What an executive KPI dashboard should show by stage, where each number lives, why dashboards go stale and how to keep one current without an analyst.
An executive KPI dashboard is one view of the four to eight numbers leadership uses to decide things: revenue, cash runway, acquisition cost, pipeline, churn. Picking the chart type is the easy part. The hard part is that each number lives in a different tool, somebody has to go and get it every week, and the dashboard is only as true as its last refresh.
What belongs on an executive KPI dashboard?
An executive dashboard answers one question: is the company on track, and if not, where is it off? It is not a reporting dump. Every ticket count, every campaign, every support queue has a home somewhere, and that home is not this page.
The useful way to choose the numbers is to work backwards from decisions. List the calls leadership makes in a quarter: raise or hold hiring, increase or cut ad spend, change pricing, double down on a segment. Then ask which number would change each decision. The numbers that survive that test belong on the dashboard. Anything else is context, and context goes in a drill-down or a different dashboard.
Executive vs operational and tactical dashboards
The three kinds of dashboard differ in who reads them, how often, and what they are for.
Dashboard | Who reads it | Question it answers | Cadence | Typical numbers |
|---|---|---|---|---|
Executive | CEO, CFO, board, department heads | Is the company on track? | Weekly, monthly, board cycle | ARR, cash runway, CAC, pipeline coverage, churn, margin |
Operational | Team leads, managers | Is this function running well today? | Daily | Tickets open, deploys, campaign spend by channel, invoices overdue |
Tactical | Individual contributors | What do I work on next? | Hourly or live | Leads assigned, tasks due, ad set performance, queue depth |
How many numbers belong on one?
Cap the executive view at 8 to 12 numbers and treat 4 to 8 as the core that gets read. Beyond twelve, the page reads as a list rather than a brief, and a list does not change decisions.
Within that cap, pair results with drivers. Lagging numbers tell you what happened: revenue, churn, EBITDA. Leading numbers tell you what is about to happen: pipeline created, trial starts, activation rate. A dashboard with only lagging numbers reports on a quarter that is already over. One or two drivers per result is enough.
The cap matters more, not less, when the person pulling the numbers is an AI employee rather than an analyst. Every extra number is another source system to keep connected and another query that can quietly break.
The numbers a founder reads on a Monday
Five numbers cover the decisions a founder or senior leader makes in a week. Each one is tied to a call that gets made because of it.
Revenue, usually as MRR or ARR. The decision: is growth on plan, and does the plan still hold? Show the current figure, the change against last month and the change against the plan. Net new MRR (new plus expansion minus contraction minus churn) shows whether growth comes from new customers or from keeping existing ones.
Cash runway. The decision: when does fundraising or a cost cut have to start? Runway is cash in the bank divided by net monthly burn. It is the one number that can end the company, so it sits near the top and shows the date it implies, not just the month count.
Customer acquisition cost. The decision: increase ad spend, hold it or move budget between channels? CAC is total sales and marketing spend over a period divided by new customers won in that period. By channel it says where the next dollar should go. Blended it says whether the whole engine is getting cheaper or more expensive.
Pipeline. The decision: is next quarter’s revenue already in motion, and does the sales team need help now? Pipeline coverage is open qualified pipeline divided by the revenue target for the period. The coverage ratio a company needs depends on its historical win rate, so show the ratio alongside the win rate it is based on.
Churn. The decision: fix retention before spending more on acquisition? Customer churn is customers lost in a period divided by customers at the start of the period. Revenue churn weights that by what those customers paid, which is the one that matters when a few large accounts carry the business.
Get started free and ask Viktor to pull your first Monday numbers →
Which numbers by company stage and role?
The set shifts as the company grows. Before product market fit, the dashboard is about survival and signal: cash runway, active usage, a retention cohort and whichever early revenue number exists. Acquisition cost barely matters when the marketing budget is near zero, and a churn percentage computed on twenty customers moves wildly every month.
Once there is repeatable revenue, the acquisition and retention economics take over: CAC, LTV, the ratio between them, net revenue retention and pipeline coverage. At scale, profitability joins the page: gross margin, operating margin and EBITDA, plus runway reframed as months to profitability rather than months to zero.
