Data is the new gold, for quite a while now. But why is data of gold value? And how on earth do I start my search for that gold mine? We find out from Robin Pauwels, expert in Data Analytics.

• First, we centralize all available data through data platforms, lakes and pipelines in the most efficient way possible. Most companies have many data sources, from a CRM system over the website to excel lists, ... All this data has to end up in one mountain and then be put into a certain format.
• Then you get to the second layer and start analyzing the collected data to extract valuable infomation for your business. This can be done via simple BI reports but it is of course more interesting to do more advanced analytics. Thanks to Artificial Intelligence (AI) or Machine Learning (ML) for example, you can extract insights from the data that make it possible to make predictions and recommendations. That's why they call data the new gold: if you have good data - collected in the right way - you can extract very valuable things for your business.

After the AI-hype

The fact is that companies working with data analytics have undergone an evolution. When the value of data became clear, a real hype arose: big companies spent a lot of money to get started with AI and ML - preferably before their competitors. The data analysts tried to do as many technical implementations as possible and answer the many customer questions. But by jumping right into the deep end without the companies really knowing what direction they wanted to swim in, not all projects achieved their potential.
Thus, the awareness arose that as a data analyst you have to work strategically and first listen very carefully to the ambitions of a company. What are the goals and how can data and possibly AI and ML be a solution? It is important that from data analytics we first make the link with the business and only then decide which tools and processes to install.

Eyes on the target

Once you get started, everything starts with data engineering. You create a good data flow on which you set up a future-proof data architecture that is scalable and on which you can build further and easily plug in new business cases. Only when this first step has been taken can you decide which more innovative cases you want to set up. For example, we can look for patterns within the data, anticipate customer behavior, forecast sales, predict market demands or target customers more specifically through marketing. We can also segment those customers, cluster them into certain groups with similar characteristics, etc. But on the other hand, we can also optimize the production environment internally, make the processes run more efficiently, predict breakdowns in advance or intelligently monitor when machines will break down.

Data analytics therefore offers a lot of interesting possibilities, but everything starts from the company's objectives.

Custom-made

What evolutions will we see in the coming years? The demand for data engineering is very high at the moment and it will evolve more towards data architecture because more and more companies want to bet on that. But also the platforms that offer out of the box solutions are definitely going to boom in the coming years.

But data analytic is and will continue to be customized. You always need a good data scientist who can communicate with the business and build the solutions to measure. After all, every company and every sector is different. However, we are evolving more towards so-called 'productized services', which is really the mix between a ready-made product and a customized service. To save time and money, the customer receives a fixed package that can be further customized, always based on the needs and objectives of the company.