Data Analytics offers companies a huge opportunity
Needless to say, the potential of Data Analytics is huge. Even if you don’t believe the promises made by consulting firms and take the quote “data is new oil” with salt, you can’t deny that information already has a huge impact on business models. Uber, Airbnb, and Netflix are obvious examples: They are not transportation, restaurant, or entertainment companies, but Data Analytics-driven companies.
However, the data is useless unless the organization is able to use it with (advanced) analytics to make better or faster decisions. Examples can be found in different industries and throughout the value chain.
Why is Data Analytics an Opportunity NOW? The answer is quite simple:
– Digitization of organizations: Making large amounts of data available that can be exploited through analytics. This applies to the classical digitization of processes, but also in new areas such as health information (health monitoring), the Internet of Things and open source services (such as PSD2).
– Easy-to-use analysis tools are available. A few are mentioned: DataRobot, H2O, Domino Datalab, Dataiku and Sagemaker; all enable automatic machine learning. Note that “the stupid tool is still silly,” but these solutions make ML more accessible to many analysts and organizations.
– The data transfer of organizations has increased. The potential of new knowledge-based business models is increasingly recognized. You do not need to study econometrics or computer science to adopt Data Analytics.
Lack of compliance with the company
Many research reports show that organizations are still unable to realize the potential of Data Analytics. The main reason is the lack of consistency between Data Analytics and the company. Reluctance does not cause this distortion.
On the other hand, a company is usually unable to articulate clearly what they expect from Data Analytics. Often, they expect a silver list, while a business challenge can be too complicated to solve completely using only Data Analytics. Or they distrust data and analytical insights in advance and rely on their professional experience and knowledge. On the other hand, many data scientists get their energy from building the most advanced analytical solutions without thinking about the consequences of using it in real life. The best solutions are created only through an open mindset and an iterative process between business and analytics.
Improving the efficiency of data analysis
While the above findings are fairly universal and permanent, there are plenty of opportunities to make analytics work in your organization.
First, focus on people over technology. The challenge of data analytics and business alignment is a matter for people. Make sure the technical people have sufficient soft skills to listen to the company, turn business opportunities and challenges into technical analytical requirements. In addition, schedule regular meetings on progress and findings so the company can take advantage of this. If your organization has a low level of analytical maturity, don’t hire a high-end data scientist who is only interested in the latest ML applications. Instead, hire analysts who can dispute the data, provide rudimentary insights, and work with the company to use those insights to make better decisions.
Second, focus on the effect instead of the most sophisticated designs. While this sounds trivial, it’s easier said than done. Most data scientists are more interested in building advanced models using the latest technology and ensuring that the model is integrated into the end-to-end process with an appropriate process to monitor performance and maintain the model. How to focus on impact?
– Regularly prioritize Data Analytics with your management team. Reviewing the impact of past and current initiatives and assessing the impact of possible new initiatives is a key part of this process.
– Be realistic about the contribution of Data Analytics to the problem at hand. In some cases, such as reallocating digital marketing, analytics can be completely descriptive. In other cases, e.g., rejection of a payment due to fraudulent claims, the person is required to review the case. The balance between analytical models and humans depends on the nature of the decision, the data available, the complexity of the decision, etc.
– Take an overview of the use of analytical insights. True capture of effects is as important as building analytical insights. For example, predicting customer switching can only increase restraint if the organization takes action on the forecast; a dynamic pricing model is very useful, but only if it can be applied in the market.
Third, embedding Analytics in an organization requires top management support. They need to lead by example, promote understanding and belief, build a re-use mechanism and develop skills:
– Combining analytical teams with management teams will help to create a mutual understanding of the barriers that will allow the analyzes to be fully exploited.
– Sharing success cases helps showcase the potential of Analytics
– Training at all levels of the organization increases awareness, conviction and willingness to experiment
Finally, Analytics is not a one-time operation but a journey of change. As with all changes, its success requires time and endurance. Or even better worded: “There are very complex issues to take full advantage of the analysis. These issues are not resolved through a conceptual approach, but through an experimental approach. Those companies that take the plunge, test, replicate their success on a larger scale, and learn from their failures could well create an unbeatable edge over their more hesitant competitors. “(McKinsey, Carpe Data, 2017)