We’re having an internal debate about performance metrics for our workflow processes in AM 2023.1. The Analytics Dashboard provides a solid set of standard workflow KPIs - cycle time, throughput, exception rate, completion percentage. These work well for executive reporting and comparing performance across different workflows.
However, our continuous improvement team wants to create custom metrics that are more specific to our business context - things like ‘first-pass yield by operator skill level’ or ‘workflow rework cost by material category’. They argue that standard KPIs are too generic and miss the nuances that drive real improvement opportunities.
The counter-argument is that custom metrics create dashboard clutter, require ongoing maintenance, and make it harder to benchmark against industry standards. What’s been your experience? Do standard workflow KPIs provide enough insight for meaningful continuous improvement, or do you need custom metrics to really understand what’s happening in your processes?
We use both, but strategically. Standard KPIs go on executive dashboards for high-level monitoring and trend analysis. Custom metrics go on operational dashboards for the teams actually working on improvements. The key is not mixing them - each audience needs different information at different levels of detail.
From a pure analytics perspective, custom metrics are essential for root cause analysis. Standard KPIs tell you that something is wrong, but custom metrics tell you why. For example, overall workflow cycle time might look acceptable, but a custom metric showing ‘cycle time variance by shift’ might reveal that night shift has significantly longer processing times due to staffing or training issues. You can’t get that level of insight from generic KPIs alone.
Consider the lifecycle of your metrics. Standard KPIs are evergreen - they’re always relevant for monitoring ongoing operations. Custom metrics should be treated as temporary tools for specific investigations or improvement projects. Build them when you need them, use them intensively during the project, then archive them when the project closes. This prevents metric sprawl while still giving you the analytical depth you need.
The standard versus custom metrics debate is really about balancing consistency with specificity, and both are necessary for effective continuous improvement.
Standard vs Custom KPIs - Strategic Use: Standard workflow KPIs are essential for three purposes: establishing baselines, enabling benchmarking, and providing consistent executive visibility. Metrics like cycle time, throughput, and exception rate are universal enough that everyone understands them immediately and they facilitate meaningful comparisons across workflows, plants, or even companies. However, they’re deliberately generic and often miss the contextual factors that drive performance in your specific environment. Custom metrics fill this gap by capturing business-specific relationships - like how operator certification level impacts first-pass yield, or how material supplier variability affects workflow rework rates. The key is using standard KPIs as your foundation for monitoring and governance, while deploying custom metrics tactically for deep-dive analysis and improvement initiatives.
Dashboard Clarity Principles: The biggest mistake organizations make is trying to serve all audiences with one dashboard. This creates information overload and reduces the utility for everyone. Implement a three-tier dashboard strategy: Executive dashboards show 5-6 standard KPIs focused on business outcomes (OEE, on-time delivery, cost per unit). Operational dashboards show 8-12 metrics mixing standard workflow KPIs with custom operational metrics relevant to daily management. Analytical dashboards for improvement teams can show 15-20+ metrics including detailed custom calculations needed for root cause analysis. Use AM 2023.1’s role-based dashboard filtering to ensure each user sees only their relevant tier.
Continuous Improvement Integration: For continuous improvement to work effectively, you need both types of metrics working together. Standard KPIs identify that improvement is needed (cycle time is trending up, exception rate increased). Custom metrics then enable root cause analysis (cycle time increase correlates with new operator assignments, exceptions cluster on specific material types). Create custom metrics with explicit sunset dates tied to improvement projects. When you launch a Six Sigma project to reduce rework, create custom metrics to track the specific factors you’re investigating. When the project closes and improvements are standardized, archive those custom metrics and rely on standard KPIs to monitor sustained performance.
Practical implementation approach: Start with Analytics Dashboard’s standard workflow KPIs for all ongoing monitoring. These should never be removed or modified - they’re your stable foundation. Then create a ‘Custom Metrics Library’ with clear governance - each custom metric must have: defined business question it answers, data sources and calculation logic, owner responsible for maintenance, and review date (typically quarterly). Approve new custom metrics through a review process that evaluates whether existing metrics can answer the question or if new calculation is truly needed. This prevents metric proliferation while ensuring you get the specific insights needed for improvement. Finally, design your dashboard layouts to clearly separate standard and custom metrics - use different visual sections or tabs so users understand which metrics are permanent monitoring tools versus temporary analytical aids.