What happened
What this says
Current to 13 Oct 26Siân Pritchard asked how to structure a dashboard with multiple dimensions to avoid sparse matrices. Bethan Lister recommended region as the first dimension with resolution type as breakdown, which Siân confirmed worked perfectly.
Problem: Siân wanted to build a complaints dashboard with dimensions for region and resolution type over time, unsure how to structure it to avoid sparse matrices.
Handling: Bethan recommended setting region as the first dimension with resolution type as a breakdown within that, with time on the x-axis and filters available to drill into regions. This design pattern avoids the sparse matrix problem.
Watch: Siân confirmed the approach worked perfectly and found it clean and straightforward. This shows active dashboard building and product engagement.
Details
Updated 9 Oct 26Trail
- created · Siân Pritchard 11 Mar 2026 13:53Our collections team wants a dashboard showing complaints by region and resolution type over time. I've started using the dashboard builder but I'm unsure about the best way to dimension this to avoid a sparse matrix or too many edge cases. How have you seen other teams approach this?
- public_reply · Bethan Lister 11 Mar 2026 18:34I've picked this up. Just so I've got this right: you want to show complaints by region, then broken down by resolution type, trending over time? If you give me a couple of examples of the regions and resolution types you're working with, I can walk you through building it. Thanks, Bethan
- public_reply · Bethan Lister 13 Mar 2026 12:06Here's what I'd suggest: set region as your first dimension, then add resolution type as a breakdown within that, with time on the x-axis. That way you avoid a huge sparse matrix. If you add filters above the chart you can drill into specific regions. Can you try that and tell me what you see? Thanks, Bethan
- public_reply · Siân Pritchard 13 Mar 2026 12:07That's worked perfectly. Clean and straightforward. Thanks, Bethan.
- status_change · 27 Mar 2026 23:59