Competitive advantage knowledge: How to gild your data stock

Companies today have a mass of data at their disposal that analysts could only dream of years ago. But this resource is rarely used consistently. Managing data as an asset is a paradigm shift for many companies. We show a pragmatic way to turn zeros and ones into gold.

In the past, valuable company documents were physically tangible: a construction plan on microfilm, an industrial patent on the wall, customer files and business books. Now companies generate and handle most of their knowledge digitally, and to an ever-increasing extent. New technologies, such as connected devices and intelligent software make it possible to transmit information in real time, analyze processes more precisely and enable further automatization. Purchasing, production, logistics, administration, customer contact – all these areas can be controlled more easily and effectively with the aid of precise data and suitable programs. In the best case scenario, data contributes to getting the most out of an organization. Or it can even build the foundation for new business areas.

Assessing and exploiting digital assets

But data in itself is neither the new oil nor is it pure gold. Companies first have to transform it into information that can be used for a defined purpose. This purpose determines the value of the data. For example, real-time motion data of vehicles is of great importance for a truck manufacturer that also offers a logistics management system. It is less valuable, however, for a competitor that is not aiming for the software market. Some companies already have their digital assets assessed by specialized rating agencies. Nevertheless, when it comes to an efficient management and profitable use of all data, most companies are still in the early stages. Valuable, but hidden information remains unused; less relevant data is maintained at great expense, and data that is no longer relevant is being kept and not deleted.

Companies that want to manage their digital assets more effectively need to take three steps:

  • The organization needs Data Governance-clear rules on how to generate, store, manage and process data. For each data category, for example, it must be determined who can access it, how it is secured, and how long it will be retained. Legal requirements define the minimum standards, while cost-benefit considerations determine further investments.
  • Knowledge silos must be opened. Employees should be able to easily view, retrieve and use data from other departments. To do this, the data must meet certain technical requirements.
  • – The company needs to strengthen its expertise: Not only data scientists are required, but an internal social network in which data processing experts work together with specialists from different business areas. Among this community, information on current business issues and the origin and significance of available data can be exchanged on a regular basis.

Many details have to be taken into account: What kind of information does the company collect all around the world, for example about suppliers? Which employees and external partners have access to this data? Which different regulations must be followed in the countries when information is shared, such as addresses? In order to keep the overview, an agile approach is recommended: A project team starts with basic Data Governance guidelines; together with the relevant business departments, they can work on individual use cases. Their initial findings will flow back into Data Governance and expand the company’s knowledge base.

Making machine data usable for employees

But there’s still more to it. Usually, companies first have to prepare and manage their data differently in order to fully exploit their value. For employees in different departments and locations, data content, formats and databases should be as easy to use as possible. For example, it helps if the company uses a common reference currency in sales reports. But there is another problem which is less trivial: unstructured information. On the servers of many companies, you will find masses of data generated by machines for machines. A different programming may be necessary here to add meaningful meta data, so that employees can get to grips with the figures. In the best case, the data meets the so-called FAIR criteria for international data science.

Sounds like a major project? That’s true, but companies do not have to prepare it in a big bang manner. They can start pragmatically with a work area that will benefit directly from improved data analysis, such as purchasing. Country by country, the corresponding data is classified, evaluated, made “fair” and shared among colleagues. Every buyer should be able to find any information he or she needs – the company, therefore, has to set up an internal data search portal (modelled on Google Dataset Search, for example).

Employees must internalize new routines

In such a project, normally the Business Intelligence unit within IT will be in charge. As technical experts, they have to coordinate with contacts in the business areas who know the context and origin of the data. It is very important that employees in the business units make themselves familiar with the new data guidelines. They have to understand the reason why they must handle their data differently. The more valuable the information, the more important it is that employees are motivated to maintain their data. Often it will not be enough to offer user-friendly software that guides employees through the new routine step by step. Supplementary workshops and training sessions, online help and special IT support may also be necessary.

Once the pilot project is completed, the company can build on these experiences. A good side effect of this project: It becomes more transparent for employees how their company handles data. Workers make a positive experience of digitization – as an opportunity for the company and as an improvement that makes their own everyday tasks much easier.

18.10.2018, Grosse-Hornke

More articles