A KPI with no owner does not get investigated when it moves. The table gives one standard formula per number and the role that answers for it.
KPI | Formula | Owning role |
|---|---|---|
ARR | MRR at period end multiplied by 12 | CFO |
CAC | Sales and marketing spend in the period divided by new customers won in the period | CMO (blended), CRO (sales-led) |
LTV | Average revenue per account multiplied by gross margin, divided by customer churn rate | CFO with CRO |
Churn | Customers (or revenue) lost in the period divided by customers (or revenue) at the start of the period | CRO or head of customer success |
EBITDA | Net income plus interest, taxes, depreciation and amortization | CFO |
Operating margin | Operating income divided by revenue | CFO |
Cash runway | Cash balance divided by net monthly burn | CFO |
A COO owns the operational drivers that feed these, such as fulfilment cost or support load, which live on the operational dashboard and roll up to margin on the executive one.
Where each number actually lives
This is the part a dashboard tool does not solve, because it is not about the chart. Each number on the executive page is computed from records in a different system, and someone has to log in, pull the records, apply the formula and type the result into a cell.
KPI | Source system | What has to be pulled |
|---|---|---|
MRR, ARR, revenue churn | Billing system such as Stripe or Chargebee | Active subscriptions, upgrades, downgrades and cancellations in the period |
Customer churn | Billing system plus the CRM for the customer count | Accounts at period start, accounts cancelled |
Pipeline and coverage | CRM such as HubSpot, Salesforce or Pipedrive | Open deals by stage, amounts, close dates, win rate |
CAC | Ad platforms (Google Ads, Meta Ads, LinkedIn Ads) for spend, CRM for new customers, payroll for sales cost | Spend by channel, new customers by source, period sales and marketing headcount cost |
ROAS | Ad platforms plus the billing system | Spend by campaign, revenue attributed to that campaign |
Cash runway, EBITDA, operating margin | Accounting system such as QuickBooks or Xero, plus the bank | Cash balance, monthly P&L, burn |
Three things make this harder than it looks. The same customer can carry a different name in the CRM and the billing system, so churn counts drift. Ad spend is reported in the ad account’s time zone and currency, revenue in the billing system’s, so CAC for a given week rarely lines up exactly. And the accounting close lands after month end, so margin runs a cycle behind sales. The sourcing problem, not the chart, is the real work, and it is the work that gets dropped first when the person doing it is busy.
Viktor’s Integrations page lists Stripe, HubSpot, Salesforce, Pipedrive, Google Ads, Meta Ads, LinkedIn Ads, QuickBooks, Baremetrics, Google Analytics and PostHog among the tools he connects to, with 27 native integrations and 3,200+ tools through managed connectors, which covers every source in the table above. He can also be asked to build a custom integration for a tool that is not in the catalog.
See the 3,200+ tools Viktor connects to →
Why executive dashboards go stale
Bad design is not what kills a dashboard. A dashboard dies when nobody refreshes it. The first version is built with care, gets opened in a leadership meeting, and is trusted. Then the person who pulled the numbers is out for a week, the Stripe export changes format, or the refresh slips from Monday morning to Tuesday afternoon. Readers notice the date stamp, stop trusting the page, and go back to asking for the number in Slack.
Three other problems follow, usually as symptoms of the first.
Overload. When refreshing is manual, nobody wants to maintain two dashboards, so operational numbers migrate onto the executive page. It grows to thirty tiles and stops being read.
Data quality. A manual pipeline has no tests. A changed column name, a renamed CRM stage or a duplicated customer record changes a number silently, and the error is found weeks later in a board meeting.
Low adoption. If leaders cannot trust the date or the figure, they do not use the page to decide anything, and a dashboard nobody decides from has no reason to be maintained.
Measure usage so this is visible before it is fatal. Put the refresh timestamp on the page. Count how often it was opened in the week and by whom. In the Monday meeting, note which decisions referred to the dashboard and which were made from a screenshot someone pulled separately. If the second list is growing, the dashboard is already stale.
How to keep it current without an analyst
The fix is to take the refresh off a person’s to-do list. Four practices do most of it.
Match refresh frequency to the number. Pipeline and ad spend can refresh daily. Revenue and churn refresh when billing closes, usually daily or weekly. Cash and margin refresh when the books close, so monthly, with a weekly cash balance in between. A dashboard that promises live numbers for everything will show a stale margin next to a live pipeline and lose trust for both.
