Finance & pilotage
Your dashboard can be perfectly wrong.
A dashboard can look flawless and still lead to poor decisions. The real issue is not the screen. It is the reliability of what it represents.
The number is not reality.
A dashboard can display technically accurate data while still giving a misleading representation of the company.
By the end of this read, you will be able to test whether the figures used for decision-making truly represent the same reality for finance, operations, sales and management.
There is something reassuring about a dashboard. Neatly aligned figures. Curves. A few indicators in green. Revenue rising, margin stabilising, a properly filled sales pipeline.
The information appears to be under control.
Then someone asks a question.
“The €2.4 million in revenue — is that invoiced, delivered or ordered?”
Silence.
Sales looks at its CRM. Finance opens its ERP. The operations manager checks their own file. Three figures appear.
None of them is necessarily wrong. And that is precisely where the problem begins.
On several indicators, Belgium is above the European average in terms of digitalisation. The European Digital Decade report notably states that 74.5% of Belgian SMEs have at least a basic level of digital intensity, compared with 57.7% in the European Union. For data analytics, 44.5% of Belgian companies were using it, compared with 33.2% in the EU. [1]
of Belgian companies were using data analytics, compared with 33.2% in the EU. [1]
ERP in 2025: small businesses versus large enterprises. For BI: 11% versus 69%. [2]
The same company can speak several languages.
A figure can be correct and still be the wrong figure.
An exact figure can be wrong for the decision
Take revenue. It seems difficult to imagine a more objective indicator. Yet ask the sales director, operations manager and CFO for the month’s revenue at the same time.
The first may be talking about signed orders. The second about goods delivered or services performed. The third about recognised revenue.
Everyone is talking about performance. No one is measuring exactly the same thing.
A dashboard does not therefore need to contain an error to be wrong. It is enough for the displayed definition not to match the question the decision-maker thinks they are asking.
A figure is never just a value. It has a definition, a scope, a date, a source and a calculation method.
Research on data quality in management accounting shows that the quality of the information system and the quality of the data used for management are closely linked. A study of 143 medium-sized and large German companies also highlights that a variety of sources can add complexity when the information architecture is not under control. This result is not a benchmark for Belgian SMEs, but the mechanism deserves attention. [3]
First weakness: definitions
In many organisations, disagreements about figures do not come from a lack of competence. They come from the information architecture.
The CRM has one definition of an active customer. The ERP has another. The sales file retains certain historical categories. Financial reporting applies accounting rules. Operations work from orders that have actually been launched.
Then Power BI, Excel or another tool aggregates all of this. The result may look coherent on screen even though the objects that make it up are not.
More data does not automatically produce more truth.
The difficulty begins when several definitions coexist and no one knows which one drives the decision.
Ordered, delivered, invoiced or recognised?
Gross, contribution, after discounts, logistics or fully loaded?
Raw, qualified, weighted, proposals sent or probable commitments?
The average can be reassuring. Until you look underneath.
A dashboard is not a KPI museum. It should be a decision-making instrument.
Second weakness: time
An indicator can be accurate and already too old. Yesterday’s cash position is not necessarily Friday’s. Monday’s order book can lose a major deal on Wednesday. The margin shown on a sale may ignore a recent rise in costs.
The answer is not to put the entire company in real time. The real question is: how quickly does this information need to change for the decision it serves?
A strategic indicator can perfectly well be reviewed monthly. Cash pressure may sometimes require a daily frequency. A production problem may require an almost instantaneous signal.
Recent research among German management accountants clearly distinguishes information quality from decision quality. The study remains limited by a sample of 61 respondents and its German context, but it confirms the value of treating the dashboard as a decision-making tool, not simply a visualisation screen. [4]
Third weakness: aggregation
An average margin of 28% can look reassuring. Until you discover that three product families are highly profitable and two others are barely profitable at all.
A stable absenteeism rate can hide a concentration in a critical team. Rising revenue can mask the gradual disappearance of the most profitable customers. Correct overall customer satisfaction can coexist with an explosion of complaints in a strategic segment.
