Is a company classed as "data-driven" solely because it collects vast amounts of data? And are data-driven projects the sole responsibility of IT? No, says Eberhard Lösch. In his article, he clears up these and other misconceptions and reveals what is needed for successful data-driven business.
Every company has data that is either not used at all or that is used (and reused) in a manner that is sometimes sensible, sometimes not. And "sensible use" is the decisive aspect here as well as the focus of this article on data-driven business. After all, reusing data in a way that adds value as opposed to simply collecting data is what tips the scales and determines whether a business is "data-driven" – or not.
Company: "We've been a 'data-driven' company for a long time. After all, we've been using databases for years."
Merely being in possession of technical infrastructure and data is not enough to make a company a data-driven business. It's almost equivalent to having a football kit and a pair of boots in your wardrobe and thinking that makes you an amazing footballer. A data-driven company uses data consciously and with the explicit intention of generating added value for the company by making intelligent use of data. The technology is merely the tool for achieving this.
Company: "We don't really see the value or objective of data-driven business. Our business has worked well without it so far, so why should we make changes to the existing setup?"
This argument takes past success and projects it into the future. However, experience has shown that relevant changes can occur over time (even in highly regulated environments). The company then finds itself in a dash to catch up so that it can at least keep pace with the competitive edge gained by its more innovative competitors.
Company: "We [department] are not responsible for ‘data-driven' projects, but a large IT project is due to be launched in the company soon."
IT projects and data-driven projects are two separate things. In Data Driven Business (as the name suggests), the focus is on data (and not technology). Of course, these projects are supported by technology, but the main benefit is generated from the intelligent use of data, i.e. the data from the department in question.
If new insight is gained from the data during the course of the project, it is normally only the department who can properly assess whether the type of information obtained is entirely in line with the project's advancement. Consequently, the department plays the main role in a data-driven project.
Did any of those misconceptions sound familiar to you?
We now know that a data-driven organisation is not created through the collection of data, that being data-driven makes a company competitive and that, while this topic may be related to IT, it concerns every division – from marketing to financial controlling through to senior management. But what specifically does it take for data-driven business to succeed? Here are four of the most important factors for success:
Data is an important part of a corporate strategy: Being data-driven must be permanently embedded in the organisation's most senior management committee.
Ideally, the focus on consistent data use should be carried forward by the organisation's most senior body. The added value generated by intelligent data use should be clearly acknowledged and achieved intentionally. Typically, important aspects of intelligent data use are embedded in the entire organisation's objectives and strategy. This ensures that the organisation's management team is fully aware that most of its thoughts and actions are guided by the use of data.
Out of the four factors listed here, this is the one with the most central significance and thus the greatest impact.
Involve the organisation by setting targets: The individual divisions draw up specific suggestions on ways to improve based on the use of data.
For an entire organisation to be considered as data-driven, the issue must also be embedded within the entire organisation. Experience has shown that using targets (e.g. OKR) to embed the topic can be very effective. For example, the management team could task the individual departments with thinking about how a significant improvement in the department's contribution to the organisation's performance can be achieved through the intelligent use of data. In response, the departments present carefully formulated improvement objectives.
Suitable methodology and working culture: Suggestions/theories are implemented in a goal-oriented manner by the relevant participants.
Good ideas or promising theories are no guarantee that these will actually succeed and that the desired benefit will be achieved. From experience, the form of implementation plays an important role. For instance, when it comes to the make-up of the project team, the close involvement of all specialists from the relevant divisions has proven to be essential. By dividing the project into small, self-contained steps (as is the case with an agile approach, for example), the specialists can immediately verify whether the project is on the right track. From experience, data-driven projects are no longer IT projects in the traditional sense.
Methodology: Take an open and flexible approach to finding a solution and have the courage to come up with crazy ideas
Methods such as Design Thinking have proven effective in finding the right solution. Ideally, the teams are made up of members from a variety of areas and feed different perspectives into the solution-finding process. It is important that the affected/benefiting department is represented in the project team. What's more, the intention is to "Fail early and often" – to turn learning from "mistakes" into a methodology. It is part of the concept to draw experience from previous attempts and then take these findings into the next stage.
Often lots of ideas and solutions do not come to fruition because they are deemed "crazy" from the outset based on past experiences and therefore aren't pursued any further. Generally, you do not make a real breakthrough in the brainstorming process unless you leave existing patterns/thought processes behind – in other words, freeing yourself from reservations and considering a wide array of ideas.
Making the best use of state-of-the-art technology: The benefits of pioneering technology are fully exploited at an early stage.
And of course, last but not least, technology also comes into play as a factor for success.
For example, smart data automation tools like biGENIUS can automate essential tasks throughout the entire lifecycle of a data analytics platform. Modern analytics solutions can thus be set up quickly and cost-efficiently.
Using Cloud computing makes it possible, for example, to deliver a suitable development environment (for data science, e.g. Machine Learning Studio) for creating a prototype within a very short space of time. State-of-the-art infrastructure therefore no longer has to be maintained and operated on a permanent basis.
It is important to know that technology in itself is a vehicle and is just there to provide support. This means that even the best technology is of no use if, for example, the underlying data is not suitable.
Despite all the technology, the focus should still be on the data and making use of the right technology on a case-by-case basis.
"Having a football kit and a new pair of boots in your wardrobe is nowhere near enough to make you a pro footballer!" And as this example perfectly illustrates that the mere presence of data does not mean that an organisation is data-driven. Nor is it an issue that only affects IT. The principle of being data-driven must be embedded in the corporate strategy, targets must be aligned with this, the working culture and methodology have to match up and, of course, the right technology, tailored to the existing data, is essential, too.
Would you like to find out more about the topic on the basis of a specific customer story?
Then register for IT Days 2020 now and attend Eberhard Lösch's session.