The SME CFO dashboard: built to steer, not just report

A financial dashboard for an SME CFO is a decision instrument that signals each week which actions are needed on cash, margin, receivables and obligations. A report is a different thing: it describes the past. The difference is not in the visualisation, it is in what you do with it. A good dashboard lets you scroll past the KPIs that are fine and pulls you toward the one that is off. Eight to twelve KPIs is enough; more than that and you are reading again instead of deciding. This pillar explains which KPIs actually steer per growth stage, where dashboards go wrong, and how to move from a monthly spreadsheet export to always-on insight without a six-month platform project.

Why monthly reporting falls short

The numbers in a dashboard are only as good as the data behind them. In practice most SME finance teams are at least a month behind on bookings: bank lines not yet imported, supplier invoices not yet processed, revenue from the CRM or billing tool not yet reconciled. That means a monthly report only flags the worst surprise after it has already happened.

Three patterns where monthly is structurally too late.

Cash surprise. A €3M revenue business with healthy margins finds at the start of the month that its bank balance is lower than expected. The reason: two large supplier invoices were not yet booked, and one customer paid ten days later than their baseline. Last month's report showed nothing unusual. Only on the tenth of the new month does the CFO see that working capital is two weeks too tight. A dashboard with daily cash on hand and a rolling forecast would have caught it on the second.

Margin erosion. At a €15M services business, gross margin drifts from 42 to 38 percent over three months. In the monthly numbers it looks like noise each time. Only at the quarterly review does the pattern become visible. By then the price list for the next quarter is already out, built on the old margin assumption. The CFO has lost three months because the dashboard layer aggregated only at quarter level, not on a rolling three-month average per service line.

Customer concentration. At a SaaS business with €5M ARR, the largest customer accounts for 22 percent of revenue. Their ordering pattern shifts: fewer seats, longer payment term, fewer support tickets (which looks good at first). Monthly revenue looks unchanged because another customer just expanded. The big customer's pattern is an early warning for churn that the finance layer cannot see because the aggregate number stays stable. A dashboard tracking customer concentration and per-customer payment behaviour would have flagged it three months earlier.

In all three cases the problem is not the data. The problem is the frequency and the granularity. A dashboard that is too coarse gives you the feeling of control without the signals.

The four layers of a good dashboard

A dashboard that shows only total revenue and total costs is a report. A dashboard that steers has four layers that explain each other. Per layer, two KPIs minimum, and those KPIs must act on each other so that one deviation leads you to the underlying question.

Cash and liquidity. The foundation. Two KPIs minimum: bank balance now, and a rolling 13-week cash forecast. A third often needed: burn rate in the current month. This layer answers "how much runway do I have?". If cash on hand deviates from the forecast, you know something is coming in or going out that you did not plan; the drill-down sits in procurement, receivables or working capital.

Margin and profitability per cost centre. Two KPIs: gross margin per business unit or service line (not total gross margin), and a 3-month rolling average of that same margin. Optionally a third: contribution margin per product or service line. This layer answers "where do we make money and is that shifting?". A stable total margin can hide a mix where one line goes up and one goes down.

Customer and receivables health. Two KPIs: DSO (days sales outstanding) and an aging breakdown of open invoices in buckets (0-30, 30-60, 60-90, 90-plus). Optionally a third: customer concentration as percentage of revenue from top 5 or top 10. This layer answers "who earns me money and who costs me working capital?". A high AR is not a problem if it is predictable; it becomes one as soon as the pattern deviates.

Obligations and forecasting. Two KPIs: open accounts payable commitments for the next 30 days, and variance between forecast and actuals for the past month. Optionally a third: forecast accuracy over the last three months. This layer answers "what do I owe, can I pay it, and how well do I know what is coming?". Low forecast accuracy is not a disaster if you know about it; it is a disaster if you base your reporting on it.

