The same defined outcome before and during the trial
- Was the result still usable?
- Required checks, corrections and recipient feedback
- Could others use the route?
- Training, cover, access, time and worker experience
Use existing records where they hold the needed information. Include open items at review time; completed work alone can hide a growing queue. Record case number and type so a shift towards easier work is visible.
If a comparable group keeps the old route during the same period, examine differences in work mix, staffing and starting performance. If the only evidence is before and after in one group, describe the observed change with its limits.
Demand, personnel or another policy may also have changed. A small trial can show feasibility and promising signals without establishing an effect for every team.
Before vs During Trial: Key Metrics for Process Evaluation
- Was the change used as intended?
- Track eligible cases, exceptions, and actual routes taken
- Did the intended result improve?
- Measure consistent outcome definition (e.g., handover completeness)
- Was the result still usable?
- Monitor required checks, corrections, and recipient feedback
- Could others use the route?
- Assess training, access, time, cover, and worker experience
Key Evidence Sources for Process Change Evaluation
- Existing records
- Use case logs, completion times, and error reports
- Open items at review
- Track pending work to avoid hidden backlogs
- Recipient feedback
- Assess usability and completeness of outputs
- Worker experience
- Evaluate practicality and workload impact
Make a proportionate decision
Inspect ordinary cases and exceptions. Ask workers whether the steps were practical, and recipients whether the result was usable. Distinguish a flaw in the route from a failure to explain or support it; each suggests a different remedy.
Decide whether to extend within the tested setting, revise and test again, try another setting with appropriate cover, or return to the earlier process. Record what was observed, what remains uncertain and what conditions must be met before wider use. If the wider group has different demand, authority or skills, check those differences before adopting the change there.
