How do we measure whether a change initiative is actually working?

Most organizations measure change with surveys (awareness, sentiment, satisfaction) and outcomes (revenue, efficiency, adoption percentages). Surveys measure what people think. Outcomes measure the end result. Neither tells you whether the behavioral change is happening or why.

Behavioral measurement sits in the middle.

It tracks whether the specified target behaviors are occurring, with what frequency, and under what conditions. This is where the SHIFT framework's "T" (Test and Iterate) becomes operational.

Start with the target behaviors you specified during diagnosis. If you specified "managers conduct weekly one-on-one feedback sessions using the new framework," then measure that behavior directly: how many managers held the session this week? How many used the framework versus defaulting to unstructured conversation? What percentage completed the follow-up documentation?

Next, measure the barriers. If your diagnosis identified that the primary barrier was capability (managers did not feel confident using the framework), track proxy indicators: how many accessed the reference guide, how many attended the practice sessions, how many reported feeling confident in post-session check-ins? If the barrier was opportunity (no time in the calendar), track whether protected time was actually created and maintained.

The SHIFT framework's Test and Iterate stage treats measurement as a feedback loop, not a report card. Behavioral data tells you whether your intervention is hitting the right barrier. If managers are attending practice sessions (capability intervention is being delivered) but still not conducting feedback conversations, the barrier was probably not capability. That data triggers re-diagnosis and strategy adjustment. Measurement without diagnostic context is just surveillance. Measurement linked to COM-B diagnosis is a learning system.