A financial dashboard consolidates revenue, cash position, gross margin, operating expenses and a small set of leading operating metrics into one refreshed view. Its value depends on consistent source data and on a named owner who explains the movements, not on the number of charts it carries.
What belongs on a financial dashboard
A useful dashboard is short. Revenue against plan, cash on hand and runway, gross margin, operating expenses by function, and three or four operating metrics that actually drive those numbers in your business. Everything else is a report someone can open when they need it. A dashboard that shows forty tiles is a dashboard nobody reads, because it has made no decision about what matters.
It should also be obvious what is an actual and what is a forecast. Actuals come out of the ledger and are bound by the record-keeping rules the ATO sets for business records. Forecasts are a model. Mixing the two on one tile without labelling them is how a board ends up arguing about a number that was never claimed to be final.
The data underneath decides whether it works
Most dashboard projects that fail did not fail at the dashboard. They failed at the source. Non-standardised reporting styles, processes and data definitions are not really a data problem, they are a business problem,[1] and no visualisation layer fixes two teams counting revenue differently. It just renders the disagreement faster.
When the plumbing is done properly it is not exotic. What an autonomous finance function actually needs is a chat AI with projects, a data connector or middleware such as CData linking the source systems like Xero, the CRM and timesheets, IT support to keep it standing up, and validated review skills for anything legal.[2] That is a build, with an owner and a maintenance cost, not a subscription.
| What the dashboard shows | What a person still has to answer | |
|---|---|---|
| Revenue against plan | The gap, in dollars and per cent | Whether the gap is timing, pricing or demand, and which of those changes what the business does next month. |
| Cash and runway | The balance and the burn | Whether the runway assumption still holds given hiring commitments and collection performance that are not on the tile. |
| Gross margin | The percentage, by period | Whether a move is mix, cost of delivery or a one-off, and whether the cost lines behind it are being accrued consistently. |
| Operating expenses | Spend by function against budget | Whether an underspend is a saving or a delay, which is a conversation with the budget holder rather than a query on the ledger. |
Who owns the dashboard
Ownership splits in a way founders often miss. Someone has to keep the feeds and definitions correct, which is a control job. Someone else has to interpret the movement in front of the leadership team, which is a commercial job. In a lean function one person does both, and the interpretation is what gets dropped first, because the feeds break loudly and the missing commentary does not.
The direction of travel favours the interpreter. The technical skill set in finance is becoming less critical as systems and AI absorb more of it, and the weight is shifting towards business partnering and influence.[3] If you are hiring around a dashboard, hire the interpreter before the builder.
That interpreting half of the job has a name and a career track. Here is what a finance business partner actually does.
Why some businesses get less out of this than others
Be realistic about your own shape before you budget for a build. The businesses at the front of this are digital and tech heavy, software and similar, because they are simpler to automate and plug into. An inventory-heavy business selling physical goods is much harder, there are far more nuances, and with legacy data issues or imperfect infrastructure it gets difficult at scale, so most larger businesses with a stock component are not very advanced at all.[4]
None of that is a reason to skip the build. It is a reason to sequence it: fix the definitions and the source systems first, then build the view.
A dashboard nobody trusts is worse than none
The failure mode is quiet. The numbers stay technically correct, the refresh keeps running, and the leadership team goes back to asking finance directly, because they do not quite believe the screen. Numbers matter, but for finance leaders trust matters more, because without it even the best reporting will not get used.[5] A dashboard inherits the credibility of the person behind it. It does not create any of its own.
The commentary is where that trust is earned or lost, which is why variance analysis that actually gets read is the harder half of the work.
Common questions
What should a financial dashboard include?
Revenue against plan, cash on hand and runway, gross margin, operating expenses by function, and three or four operating metrics that genuinely drive those numbers in your business. Keep it short enough that someone reads all of it. It should also be obvious on the face of each tile which figures are actuals from the ledger and which are forecast, because mixing the two unlabelled is how a board ends up arguing about a number nobody claimed was final.
Why do financial dashboards fail?
Almost always at the source rather than the screen. Non-standardised reporting styles, processes and data definitions are a business problem rather than a data problem, and no visualisation layer resolves two teams counting revenue differently. The second failure mode is trust: the numbers stay correct, the refresh keeps running, and the leadership team quietly goes back to asking finance directly because they do not believe the screen.
Who should own the financial dashboard?
Two jobs sit inside it. Keeping the feeds and the metric definitions correct is a control job. Explaining to the leadership team what a movement means is commercial work, and it is a different skill. One person usually carries both roles in a lean finance function, so the explaining gets dropped first, because a broken feed is obvious and missing commentary is not.
Can AI build and maintain a finance dashboard?
It can do more of the assembly than it could two years ago. A practical setup is a chat AI with projects, a data connector or middleware linking the source systems such as the accounting ledger, the CRM and timesheets, and IT support to keep it running. Treat that as a build rather than a subscription, because it needs an owner and an ongoing maintenance cost. Digital, tech-heavy businesses get there far more easily than inventory-heavy ones, where legacy data and physical goods make automation much harder at scale.
References
- My view on this: the challenge of non-standardised data and processes is not merely a data issue, it is a fundamental business issue.
- What implementing an autonomous finance function actually requires: a chat AI with projects such as Claude, a data connector or middleware such as CData to link source systems like Xero, the CRM and timesheets with IT support, and validated legal-review skills for contract automation.
- What I am seeing in finance hiring: the technical skill set is becoming less critical as systems and AI handle more of it, shifting the focus towards business partnering and influence. Tom Hunter hosts The CFO Track, a weekly interview series with Australian CFOs and finance leaders.
- On who is really ahead with automation: the businesses at the front of AI adoption are very digital and tech heavy, like software, because they are much simpler businesses where it is easy to automate and plug AI in. When you are inventory heavy and selling physical goods that is very difficult, there are so many nuances, and with legacy data issues or imperfect infrastructure it is very difficult at scale, so most bigger businesses with an inventory component won't be very advanced at all.
- Something I say often: while numbers are important, trust matters more for finance leaders, because without it even the best reporting will not get used.