The four layers only work if they explain each other. A high AR is a signal for the cash layer; a margin erosion in one business unit is a signal for the forecast layer; a shift in customer concentration explains a change in DSO. Anyone reading the layers in isolation misses the real stories.

KPIs per growth stage

The KPIs that actually steer differ by company stage. A dashboard built for growth is not suitable for steady-state, and vice versa. The brief refers to three stages: growth, steady-state, turnaround. Below per stage a KPI set with formula, refresh cadence and owner, plus explicit markings for KPIs that are popular but largely vanity.

Growth stage (post-PMF, growing above 5 percent per month)

In growth the main question is cash: you spend faster than you collect, and the error margin on working capital is small. The dashboard should be built around that.

Vanity at this stage: total revenue year-to-date (lagging, no next action), website visits (not finance), team size as headline (only relevant in relation to capacity and burn).

Steady-state (steady, margin matters more than growth)

In steady-state the centre of gravity shifts to margin and operational efficiency. The question is no longer "will we make it" but "are we getting out of it what is in it". The dashboard should support that.

Vanity at this stage: trailing-twelve-month revenue as headline (lagging, no action), total customer count without mix (a major customer loss can stay invisible), industry benchmarks at organisation level (too coarse to steer on).

Turnaround (cash and costs under pressure)

In turnaround the question is "do we survive the next 13 weeks and can we structurally improve after". The refresh cadence goes up on everything related to cash and commitments.

Vanity in turnaround: anything that takes more than five minutes to update, quarterly aggregations as headline (too slow at this stage), industry benchmarks (irrelevant, you have your own problem to solve).

Common design mistakes

What makes a dashboard useless rarely sits in the tooling. It sits in the choices. The following mistakes show up at SME finance teams more often than they would like.

Vanity metrics. A KPI with no action attached. "Total revenue year-to-date" is vanity if there is no threshold that triggers a conversation. Ask of every KPI on your dashboard: "Which action follows when this deviates?". No answer? Drop it.

Non-actionable charts. A beautiful trend line that only says something changed, without a reference line (target, threshold, prior period at the comparable point), produces no decision. A chart should always show a comparison, not just a course.

Stale data. Reporting is only as good as the data behind it. Most SME finance teams are at least a month behind on bank bookings, invoice processing and CRM synchronisation. A dashboard built on stale data produces more misleading than no dashboard. Do not start with visualisation; start with the source cycle: make sure bank bookings run daily, invoice processing within 48 hours, CRM revenue weekly. Then the dashboard.

No drill-down to the books. A dashboard that shows a deviation but does not link to the underlying ledger lines forces you to open a second tool. Deviations get picked up late or not at all, because the threshold to investigate is too high. Every KPI should be one click away from the ledger lines or source lines that produced its value.

Mismatched refresh cadences. A dashboard with a daily cash stand next to a monthly margin trend, without clear distinction, leads to interpretation errors. State the freshness next to every KPI. A rule of thumb: a colour or a timestamp on each tile that says "last booking X hours ago" prevents anyone from comparing KPIs with unequal recency.

Missing exception layer. Most dashboards show only the current state. What is missing is what deviates from what is normal for that KPI at this point in the cycle. A good dashboard has a separate "what is different this week" section, populated by thresholds per KPI. That layer is what the CFO actually reads.

Built for the board, not for the operator. A dashboard that aggregates only at quarter or year level is suitable for supervisors, not for whoever takes operational decisions. Build two versions: an operational one for the finance team and the CFO, and a condensed one for the management team and the supervisory board. Do not try to combine both.

Too many KPIs. Above 12 to 15 KPIs per screen, the dashboard becomes a report. The threshold to look rises; the threshold to act drops. Always start with too few and only add something when you have wanted that specific KPI three times in the past month.

Build once, never iterate. Dashboards in Power BI or a similar BI tool are often built by someone with the technical skill but without the daily finance work. Changes require another ticket and days of turnaround. Result: the dashboard rots, irrelevant KPIs stay, new questions get answered in a spreadsheet again. Iteration should be within the CFO's reach.

