Wiki
Dashboard and Metric Layer Project Checklist
Archive-backed checklist for a dashboard or metric-layer portfolio project that proves stakeholder decisions, event definitions, metric ownership, tested models, BI consumption, and adoption.
Related Wiki Pages
Definition
A dashboard and metric-layer project proves that analytical data helps people make a decision. The project should start with one stakeholder decision. Then it should show source definitions, metric grain, and tested transformations. It should also show a BI surface and adoption evidence.
Use this checklist with Analytics Engineering Portfolio Projects when the target work is analytics engineering, product analytics, or data products.
Caitlin Moorman gives the adoption standard in Last-Mile Data Delivery. At 8:48, she defines the last mile. At 26:21 and 34:00, she works backward from user research and outcomes. At 38:15, she places metrics inside the meetings where decisions happen.
Common Definition
Guests treat a dashboard as the final surface of a data product. The useful proof is the path from events or source tables to tested models, metric definitions, dashboard consumption, and stakeholder behavior.
Arpit Choudhury gives the event path in Data-Led Growth Stack. He covers tracking plans at 13:34 and anomaly investigation at 18:27. He covers data flow at 22:50, SaaS event examples at 24:43, warehouse and BI work at 28:52, and activation at 30:03.
Adam Sroka gives the metric standard in ML Engineering KPIs and Metrics Strategy. At 22:41, he discusses KPI definition. At 28:04, he covers gaming risk. At 30:30, he covers derived KPIs. At 41:07, he discusses dashboards and visibility.
Guest Differences
For Caitlin, adoption is the starting point. A dashboard fails when people can’t find it or trust it. It also fails when the meeting workflow never uses it.
Arpit starts from product and growth data. He ties event definitions, warehouse transformations and BI into the same system. Reverse ETL belongs there too. That makes event ownership part of the project.
Victoria Perez Mola and Juan Manuel Perafan start from analytics engineering. Victoria connects modeling, data quality, and Looker. She also covers dbt docs, DAGs, and tests (Master Analytics Engineering, 4:05-38:53). Juan adds robustness, generic tests, and singular tests. He also covers CI, KPI tests, and semantic-layer thinking (Foundations of the Analytics Engineer Role, 38:41-1:14:40).
Stakeholder Decision
Start with one named decision. A growth manager or finance lead should be able to explain what changes after looking at the metric. A support team, product manager, or executive sponsor works too.
Caitlin’s outcome-first discussion at 34:00 supports this structure. Use Data Product Adoption and Data Product Management when the project needs adoption and stakeholder framing.
Metric Definitions
The project should define metric grain, units, dimensions, and filters. It should also define ownership, refresh cadence, and caveats. Separate primary decision metrics from diagnostics and guardrails.
Jakob Graff gives the experimentation guardrails in A/B Testing and Product Experimentation. He covers randomization at 8:13 and causality at 11:48. He covers A/A tests at 27:52, metric stability at 33:23, and power analysis at 37:44.
Use Metrics, Experimentation, and A/B Testing for the deeper measurement topics.
Tested Model Layer
A dashboard-only artifact is weak unless the metrics sit on reusable and tested models. Show staging models, intermediate logic, and marts. Also show metric definitions, tests, documentation, and lineage.
Nikola Maksimovic gives a practical version in Marketing to Analytics Engineering. He describes product support and A/B testing at 14:14. He covers data modeling at 18:34, Snowplow, dbt, and Looker at 20:34. He covers product analytics at 38:27.
Use dbt, Tracking Plans, Event Tracking, and Data Quality and Observability to connect the model layer to source quality.
Trust And Adoption
The project should show how errors are detected and how stakeholders learn to use the dashboard. Include a usage guide, definition page, warning note, or meeting workflow.
Tammy Liang gives the team version in Building and Scaling a Data Team. She discusses business health dashboards at 7:22 and reporting collaboration at 8:51. She covers a stack and Notion wiki at 22:32, dbt tests at 40:09, and workshops at 49:00.
Related Pages
Use these pages to follow the analytics and product context.