Dashboards

Published: 2026-09-12 · Last reviewed: 2026-09-12 · Reading edit: 2026-09-12

The weekly dashboard has twenty charts, but the meeting ends without a decision. A more useful view starts with one question someone can act on. Choose that question first, then include the measures and underlying records needed to answer it.

Choose the operating question

Examples include which requests need ownership today, which group of inquiries from the same campaign period needs investigation this week, or whether next quarter’s pipeline assumptions remain plausible. A single view rarely serves all three audiences well.

Build the decision view

1. Choose who uses it and how often. State who uses the dashboard, when, and what they can change. If no action follows, the information may belong in a reference report instead.
2. Define each metric. Record numerator, denominator, unit, population, time window, source, refresh schedule, and exclusions. Distinguish zero from missing or delayed data.
3. Separate outcomes and diagnostics. Show the business result with a few measures that help explain it. Avoid treating every activity counter as an equal success metric.
4. Provide context. Include comparison periods, cohort maturity, targets or ranges with their basis, and annotations for material changes. A red indicator without a defensible threshold creates noise.
5. Let the reader inspect the underlying records. Allow an authorized user to reach the underlying cases while respecting access rules. Aggregates should reconcile to the source and use consistent definitions across teams.
6. Record the decision and check what happened. Record the decision, owner, deadline, and follow-up outcome in the operating meeting. Retire measures that repeatedly fail to inform decisions.

Worked example

A fictional daily routing dashboard shows eligible requests awaiting acceptance, their age under the agreed clock, and the responsible queue. It distinguishes five overdue requests from three requests whose region is still unknown.

The operator fixes ownership and escalates the missing-data pattern. The weekly campaign dashboard uses mature inquiry cohorts and qualification outcomes instead. Combining both into a single “marketing score” would hide the different actions each requires.

Metric contract

For the fictional routing example, this is the definition that sits behind the number:

Field	Definition
Decision and user	The operator decides which unanswered demo requests need attention today
What is counted	Unique demo requests with no meaningful response after the agreed response window
Included records	Open demo requests; exclude test records, duplicates, and support requests
Source and freshness	Request receipt and response times from the CRM; show the last successful refresh
Time rule	Apply the team’s documented business-hours policy; show missing timestamps separately
Next action	Open the overdue request, confirm its owner, and arrange the response

Check trust before design polish

Manually reconcile a sample period, test filters, and inspect timezone boundaries. Confirm that exports preserve definitions and that stale data is visible. Attractive charts cannot compensate for inconsistent stages or double-counted pipeline. When a metric changes definition, annotate the break rather than displaying an uninterrupted trend.

Try it with your own work

Pick one dashboard and ask its owner what they changed after the last review. Keep the measures that helped, and identify one missing detail that would make the next decision easier.

Sources and scope

• HubSpot: manage dashboards provides one platform implementation reference. The decision and metric contracts here are original guidance.

The example is fictional; any numbers illustrate the method rather than a benchmark. Adapt the worksheet to your own situation.

What to read next

Routing and SLA · Funnel model · Sales operating cadence

Chapter guide · All playbooks

---

Copyright © 2026 Ivan Xu. All rights reserved. See the copyright and reuse terms.

Canonical source: github.com/weilun88313/B2B-Playbook