drafty

KPI dashboard template

Ask your agent for a KPI dashboard, get a clean chart-and-metrics view back — then share one link stakeholders can comment on inline. No login required to reply.

What it is
A KPI dashboard is a single view of the metrics that matter most for one goal — current value, target, and trend — so a stakeholder can read it in 30 seconds and know whether to act.

A KPI dashboard template answers one question: are we on track? The common failure mode is not picking the wrong chart type — it's having six versions of the dashboard living in six places, and nobody's sure which numbers are current. Generate it with your agent, share the Drafty link, and feedback lands on the exact metric instead of a forwarded PDF.

Generate it with your agent

Paste this into Claude, Cursor, or any agent:

claude
❯
Build a KPI dashboard for {team or goal}. Include: 5–7 metrics only (no vanity metrics — each one should prompt a clear action when it moves), the current value, the target, and a simple trend indicator (up/down/flat). Group them by category if there are more than four. Add a one-line owner and reporting cadence for each metric. Then publish it to Drafty so I can share a link and collect inline comments — no account needed to reply.▌

See it on a real one

Live canvas — comment on any elementOpen ↗

What belongs in a KPI dashboard

Five to seven metrics is the practical ceiling. More than that and the dashboard stops being a dashboard — it becomes a report nobody finishes reading.

Each metric needs four things to be useful:

  1. Current value — the actual number right now, not as of last week's export.
  2. Target — what "good" looks like. Without a target, a number is just a number.
  3. Trend — up, down, or flat since the last period. Context beats a snapshot.
  4. Owner — one name. A metric without an owner is nobody's problem until it becomes everyone's.

The most common mistake: tracking what's easy to measure instead of what matters. "Page views" is easy; "qualified trials started" is what tells you if growth is working. If moving a metric wouldn't change any decision, cut it.

The sharing problem most dashboards never solve

Most teams get the metrics right and then lose the feedback loop. The PM posts a screenshot in Slack. Someone replies with "the churn number looks off." Three days later there's a thread in a thread, a follow-up in email, and a note in a doc the original author doesn't know exists.

The dashboard that lives on a shared link means every reaction is anchored to the exact metric it's about. When your agent updates the numbers, the URL stays the same — no "which version are we on?" conversation at the start of every weekly.

When to use a KPI dashboard vs. a full report

Use a dashboard when you need a fast read — a leadership check-in, a weekly team standup, a board slide. Use a report when you need to explain the why behind the numbers.

The rule: if someone needs more than 30 seconds to get the point, it's not a dashboard — it's a report. A one-pager is the right format when you need to add context; a dashboard is for the numbers that speak for themselves.

FAQ

How many KPIs should be on a dashboard? Five to seven is the practical limit. Research consistently shows that dashboards with more than ten metrics produce worse decisions than those with five — the brain treats more numbers as noise and stops reading. If you have twelve metrics you care about, build two dashboards by audience, not one dashboard with twelve tiles.

What's the difference between a KPI dashboard and a KPI report? A dashboard is live and scannable — built for a recurring check-in. A report is a snapshot with narrative, built to explain. Dashboards answer "are we on track?" in under a minute. Reports answer "why" and take longer. Most teams need both; most teams only build the report.

What metrics should go on a KPI dashboard? Start from the decision you're trying to make, not from what's easy to pull. Ask: "If this number moved 20% this week, would I do something different?" If yes, it's a KPI. If you'd shrug, cut it. For product: activation rate, retention, and revenue-per-user will tell you more than sessions and page views.

How often should a KPI dashboard be updated? Match the cadence to the decisions it informs. A daily ops dashboard (support tickets, conversion, deployment health) should update daily. A strategic dashboard reviewed in a monthly business review can update weekly. Updating more often than decisions get made is wasted effort.

Can I have different dashboards for different audiences? Yes — and you should. An executive dashboard shows quarterly targets and trend lines. A team dashboard shows weekly actuals and the metrics each person owns. Showing an engineer the same view as the CFO doesn't serve either of them. The agent prompt above groups metrics by category; you can easily fork it by audience.

What's a leading vs. lagging KPI? A lagging KPI measures what already happened — revenue, churn, NPS. A leading KPI predicts what's coming — trial signups, activation completion rate, pipeline coverage. A good dashboard has both: lagging tells you the score; leading tells you whether the score is about to change. Most teams over-index on lagging indicators and wonder why the dashboard always feels like bad news after the fact.