Measuring What Actually Matters

Leading and Lagging Indicators for Integrated PM+CM Projects 

One of the most consistent themes in conversations I have with project and change practitioners is this: we know how to measure delivery, but we don’t know what to measure for change. Go live date, budget variance, scope adherence; these are well-understood. But “did the change stick?” is a question most teams can’t answer with data because they never defined what the data should look like.

This is a solvable problem. And it starts with understanding the difference between two types of measures and then applying them across the right dimensions.

Leading vs. Lagging: The Distinction That Changes Everything

A lagging indicator tells you what happened. It’s a rearview mirror; accurate, but by definition, late. By the time a lagging indicator reveals a problem, you’re already past the moment when intervention was cheapest.

A leading indicator tells you what’s likely to happen. It’s a signal, not a result. It gives you the ability to course-correct before outcomes are locked in.

Most project measurement frameworks are heavily lagging. For delivery, that’s often sufficient. For change, it’s almost never enough. You need to know whether adoption is building while you still have time to do something about it.

The goal is a balanced measurement architecture, one that pairs leading and lagging indicators across every dimension of change impact: people, process, data, systems, and behavior.

People Dimension

This dimension is where most change practitioners instinctively focus and for good reason. People are where change either takes root or fails.

Leading indicators:

•         Attendance and active participation rates in training and readiness sessions

•         Pre-training assessment scores

•         Stakeholder sentiment scores gathered through pulse surveys during Execute

•         Volume and nature of questions coming in to change champions or help channels

•         Manager readiness ratings; are people leaders prepared to coach their teams through the transition?

Lagging indicators:

•         Post-go-live adoption rates at 30, 60, and 90 days

•         Help desk ticket volume and categorization (user error vs. system error) 

•         Turnover or role change in affected populations

•         Employee satisfaction and engagement scores measured post-implementation

•         Net Promoter Score trends for the change initiative itself

Process Dimension

If a new process was the point of the project, then process execution data is adoption data.

Leading indicators:

•         Process compliance rates during pilot or phased rollout

•         Workaround frequency; are people finding ways around the new process?

•         Process step completion rates in workflow tools during UAT and soft launch

•         Escalation rates to supervisors for process-related questions

•         Deviation tracking in controlled environments

Lagging indicators:

•         End-to-end process cycle time compared to baseline

•         Error rate and rework volume at 60–90 days post-implementation

•         Exception and override frequency in automated workflows

•         Process audit results; formal confirmation the new process is being followed as designed

•         Cost per transaction compared to pre-implementation baseline

Data Dimension

Data quality is one of the most revealing and most underused lenses for measuring change adoption. If people aren’t using the new system or process correctly, it shows up in the data before it shows up anywhere else.

Leading indicators:

•         Data entry completion rates in new systems during the first weeks of use

•         Duplicate record creation, a classic signal that users are creating workarounds

•         Field-level accuracy during early adoption

•         Volume of data corrections or manual overrides in the first 30 days

•         Migration data quality scores during cutover

Lagging indicators:

•         Reporting accuracy at 60 and 90 days

•         Audit finding rates related to data integrity

•         Frequency of manual data reconciliation between old and new systems

•         Data completeness scores in key reporting fields

•         Cost and time associated with data cleanup activities post-implementation

 

System Dimension

System metrics are the most technically accessible; most platforms produce them automatically but they’re often siloed in IT and never connected to the change story.

Leading indicators:

•         Login frequency and session duration in new systems during early adoption

•         Feature utilization rates; are users accessing the full scope of the system?

•         System error rates during the first weeks (often indicate training gaps, not technical failures)

•         Support ticket volume and resolution time

•         User-reported confidence scores

Lagging indicators:

•         Monthly active user rates at 30, 60, 90, and 180 days

•         Feature adoption rates across the full system scope

•         ROI realization tied to specific system capabilities

•         Deprecation of legacy system access; if people can still log in to the old system, some of them are using it

 

Behavior Dimension

Behavior measures are the hardest to collect and the most important to track. Everything else; system logins, process compliance, data quality is a proxy for the underlying question: have people actually changed how they work?

Leading indicators:

•         Observation data from structured leader walkthroughs or gemba walks during early adoption

•         Manager coaching frequency; are people leaders having conversations that reinforce the new behaviors?

•         Change Champion network activity

•         Self-reported confidence and competence ratings (distinguish between “I understand it” and “I feel confident using it in real work”)

•         Peer modeling; are informal leaders in the affected population visibly adopting the new behaviors?

Lagging indicators:

•         Behavioral observation results at 90 and 180 days; structured, not anecdotal

•         Performance review data reflecting new-state competencies

•         Culture survey results in relevant dimensions

•         Regression rates; what percentage of the population reverted to old behaviors after initial adoption?

•         Business outcome data that only moves if behavior changes: customer satisfaction, cycle time, error rates, cost savings

 

Building Your Measurement Architecture

Measurement planning cannot happen at go-live. It must happen at Plan, when adoption metrics are being defined alongside delivery metrics.

For each dimension, the questions to answer at the start of every project:

•         What does good look like at 30, 60, and 90 days?

•         What signal will tell us; while we still have time to act, that we’re off track?

•         Who owns this measurement, and where does it go when the project team disbands?

That last question is the most important one. Measurement without accountability is just data. The real work is designing a measurement system that outlasts the project; one that belongs to the business, not the project team. Because the change isn’t done when the project closes. It’s done when the new way of working is the only way anyone remembers.