Key takeaways
- More dashboards are not the solution: according to research only 10 to 15% of organisations really know how to take data-driven decisions consistently.
- A “wrong” dashboard gives insight but no course of action; a “good” dashboard is linked to a daily or weekly decision and shows the possible action at once.
- The seven biggest problems are: accountability instead of steering, no genuine KPIs, no norms and targets, a business that does not understand BI, decisions out of sight, tools that allow everything, and neglected data analysis and AI.
- At organisation level there are only two or three genuine KPIs. Setting norms and targets and adjusting them regularly is the number one success factor.
- Ask not only “what do you want to see?” but also “what will you do with it and which decision improves?”. Then you automatically build fewer dashboards that add more value.
Everyone talks about data-driven working
In an ideal world we have all the information at hand to take good decisions every day. Practice is different: consistently taking data-driven decisions gets off the ground only with difficulty, however much we talk about data-driven working. Research shows that only 10 to 15% of organisations really know how it should be done; the rest muddle along. They build dashboard after dashboard and report after report to give decision-makers ever more detailed insight, hoping it will work out. It is an endless road that leads to a gigantic report factory and an enormous waste of money.
A wrong (dumb) and a good (smart) dashboard
A “wrong” dashboard gives insight but offers the user no course of action. It shows how things develop over time, but not whether that is good or bad. It takes the hierarchy (the silo or department) as its starting point and tries to account for things, but does not make the performance of the process visible. And it shows only aggregated information, which is far too abstract to steer on. What use is the insight into the number of parking permits issued over the past twelve months? At that level it is about strategy, budgets, opinions and looking back.
A “good” dashboard not only gives insight but also shows the possible actions straight away. It notes, for example, that the stock of an article has been shrinking fast in recent days and shows that you need to purchase ten extra today, on top of the regular order, to bring the stock back up. A good dashboard is linked to the daily or weekly decisions in a process. It looks ahead.
A wrong dashboard versus a good dashboard
The difference in one overview.
Wrong (dumb)
Good (smart)
What it gives
Insight only
Insight plus a possible action
Starting point
The department or silo
The process and its decisions
Level of detail
Aggregated
Fine-grained
Norms
No good or bad
Targets on every KPI
Direction of view
Backwards
Ahead
In the initial phase of business intelligence you have to start somewhere, often on the data side. That is not wasted effort: it shows users how powerful dashboards can be. But switch to smart dashboards as soon as possible. The seven problems below stand in the way.
Problem 1: a lot of accountability but little steering
Organisations drown in the detailed information they have to supply to account for themselves. In doing so they often forget to steer themselves, or the steering gets polluted with accountability information. They put their fate in the hands of auditors and inspectors instead of taking the helm. Think of banks that have to deliver thousands of data items to the ECB, with more added every year, or healthcare with its hundreds of quality indicators that are of little use internally and cost a fortune to supply. Control is good, but grip is better. Reporting for the sake of reporting is pointless.
Problem 2: no genuine KPIs
Effective dashboarding stands or falls with genuine KPIs. If you keep calling everything a KPI, you get bulging dashboards and soon make far more dashboards than necessary. At organisation level there are only two or three genuine KPIs.
Too many dashboards or reports actually reflect a problem in management.
If managers do not understand which buttons they can turn in which situation, the choice is often to put everything on the dashboard, which only makes the steering more complex. So first get the genuine KPIs clear together, and design your dashboards around them.
Problem 3: no norms and targets
Without norms and targets on your KPIs you have no goal and cannot steer. Setting norms and targets and regularly evaluating and adjusting them is the number one success factor in performance management and data-driven working. Without target values the conversation about performance is about nothing. So work on the target values before you make a dashboard; otherwise you get bogged down in requests for ever more insight, which are in fact diversions. Many managers respond to problems with new ideas instead of tackling the root cause, while that is exactly where the opportunities lie: continuous improvement is innovation too. Data analysis tells you the current performance and thus what a good target is.