Build or buy: three options and a decision tree

The choice between spreadsheet, BI tool or a platform approach is not ideological, it is practical. All three fit somewhere; the mistake is picking the wrong one for your situation.

Spreadsheet plus visualisation (Excel or Sheets with a light BI layer). Strong in the discovery phase: you do not yet know precisely which KPIs will steer, you want to try multiple cuts quickly, you have one main user who also knows the data. Spreadsheets beat platforms on iteration speed because you do not need to file a ticket to change something. Weak in maintenance: as soon as more than one person uses the file, parallel versions appear. Weak in live data: you work on a snapshot, not a continuous stream.

Power BI or a similar BI tool with connectors to your ERP and CRM. Strong when you know which KPIs are stable and you want a live connection to the source. Good for multi-source aggregation. Weak in iteration speed for the CFO: changing a KPI usually requires someone with BI knowledge plus finance context, and that person is scarce in an SME. Weak in decision-led design: BI tools are originally analytical, not exception-driven; you have to build the exception layer yourself.

Platform approach focused on finance (always-on dashboards plus drill-down to the ledger). Strong in maintenance across growth transitions: as you shift from growth to steady-state the KPI set changes, and a finance-native platform lets you do that without rebuild. Strong in multi-entity and in direct drill-down to the bookkeeping. Weak in own iteration if the platform is not itself flexible: you end up bound to the vendor's roadmap.

A decision tree in five questions.

1. How often do you change your KPI set? Monthly: stay in a spreadsheet. Quarterly: spreadsheet or BI tool. Yearly or rarely: BI tool or platform.
2. How many entities or divisions? One: spreadsheet or BI tool works. Two to five: BI tool or platform. More: platform.
3. Do you need drill-down to ledger lines? Never: spreadsheet. Sometimes: BI tool with good connectors. Always: platform.
4. Who maintains the dashboard? CFO or controller themselves: spreadsheet or platform with direct access. Data team: BI tool. Nobody dedicated: platform.
5. How critical is live data? Not really, monthly suffices: spreadsheet. Sometimes, signal within a day: BI tool with connectors. Always, signal within the hour: platform.

Honest note on spreadsheets. For an SME CFO with one entity, stable services, a finance team of two to four, and a month-end cycle that closes within five days, a good spreadsheet plus a weekly cash template remains a defensible choice. Not everything has to be platform-grade. The difference shows in growth transitions: as soon as you go multi-entity, or as soon as you add a new business unit, the spreadsheet approach breaks because it needs single ownership to stay maintainable. Plan that moment in advance.

From dashboard to action: weekly cadence and the five Monday-morning checks

A dashboard not embedded in a work rhythm becomes a trophy page sooner or later. The rhythm that actually works for an SME CFO has three weekly moments and one team moment.

Monday morning, 15 minutes. Five checks in fixed order.

1. Cash on hand versus forecast. Within 5 percent: fine, move on. Outside that: drill into incoming or outgoing bookings that deviate from plan.
2. AR aging buckets. Anything moved into over-30-days since last week? Which debtor, which invoice, what is the reason? Trigger an action (personal reminder, account manager conversation, or wait-on-context).
3. KPI exceptions. Which KPIs are red or amber this week? Write them down even if today is not the day to act. An exception without a response disappears.
4. AP commitments next 14 days. Do you have cover? If not, now is the time to discuss or reschedule, not in two weeks.
5. Operational signal (per stage). In growth: pipeline pacing or new-customer velocity. In steady-state: margin shift per business unit. In turnaround: cancellations or customer conversations that went badly.

Mid-week, 10 minutes. Exception review. The points from Monday: resolved, in motion, worsened? Three statuses suffice, no long notes.

