What goes in a CTO dashboard in Metabase?

A CTO dashboard rolls engineering up to one page: delivery throughput, reliability, where engineering time actually goes, headcount and attrition, and cloud spend against budget. It is the board-meeting view — every card here summarizes a deeper operational dashboard like DORA or software delivery, and links down to it.

What does a CTO dashboard look like?

Here’s the layout this guide builds. The executive summary and attention card sit at the top so the whole story fits in the first screen; delivery and reliability come next because they are the questions a board asks first; investment mix, people, and spend fill the bottom half, ending in a per-team scorecard that names where each number comes from.

Which cards belong on a CTO dashboard?

The eight below answer the four questions an exec review always asks: are we shipping, is it staying up, where is the time going, and what does it cost.

• Deploys per month, with month-over-month and year-over-year comparison (trend)
• Lead time for changes, p50 and p85 in days (line)
• Uptime by tier-1 service, weekly (line)
• Incidents by severity per week (stacked bar)
• Engineering investment mix — roadmap vs. KTLO vs. support share of time (stacked percent bar)
• Cloud spend vs. budget by month, with the budget as a goal line (line)
• Headcount and trailing-12-month attrition (combo)
• Team scorecard — engineers, deploys, change failure rate, open incidents, and spend per team (table)

What data does the dashboard need?

• A deployments table with deployed_at, team, environment, and a caused_incident flag, from your CI/CD or Git host.
• An incidents table with severity, opened_at, resolved_at, and service, plus per-service uptime rollups from your monitoring stack.
• Issue-tracker data with an investment-bucket tag (roadmap / KTLO / support) at the epic or project level.
• An HRIS export with monthly headcount and terminations per department, for the attrition line.
• Normalized billing rows — invoice_month, team tag, cost — from your cloud billing exports.

How do you build it?

1. Land the five sources in one warehouse — deploys and incidents from your delivery tooling, investment tags from the issue tracker, headcount from the HRIS, and billing exports from each cloud account.
2. Build a monthly rollup model per domain (delivery, reliability, investment, people, spend) so every card reads a pre-aggregated table instead of raw events.
3. Create the summary cards first — deploys trend, change failure rate, uptime, open Sev-1/2 count — and check them against the operational dashboards they summarize, so the exec page never disagrees with DORA.
4. Add the investment mix, headcount, and spend cards, and set the budget goal line on the spend chart from finance’s number, not an average.
5. Add filters for team, service tier, and date range, then wire each team scorecard row to the team-filtered operational dashboard so drill-down is one click.

Example card SQL

SELECT
date_trunc('month', d.deployed_at)                    AS month,
COUNT(*)                                              AS deploys,
ROUND(
  100.0 * COUNT(*) FILTER (WHERE d.caused_incident)
    / NULLIF(COUNT(*), 0), 1
)                                                     AS change_failure_pct,
PERCENTILE_CONT(0.5) WITHIN GROUP (
  ORDER BY EXTRACT(EPOCH FROM d.deployed_at - d.first_commit_at) / 86400
)                                                     AS lead_time_p50_days,
PERCENTILE_CONT(0.85) WITHIN GROUP (
  ORDER BY EXTRACT(EPOCH FROM d.deployed_at - d.first_commit_at) / 86400
)                                                     AS lead_time_p85_days
FROM deployments d
WHERE d.environment = 'production'
AND d.deployed_at >= date_trunc('month', now()) - interval '6 months'
GROUP BY 1
ORDER BY 1;

Related

Metrics

Integrations

Dashboards

FAQ