Dernière mise à jour: 20 mai 2023
Le Data Analytics is a set of statistical methods aiming to draw conclusions from masses of information. This science is based on the innumerable data collected by an entity to understand certain phenomena and thus better anticipate them.
In the business world, this allows for example to make the most appropriate decisions as quickly as possible. Thanks to the development of science, analysts are equipped with descriptive, predictive or prescriptive tools and can thus make the data talk.
What is the difference between Data Mining and Data Analytics?
Unlike the Data Mining who searches the data, the Data Analytics focuses on raw data in order to draw more summary conclusions, without looking for hidden models.
The goal is to bring out understandable information and accessible to all, to make data talk. Data which is not observable at first glance and needs to be processed.
They are often translated by graphic representations to be even more easily understood.
The different types of Data Analytics
We distinguish 3 kinds of analytical domain. Each of them has a specific goal and participates in its own way in making the most appropriate decision.
- THEanalytical for descriptive purposes allows as its name suggests to describe a phenomenon. This is certainly the best known method, it consists of transforming data into knowledge.
By having 100 visitors to its website for 1 buyer, we deduce that the conversion rate is 1%. This data makes it possible to realize the weak transformation and therefore to set up a strategy to counteract this bad point. - THEpredictive analytics has it for objective to predict. The goal in this case is toanticipate potential events. This analysis is similar to the work of the Data Miner who provides models.
- Finally theprescriptive analytics allows him to choose between several proposed actions in order to act the final result.
Concrete examples of use
The Data analyst intervene daily in companies and constantly seek relevant data according to the directives they receive.
For example, a merchant site uses data analysis to determine the behavior of visitors to their site.
Studies conducted can uncover user behavioral trends while using multiple and complex data.
Depending on the results, the company can decide what to do next in terms of loyalty, additional offers or even the restructuring of certain pages of the site. Obviously, the more important the data collected upstream, the more precise the lessons learned.
Another example, during sales periods, the stores will decide in advance on the discounts to be made. Depending on the first sales trends, the discount may be accentuated or left at the same rate.
If an item is deemed sufficiently “fast-track” by a -20% discount, there's a good chance it will stay at that rate. On the other hand, if the Data analyst notices that an item on sale at -20% does not see its sales boosted, so the management of the offer will certainly make the decision to upgrade this item to a higher discount.