Friday, 5 minutes. Close out this week's open exceptions. What stays open moves to next week. If something has been on your list for three weeks, lower your dashboard threshold or decide it is no longer an exception. Thresholds should follow your actual pain threshold.

CFO standup, 30 minutes per week. With the finance team and optionally an operational lead. The agenda is the exceptions from the past week, not the green tiles. A standup that only discusses what went well changes nothing. Per exception: what happened, what did we do, what should the threshold be?

Escalation paths are clear per layer. An anomaly on a customer payment goes to AR plus the responsible salesperson or account manager within 48 hours. A margin slip in a cost centre goes to the cost centre owner within a week. A cash shortfall goes immediately to CFO and CEO; no weekly cycle wait.

What makes this rhythm structurally different from a report-oriented process: on Monday you look only at what is different. A dashboard that forces you to look at every KPI trains you not to look at all. One exception tile that stands out trains you to move the whole finance function forward.

Frequently asked questions

What is a financial dashboard?

A financial dashboard for an SME CFO is a decision instrument that signals each week which actions are needed on cash, margin, receivables and obligations. The difference with a report is not in the visualisation but in what you do with it: a report describes the past state; a dashboard triggers a decision.

Which KPIs belong on an SME finance dashboard?

Eight to twelve KPIs per stage suffice. The core: cash on hand now and rolling 13-week cash forecast (cash layer), gross margin per business unit with 3-month rolling average (margin layer), DSO and AR aging buckets (receivables layer), AP commitments next 30 days and budget-versus-actual variance (obligations layer). The exact set differs by growth stage: cash dominates in growth, margin in steady-state, daily cash plus AP commitments in turnaround.

How often should a financial dashboard refresh?

Not one refresh cadence for the whole dashboard, but one per KPI. Cash on hand and AR aging buckets daily. Burn rate, runway, pipeline coverage and working capital weekly. Gross margin per business unit, EBITDA, customer concentration and budget-vs-actuals variance monthly. Forecast accuracy quarterly. A dashboard with one refresh button for everything forces a wrong choice (too slow for cash, too frequent for margin).

Spreadsheet or platform: when which choice?

A spreadsheet plus light visualisation wins as long as you are still discovering which KPIs steer, have one main user who knows the data, and one entity. A BI tool with connectors wins once your KPI set is stable and you want a live connection to the ERP. A finance-native platform wins as soon as you have multi-entity, multi-division or frequent growth transitions, or want direct drill-down to ledger lines. For many SME CFOs with one entity and stable services, a good spreadsheet plus a weekly cash template remains the right call; over-tooling is a real cost.

What is the difference between a report and a dashboard?

A report describes; a dashboard triggers. You read a report end to end; you scan a dashboard and stop at the deviation. A report is built for an external reader (bank, accountant, supervisory board); a dashboard is built for the operator who has to take decisions this week. In practice most SME finance reporting fails as a dashboard because it tries to be both: too detailed to scan in 15 minutes, too summarised to act on. Build two separate products, not one hybrid.

Further reading

Related reading and the product detail behind it:

• The Dutch SMB finance stack is about to break: why disconnected tools and manual exports stop scaling.
• Your finance team doesn't need more people, it needs more leverage: the operating-model side of the same problem.
• AI in Exact Online: what you can actually automate today: the integration that feeds an always-on dashboard.
• Reporting, continuous checks, and month-end close: the product pages behind a dashboard that steers instead of reports.

Directly applicable to the receivables layer in section 3 is how Nance keeps debtors in check without spamming customers. For those exploring the move to always-on insight, Nance's Exact Online integration lets a controller-grade dashboard run continuously against your ledger. For the strategic context of financial visibility in an AI-first finance function, see the CFO solutions page, or the pricing if you are building an implementation budget.

Anyone wanting to build, revisit, or sharpen a dashboard so it steers on actions instead of reports can book a conversation to see how Nance makes a controller-grade dashboard run always-on directly from Exact Online, Twinfield or Moneybird.