The Business Intelligence (BI) industry is already around for a while but has transformed rapidly in the past years caused by one factor: technology. Large organisations, mainly tech and e-commerce companies, have embedded data and its consequent way of working as backbone/fundament of their business model. Everything is ‘data’ and ‘data is the future’, but how to extract its proper value remains a big challenge for a lot of organisations. They are often slow to adopt BI as integral part of their operations due to a lack of knowledge of what it exactly involves, where to start, and how long it will take to see any benefit [1].

The interesting part here is that, although technological advancements speed up this transformation, the way how organisations are managed should not be very different than years ago. It all starts with your corporate strategy and related business objectives that determine your business activities. These activities are nowadays surrounded by possibilities to quantify every aspect of it. However, simply loading this data into a datamart or buying dashboarding tools to make great visualizations does not make your organisation ‘data-driven’. Therefore, this article elaborates on the value and focus areas of utilizing BI on the road to establish a ‘data-driven way of working’.

The possibilities of BI and its potential value
First of all, let’s have clear what BI constitutes. According to Gartner [2], “BI is an umbrella term that includes the applications, infrastructure and tools, and best practices that enable access to and analysis of information to improve and optimize decisions and performance”.

In today’s world, AI, big data and analytics are making headlines. Technology makes everything more accessible, with the result that BI has evolved ever since. For organisations, it is a challenge to fully utilise these technological opportunities to elicit the real value. Therefore, three takeaways [3]:

• The data is there: since the data is available, failing to use BI means potentially missed opportunities;
• The past and present are relevant: as more companies put emphasis on predictive technologies, some choose to shift away from BI. However, making this choice means you are not getting an accurate understanding of the implications of past events as well as a full assessment of your current state.
• It is part of a larger whole: BI provides a strong foundation upon which to begin making a new plan of action, as it lets you know where things were, the state of the company now, and how things shifted based on previous actions.

What constitutes a BI-organisation and data-driven way of working?
To make BI part of your organisation and establish a data-driven way of working, three fundaments are necessary: dashboarding, self-service analytics and data science [4]

Get behind the steering wheel

Dashboards are a means to steer on the performance of a business, department or specific process based on certain metrics/KPI’s that are visualized properly for the end-user to take action.

As defined above, there are three success factors that make dashboarding work [4]:

• KPI’s: the dashboard content is based upon the KPI’s and targets you have set. You can do this by setting up a KPI matrix that breaks down the strategic, tactical and operational goals to all levels in the organisation (department, team, employee).
• Data visualizations: tools such as Qlik Sense, Tableau and Power BI make it very easy to empower your KPI’s with great visualizations, but this is also the biggest pitfall. Data visualization is ‘a profession in itself’ despite the fact that it appears to be simple; poorly designed data visualizations / dashboards can leave useful information and insights underexposed and can make the data even less understandable than initially intended.
• End-user relevancy: dashboards must fit seamlessly with the specific needs of users in order for to make better decisions. In practice, the focus is often on available data and the functionalities of the tool used.

Every employee an analyst

Once the dashboards are up and running, the one answer leads to the other question. In this way, employees also want to work with data to get relevant insights, make drill-downs and eventually make smarter decisions.

In order to make self-service analytics work, things become a little bit more difficult. Again, three success factors [4]:

• A data desk: by setting up a governed data desk where users can ‘shop’ data to configure dashboards, analyses and reports you can ‘facilitate’ smarter decision making. Tools such as Alteryx and Collibra help to give shape to this ‘data desk’.
• Proper tooling: being able to ‘shop’ data also means you need to have the right tools available to do something with it. The larger the organisation, the more complex this becomes. Therefore, it should be a balance between central tools that are available to all employees and the flexibility to work with tools suitable for the type of project. For example, a data scientist needs the freedom to work beyond the standard reporting tool and use programs such as R or Python. Be aware, too much freedom could also lead to a widespread of definitions, KPI’s, interpretations and data where you should strive for a ‘single point of truth’.
• Data literacy: as Gartner perfectly states [5], “imagine an organisation where the marketing department speaks French, the product designers speak German, the analytics team speaks Spanish and no one speaks a second language.” That’s essentially how a data-driven business functions when there is no data literacy. If no one understands how to work with or understand data, you will probably leave your previous efforts/investments unused. Make sure to educate all your employees to unlock all the potential. AirBnB’s Data University is a great example of how this could work.

Looking back to the future
Where dashboards and self-service analytics are about yesterday, today and tomorrow, data science is about the future. Probably one of the most hyped business words nowadays, data science is in essence the application of statistics, machine learning and artificial intelligence to develop models based on historical data that can predict future events.

The data science industry is rapidly developing. In order to tap into its full potential, consider following three success factors [4]:

• Governed freedom: by giving data science a formal place in the organisation, for example in a ‘data lab’, data scientist can work together as a team with and for clients.
• Predictive power: data science revolves around the application of mathematical models and algorithms that are able to recognize patterns in data that cannot be seen by the naked eye. With technology for machine learning, a predictive model can be built and validated based on large amounts of historical data.
• From experimentation to realization: before you start building a data science model, it is essential that you know who will work with it and especially how. You want to know if you can implement the genius model or integrate it with certain applications or digital channels. To conclude, base data science work on specific use cases coming from the business.

Organisational factors to take into account
Organisational factors are of crucial importance to really become ‘data-driven’. Therefore, make sure to pay attention to the following aspects [4]:

• Vision & Strategy: a clear role of data in the company’s vision and strategy, including a roadmap to get there
• Organisation & Process: clear operational routes and processes to make data-initiatives succeed
• Culture & users: from CEO to the front-office, a ‘data-first’ culture embedded in all the layers of the organisation
• Tools & Architecture: future-proof tools and resources where employees can work with in a quick and easy way

Next steps to establish the new way of working

1. Determine the future perspective
   Determine the vision and ambition in the field of data-driven work. What does data add to the business model and performance of your organisation?

2. Analyse the current situation
   Where is the organisation now and what are points of improvement? Are these ambitions realistic and achievable? A maturity scan regarding your data an analytics can help in this to fit the gaps.

3. Develop a roadmap
   Based on the maturity scan and future perspective, there are concrete results to plan over time. With a roadmap, guidance is created for the projects to succeed, with possible support all accross the organisation.

4. Identify the quick wins
   Which cases and small projects can immediately deliver a lot of value? Consider, for example, redesigning important dashboards or drafting (better) KPIs to adjust the steering information accordingly

As you might have noticed, BI could be complex, but once you have started and see the first results on your dashboards it is not only useful, but also fun. Curious how this could work for your organisation in practice? Get in touch with us and we are more than willing to provide you with the right advise.

Sources:
1 Maximizer
2 Gartner
3 The Armada Group
4 Kadenza
5 Gartner