Alert on movement, not on schedule. The dashboard is for the Monday review. For the rest of the week, a message when runway, churn or coverage crosses a threshold is worth more than another chart.
Make it readable on a phone. Leaders check numbers between meetings. A page that needs a laptop gets a screenshot sent around instead, and screenshots are stale the moment they are taken.
Restrict the sensitive numbers. Cash, burn and margin are not for every workspace member. The page needs sign-in and access settings so finance numbers reach the people who should see them and nobody else.
This is where an AI employee replaces the analyst’s weekly chore. Viktor hosts the dashboard as a Space: a web app he builds and hosts for the workspace, with its own database, sign-in and hosting, and an Access tab that controls who can open the app. By default he deploys an internal app that needs sign-in straight to production, while a public app goes to preview first and he asks for approval before deploying it to production. How the building works is covered in how your AI employee builds dashboards and internal tools, so this page does not repeat it.
Keeping it current is a recurring Task. Anything Viktor can do once he can do on a schedule, and the Scheduled Tasks doc gives a metrics brief every weekday morning and a pipeline report every Monday as examples. The Task pulls the subscriptions from Stripe, the open deals from the CRM and the spend from the ad platforms, applies the formulas above, updates the Space and posts a short summary with the link in the leadership channel. Viktor usually test-runs a new Task once before it goes live, with chat messages captured instead of posted and permanent writes to integrations blocked. The same pattern replaces the weekly reporting deck, as described in replacing weekly reporting with AI, and the finance side of it is in AI for finance teams. Setting the schedule up is in how to set up a recurring Task.
One boundary before the first run. Refreshing an internal dashboard is a read and a redeploy, and by default Viktor deploys an internal app straight to production, so the Task can run without an approval card unless an admin gates deploys. Anything that leaves the workspace is different: Viktor’s security page states that money moves, code pushes and customer emails wait for explicit approval in Slack and lists deploying apps among the actions that need approval; the Spaces doc above describes the public app going to preview first. If the Monday Task is meant to email the numbers to investors, that send waits for a person to approve it.
See how approvals and access work →
Pricing follows usage, not seats: credits are shared across the workspace, with no per-seat licenses, and a new workspace starts with up to $100 in free credits and no credit card. Pricing checked October 2026.
Start with up to $100 in free credits. No card required. →
FAQ
What should a CEO dashboard include?
Revenue (MRR or ARR with net new), cash runway with the date it implies, blended and per-channel CAC, pipeline coverage with win rate, and churn. Add operating margin once the company is past the growth-at-any-cost stage. Each number should map to a decision the CEO makes in a normal quarter.
What is the difference between an executive dashboard and an operational dashboard?
An executive dashboard answers whether the company is on track and is read weekly or monthly by leadership and the board. An operational dashboard answers whether one function is running well today and is read daily by the people running it. The executive view should contain the result, and the operational view the reasons behind it.
How many KPIs should an executive dashboard have?
Cap it at 8 to 12 and expect 4 to 8 to be the ones that get read. Past twelve, the page reads as a list and stops influencing decisions. Pair each result with one or two drivers rather than adding more results.
How do leading and lagging indicators work together on a dashboard?
Lagging indicators such as revenue, churn and EBITDA confirm what happened. Leading indicators such as pipeline created, trial starts and activation rate show what is likely to happen next. Side by side, a leader sees both the quarter that closed and the one that is forming, and can check whether the drivers produced the result.
Which tools and templates work for an executive KPI dashboard?
The tool matters less than the sourcing. Any BI product, a spreadsheet or a hosted internal app can display eight numbers. The decision is who pulls them from the billing system, the CRM, the ad platforms and the accounting system each week. A template built from the formulas in the stage-and-role table above, with one owner per number, outlasts any charting tool.
Can a dashboard stay current without a dedicated analyst?
Yes, if the refresh is a scheduled job rather than a person’s chore. With Viktor, the dashboard is a hosted Space and a recurring Task pulls the numbers from the source systems, updates the Space and posts a summary in the leadership channel. Sending the numbers outside the workspace waits for approval.