Problem 4: the business does not understand BI and data-driven working
The business sees business intelligence as a tool and as something for IT. The only remedy is to build a solid bridge between business and IT, and that takes time. Both need insight into the potential of data-driven working, AI and BI and what it requires of everyone. IT also has to understand how processes work, how they are connected to the strategy through KPIs, what data-driven leadership means and which competences are needed. Data literacy has to go up, from top to bottom and across the organisation.
Problem 5: decisions remain out of sight
When developing a dashboard it is customary to ask the future user “what do you want to see?” or “what do you steer on?”. That assumes the user can say exactly what they want and what they will do with it, and it quickly leads to an arsenal of dashboards. Also ask: why do you need this information, what will you do with it and which decisions improve as a result? Then you build fewer dashboards, and the ones you deliver add far more value because they are linked to process improvement. The underlying problem is the composition of many data teams: mainly technicians. To link dashboards to better, more refined decisions you need senior business analysts and consultants with knowledge of performance management.
Problem 6: the tools allow everything
Whether you use Tableau, Qlik, Power BI or another BI tool, they all give the user far too much freedom to build dashboards.
Do not be blinded by everything the tools can do; focus on which information and insights can make the most impact where in the organisation.
Making information action-oriented can sometimes be done with a simple report with three columns. That can have much more impact than a flashy dashboard with bright colours, complicated charts and animations, which looks modern but often completely misses the point.
Problem 7: AI and data analysis are neglected
The core of success with BI, data-driven working and dashboarding lies in data analysis; see without data analysis BI is pointless. Dashboarding is still too often confused with data-driven working. How do you know which measures to take to improve the score on a KPI? Only by finding the root cause in the underlying data. So invest in the skill of data analysis. Because artificial intelligence automates the analysis function, think about applying AI as well. That also keeps the number of dashboards down, because one algorithm can replace 100 reports and dashboards.
How Diamond Decisions keeps the number of dashboards small
By starting with the decision. In Business Modeling the decisions, the genuine KPIs and their norms are established first. Data analytics then builds only the dashboards and analyses those decisions need, and Decision makers ensures that teams hold the right conversation about the figures every day, week and month.
Conclusion
To tackle these problems you need people who understand not only the technology but above all the business and human side of performance management and data-driven working. You need genuine KPIs on your dashboard, norms and targets on those KPIs, a well-organised process of data analysis and a good conversation about the figures: daily, weekly and monthly. Central to those conversations: what is going well, what can be better, what is really going wrong and how do we solve that for good? Only then do you link dashboarding to continuous improvement and get real value from your dashboards.
Talk to the author
Daan van BeekDiamond Decisions
Discuss your question with Daan
Tell us which decision needs to improve. In a no-obligation conversation we show you where your data delivers the most.
Schedule an introduction →Step 3 of our approach
Diamond Data Analytics
A data warehouse that starts with your most important decisions and is ready for conversational AI: a working prototype within 24 hours.
Explore Data Analytics →Frequently asked questions
01
How many dashboards does an organisation need?
As few as possible: one per group of related decisions. If a dashboard is not linked to a decision someone takes daily or weekly, it is probably superfluous.
02
What is the difference between a good and a wrong dashboard?
A wrong dashboard only shows what happened, at an aggregated level and without norms. A good dashboard shows whether the result is good or bad, which action is possible and belongs to a concrete decision in a process.
03
Why are norms and targets so important?
Without a target value a figure has no meaning: you cannot tell whether action is needed. Targets turn a KPI into a signal and make the conversation about performance concrete.
04
Where does a small or medium-sized business start?
With the two or three KPIs that sit at the heart of the primary process, a target for each and one simple overview that shows what to do when a target is missed. That often needs no more than one dashboard.
Further reading
About the author
About Diamond Decisions
Diamond Decisions helps organizations become better by enabling better decisions. We don’t start with data, dashboards or AI, but with the decision that needs to improve, and work back from there to KPIs, data and technology. We do that in four steps: inspiration and alignment, business modeling, analytics and data platform and data-driven decision makers. That way BI and AI don’t remain an IT project, but deliver value every day.


