What should a CEO dashboard include?
A CEO dashboard is not a smaller version of the finance pack. It is the shortest set of numbers that can change a decision this week. The four that recur in almost every small and mid-sized company are the same four CEODash is built around: will there be enough money, how much to sell, what price to hold, and how to tell the story.
Updated 2026-09-24
Start from the decision, not from the data you happen to have
The usual way an executive dashboard gets built is additive. Someone asks what we can measure, the answer is everything, and the screen fills up. Six months later nobody opens it, because reading it takes longer than asking the person who already knows. The test that keeps a dashboard short is uncomfortable but simple: name the decision this number changes, and name the week you would make it. A figure that fails that test is not reporting, it is decoration. Revenue year to date is decoration. The week your cash goes tight is not. Four questions cover most of what an owner or chief executive decides on a weekly cadence, whether the business is a consultancy, a wholesaler, a workshop or a clinic. Everything below is organised around them.
1. Cash: will there be enough money?
Cash is first because it is the only one that can end the company. The section has to answer a question with a date in it — not "how is cash", but "which week does it get tight, and what closes the gap". That means a forward view rather than a balance. CEODash builds a rolling thirteen-week forecast from the opening cash position, the plan (your budget, or a run rate taken from closed periods when there is no budget), and the real timing of what you are owed and what you owe. Thirteen weeks is a quarter: long enough for a collections problem to show up while you can still act on it, short enough that the numbers are still real rather than aspirational.
- Cash position today, and the week a gap opens if one does
- Collections you are relying on, and how late they usually arrive
- Whether the gap closes on money you are already owed, or only if pipeline lands
2. Volume: how much to sell?
The volume question is not "how much did we sell" — that is in the P&L. It is "how much has to close, by when, for the plan to hold", and then whether the pipeline is big enough to produce it. What makes this section useful is coverage against a target, not activity counts. Calls made and emails sent measure effort, and effort is not the constraint in most small companies. Deals at a stage, with an amount and a close date, are the thing that either clears the number or does not. Where a CRM is connected — HubSpot is live today, and there is a CSV path for everyone else — those deals feed the same cash view above, which is what turns "we are behind on pipeline" into "we are behind on pipeline and that is the week it bites".
3. Price and margin: what price to hold?
Price is the fastest lever a small business has and the one most often pulled blind. The number that belongs on the dashboard is not the average discount — it is whether you are losing the deals you should be winning, and at what price. CEODash reads that out of the deal history rather than from opinion: win rate by deal-size band, using bands drawn from your own resolved deals rather than fixed dollar buckets; whether lost deals cluster above a price point and stall longer before they die; and whether losses are early, on fit and qualification, or late, in negotiation. Late losses above a price band are a pricing problem. Early losses are a targeting problem, and discounting will not fix them. One deliberate behaviour is worth naming: below a minimum number of resolved deals the card hides itself rather than drawing a confident trend from eight data points. A dashboard that shows a number it cannot support is worse than one that admits it does not know yet.
4. The story: what do you tell the board?
The fourth job is reporting outward — to a board, a bank, an investor or a buyer. Most companies do this in a deck that is maintained separately from the numbers, which is why the deck and the accounts drift apart and why board prep eats a week. The fix is not a better deck. It is generating the outward view from the same source as the internal one: a board pack built from the figures already in the system, and a read-only share link the recipient opens without an account. Whoever reads it sees a frozen snapshot of what you saw, which is the point — you are not asking them to trust a retyped number.
What to leave off
Per-channel marketing analytics, ticket queues, per-rep activity leaderboards, site traffic. Not because they do not matter, but because they belong to a functional owner who looks at them daily. A chief executive reading them weekly is either duplicating that person or not trusting them, and neither is a reporting problem. The honest version of this advice includes when a different tool is the right answer. If what you actually want is one screen per team, refreshed on a wall, a TV dashboard product does that better. If you want to assemble your own metrics from arbitrary sources, a build-it-yourself tool will take you further. The comparison pages make both of those cases in more detail, including where they beat us.
Frequently asked
How many metrics should a CEO dashboard have?
Few enough that you read all of them. The useful frame is not a count but a structure: one section per decision you actually make — cash, volume, price, and the story you report outward — and within each, only the figures that would change what you do this week. In practice that lands well under twenty numbers for a company under a hundred people, and the ones that get cut are almost always the cumulative totals.
How often should a CEO review it?
Weekly, at a fixed time, is the rhythm that fits the decisions. Cash timing and pipeline coverage move inside a week and are worth that cadence. Margin by product or job and the board-facing story move monthly and do not reward daily attention. Nothing on a CEO dashboard needs to be watched hourly; if something does, it belongs to an operational owner with an alert, not to the chief executive with a screen.
Do I need a BI tool to build one?
No, and for many companies a BI tool is the wrong shape. BI is a toolkit: it will build whatever you specify, which means you first have to know what to specify and then keep it maintained. That is a genuine advantage when your questions are unusual and you have someone whose job is to answer them. It is a cost when your questions are the ordinary ones and nobody owns the build. Our comparison with Klipfolio sets out that trade-off properly, including the cases where the build-it-yourself answer is the right one.