The higher a figure rises in the organisation, the more aggregated it becomes. And the more aggregated it becomes, the more likely it is to lose the details that explain what is really happening.
The CEO needs information that is sufficiently summarised to lead. But information that is too summarised becomes unable to explain.
A good dashboard should allow two movements: quickly see what deserves attention, then drill down far enough to understand why.
If the first exists without the second, the dashboard becomes a traffic light. Green. Amber. Red. But no one knows exactly what switched the light on.
When the KPI ends up replacing the objective.
Everyone can be right. The company can still be wrong.
Measurement changes behaviour. That is one of the reasons indicators are useful. It is also one of their limits.
If only revenue is rewarded, its quality can become secondary. If only the number of leads is measured, marketing can generate more contacts without generating more opportunities.
If production is assessed only on volume, inventory or quality may end up absorbing the cost of performance. If average lead time becomes the supreme indicator, some complex cases become mechanically undesirable.
An indicator can help translate a strategy into concrete behaviour. The risk appears when the measure stops being a way to read the objective and becomes the objective itself. Research on dashboards and strategy shows that focusing on a few measures can also be deliberately used to align the organisation. [5]
The problem is therefore not simplification itself, but losing the link with the purpose.
Taken separately, each indicator can be perfectly defensible. Together, they can push the company in contradictory directions.
A recent field study conducted in five large Italian companies, based on 28 semi-structured interviews, shows how business intelligence systems transform information sharing. The authors describe management accountants as potential “orchestrators” of the information flow: they help structure what is produced, transmitted and interpreted. This fieldwork cannot automatically be generalised to all SMEs, but the idea is useful. [6]
A single source of data is not yet a single source of truth.
To create a single management truth, the company must also share the definitions, rules and context that make the data interpretable. This is a governance issue. Not only an IT issue.
The role of finance or management control can become that of a guardian of consistency: definition, reference framework, reconciliation, timing and interpretation.
Optimises cost per lead.
Optimises conversion.
Optimises utilisation rate.
Protects margin.
In an SME, data governance should be simpler. Not weaker.
The LUCID Management Reliability Framework.
Before adding a KPI or rebuilding a dashboard, six questions can be used to test every truly important indicator.
A shared definition
What exactly does this figure measure? “Revenue” is not enough. Invoiced, recognised, delivered, ordered, excluding credit notes, including recurring revenue? If two managers answer differently, the indicator is not yet governed.
A reference source
Where does it come from? Which system is authoritative? Who owns the data? What transformations does it undergo before reaching the screen? A KPI without traceability is difficult to challenge and even harder to correct.
Appropriate freshness
How current is the information really? The update frequency should match the speed of the decision. Precise reporting published too late can be less useful than an imperfect signal at the right time.
A clear scope
What is included and what is not? Subsidiaries, products, salespeople, periods, inactive customers, discounts, returns, indirect costs. Differences in scope often produce differences in interpretation.
Possible reconciliation
Can we get back to the source figure? If no one can explain a variation without rebuilding three files for two days, the problem is no longer only analytical. It is organisational.
An associated decision
Which decision is this KPI supposed to improve? If no one knows the answer, it may have no reason to take up space on the dashboard. A dashboard selects the information needed to decide.
A dashboard is not an inventory of available data. It is a selection of the information needed to decide.
PRINCIPLE: the LUCID Management Reliability Framework is an operational synthesis of principles drawn from data governance, management control and performance management. It is not presented as an autonomous scientific standard.
The real cost is not the screen.
It is the decision made with excessive confidence in poorly defined information.
Cash tied up against overestimated demand.
Discount granted on a miscalculated margin.
Hiring or investment triggered too early.
Confusing revenue with cash.
Campaign stopped despite profitable customers.
Hours lost reconciling instead of deciding.
