nRF Cloud Express

Feedback
Feedback

If you are having issues with the exercises, please create a ticket on DevZone: devzone.nordicsemi.com
Drag & Drop Files, Choose Files to Upload You can upload up to 2 files.
Register

Building dashboards and comparing populations

In Lesson 3, you learned to inspect one device at a time. Managing a fleet means zooming out: instead of asking “what is this device doing?”, you ask “how is my whole population behaving, and is a new release making it better or worse?” Dashboards answer that question. This topic explains how to build them and, importantly, how to read them without being misled.

The Overview Dashboard and custom dashboards

Every project starts with a built-in Overview Dashboard that visualizes your fleet at a glance, including active devices, software version distribution, reboots, and top issues. If you edit or delete its charts, you can always restore them with the Restore Default Dashboard button on the All Dashboards page.

Beyond the built-in one, you can create custom dashboards to group data for a particular purpose, for example, a dashboard per team or one that tracks a specific cohort. You add charts from the All Dashboards page and rearrange them with Layout Mode.

Overview Dashboard

Chart types

Dashboards are built from charts, each suited to a different question:

  • Metric charts aggregate a custom metric across the fleet, with a choice of aggregation (count, sum, min, max, average) and the option to roll up by cohort or software version.
  • Issue and trace charts show how crashes and issues trend over time, including Top 5 Issues and new-issue counts.
  • Reboots and crashes charts break down reboots by reason.
  • Device Vitals cards surface the built-in vitals (stability, connectivity, battery life, flash wear) at the fleet level.
  • Device Set charts track the size of a saved search’s result population over time.

Filtering a dashboard

Every dashboard has a filter section at the top with two filters that behave differently, and the difference matters:

  • The Devices filter narrows to the devices that produced the data. Note that “Current Software Version” here selects devices by the version they are on now, so it can show all of a device’s historical data, not just data from that version.
  • The Software Version filter acts on the version that was running at the time the data was captured. This is the one to use when you want a per-version view.

Save your filters for the whole team with Save for Everyone, or revert with Reset.

Comparison Mode

To put two populations side by side, click Add comparison in the filter section. Each comparison gets a color and label (which you can customize) that appears in every chart’s legend. This is how you answer questions like “is the beta cohort crashing less than production?” or “is version 2.0 more stable than 1.9?”

Comparison of two versions in the Battery dashboard

Charts that show absolute values (counts and sums) can be misleading when the two populations are different sizes. Chart normalization fixes this by converting absolute values to a relative “per 1,000 devices” figure, shown with a 1k subscript (for example, 42₁ₖ means “42 per 1,000 devices”). Toggle it globally in the top-right of the page, or per chart. Normalization only applies to population-sensitive values; it has no effect on averages, minimums, or maximums. Always reach for it when comparing cohorts of different sizes or a population that grows over time (such as devices being activated continuously).

Comparison where normalization of charging time by population shows similar results

Drilldown

Charts with a Drilldown badge are clickable: selecting a section (or a label) takes you to the filtered list of devices behind that data point. This is the bridge from the fleet view back to individual devices, and it is the heart of the hands-on exercise below.

Drilldown from a dashboard to a device

Optional hands-on (nRF Cloud sandbox)

In the nRF Cloud sandbox, open the Device Vitals dashboard and use the Software Version filter to look at the vital scores across two versions. Spot where a regression was introduced (a version where reliability drops), then click into that section of the chart to drill down to the affected devices. Open one device to see the device-level view you learned about in Lesson 3.

This exercise demonstrates noticing a problem at the fleet level on a dashboard, then identifying the device behavior contributing to that fleet regression.

Make sure to Log in or Register to save your progress

Back
Next

Switch language?

Progress is tracked separately for each language. Switching will continue from your progress in that language or start fresh if you haven't begun.

Your current progress is saved, and you can switch back anytime.

Register an account
Already have an account? Log in
(All fields are required unless specified optional)

  • 8 or more characters
  • Upper and lower case letters
  • At least one number or special character

DevAcademy
Forgot your password?
Enter the email associated with your account, and we will send you a link to reset your password.