Imperfect data rarely has its own accounting line. Its cost appears elsewhere: in delayed decisions, poorly allocated resources, purchases made too early, hires made too late or misplaced commercial priorities.
The role of finance changes with this reality. It would be tempting to conclude that all figures should return to finance. That would be the wrong interpretation.
Operations must retain their operational information. Sales must be able to explore their pipeline. Marketing must understand its own metrics. Managers need autonomy.
Finance or management control instead becomes the guardian of consistency, not the owner of every data point. Guardian of certain rules: definition, reference framework, reconciliation, timing and interpretation.
In the five large Italian companies studied in 2026, greater autonomy for operational managers does not make management control disappear. Its role becomes more architectural: building the framework that enables other functions to use information without each creating its own reality. [6]
This function is particularly important in an SME, where there is not always a data department, a BI manager and a management control team to absorb the complexity.
The executive committee test.
Choose five KPIs that are genuinely used to make decisions. Then ask the following questions, without preparing the answers in advance.
Do the people around the table give the same answers?
Your dashboard is probably beginning to play its role. The discussion can return to the business rather than the definition of the figure.
If the figures or their definitions differ and no one knows exactly why, the problem is probably not graphical. Adding a new dashboard will not solve it.
Chosen from those that genuinely drive decisions.
Asked without preparing the answers in advance.
To identify what needs to be clarified before deciding.
A good management system reduces debates about the figures in order to increase debates about the business.
Is your management information really reliable?
Tick the statements that are true. The aim is not to produce a perfect score, but to identify the areas where an important decision still relies on fragile data.
The 7 boxes on this card are interactive in compatible PDF readers.
7 checks.
Definition, source, freshness, scope, reconciliation, decision, then collective consistency.
The foundations of decision reliability are solid.
The dashboard can be useful, but some decisions still rely on fragile areas.
Before adding charts or automation, first make the management system reliable.
Which indicator does your executive committee look at every month without having recently checked that everyone gives it exactly the same definition?
Data does not become true because it looks good.
It may be the most seductive illusion of modern business intelligence tools. They create an extraordinary impression of control: a clean screen, information available everywhere, filters, trends, alerts, automatically updated figures.
All of this is useful. But interface sophistication does not compensate for weaknesses in definitions, processes or governance.
Data is not trustworthy because it is automated. A KPI is not relevant because it appears in Power BI. A dashboard is not a management system simply because the entire leadership team looks at it.
The difference lies elsewhere: in the collective ability to know what we are looking at, understand why it moves, know when to intervene and, above all, connect the figure to a decision.
The best dashboard is therefore not the one containing the most information. It is the one that reduces uncertainty enough to enable the company to make a better decision.
Before adding a new indicator, perhaps choose just one. Put it in the middle of the table and ask: “Are we all certain that this figure means the same thing?”
Fewer dashboards. More shared clarity to decide.
[1] European Commission, Belgium 2024 Digital Decade Country Report. digital-strategy.ec.europa.eu
[4] Rieg (2026), Exploring the determinants and performance effects of digital dashboard use by management accountants. Journal of Management Control, 37, 75-110. Published online on 5 September 2025.
[2] Eurostat, Larger enterprises used more e-business apps in 2025, 20 May 2026. ec.europa.eu/eurostat
[5] Reinking, Arnold & Sutton (2020), Synthesizing enterprise data through digital dashboards to strategically align performance. International Journal of Accounting Information Systems, 37, 100452.
[3] Knauer, Nikiforow & Wagener (2020), Determinants of information system quality and data quality in management accounting. Journal of Management Control, 31, 97-121.
[6] Ascani, Sardini, Montemari & Chiucchi (2026), Orchestrating the flow of information for decision making. Journal of Management Control, 37, 175-207.
Are we all certain that this figure means the same thing?
lucid.dtsc.be
Written and created by Junior Cantos for LUCID by DTSC
LUCID · SME Management by DT Services & Consulting
10
The first 30-minute conversation is free. That's where it gets framed, with no commitment.
