Frequently asked questions (FAQ)

To make the predictions, the history of commercial behavior of each client is used: the purchases made, the type of products or items that they like the most, the moments when they like to buy the most, their receptivity to the commercial campaigns they receive from their company, etc.

In the same way that anyone can predict with enough certainty the behavior in the next short term of a person close to them, because they know their historical behavior at similar moments, a company can predict with a high level of accuracy how a client of his will behave from the commercial point of view.

By means of these predictions, a company can determine with a certain degree of reliability if a customer of its will like one of its products, if they will be receptive to a commercial campaign, if they will be comfortable with a specific offer, etc. Most consumers tend to like that the companies of which they are customers communicate offers, new products, etc…. as long as they are about interesting products for them.

Through Predictive Marketing, companies can individually choose the offers and products that are most likely to be of interest to their customers and are suitable for them, thus avoiding the inconvenience of offering them products that do not fit their profile, and helping them to choose among the large number of products and articles offered by mass consumption companies.

Predictive Marketing requires the use of tools and technology aimed at developing and implementing Advanced Statistics techniques, the application of Artificial Intelligence algorithms, and the handling of large volumes of data of different nature, which constitute the input or “raw material” so that these algorithms “learn” to make predictions. The necessary data are those that constitute the consumption history of each client and, in general, their historical commercial behavior in the company (operations, claims, purchases, etc.).

The use of these tools and of this type of data requires specialist profiles, who must combine their technological expertise with knowledge of the business of the company for which the algorithms are developed.

The company must provide two important elements: anonymous data on the historical behavior of its customers, and the interpretation and orientation of the algorithms from the point of view of its business.

The interpretation of the algorithms with a business perspective is very relevant, so that they are actually useful. In our experience, algorithms should not be applied “blindly”, but with the supervision of people in the company with deep knowledge of your business. It is the best way to take advantage of them and recommend products and services that are in the interest of your customers.

It is common to measure the ROI (Return on Investment) of the campaigns or commercial actions carried out by the company, to know the additional income that each campaign generates in direct sales and induced business. The ROI of a campaign is usually calculated by trying to answer this question: “How much additional revenue has this campaign generated, compared to the natural sales that are obtained without running the campaign?” Similarly, the ROI of Predictive Models can and should be measured, answering questions like these: “How much revenue has this model generated, used to select the target audience for a campaign?”; “How much revenue has this model generated, used to offer a set of personalized recommendations for each client?”

Bionline Apps automatically calculate the ROI of the models, so that our clients know at all times the return on their investment.

None. The company only has to contribute knowledge of its business, since all the specialized technical work of data analysis and application of Algorithmic and Advanced Statistics, together with the necessary technology to carry it out, is provided by Bionline if our client so requires. Bionline is the only company prepared to offer this service externally, when our client requests it, to provide a comprehensive Predictive Marketing service or to complement and enhance the capabilities that our client may have developed internally.

From the business point of view, Predictive Marketing tends to have a direct impact on sales, producing an increase in its figures through customer loyalty and loyalty, either by increasing cross-selling, upgrading consumption, reducing figures. of abandonment, etc.

From an internal point of view, the company’s business experts must have a good understanding of the models they use, the customer universes they are aimed at, the predictions they make and their business results, and they must acquire a work approach where decisions take into account what your customer data says.

Yes, our applications allow the company’s business analysts to manage the analytical engines of Predictive Models. They are designed from a business point of view, although they include technical aspects that help them understand and get the most out of these engines.

The applications act as intermediaries between the Models (whose direct management would require a specialized technical profile) and the business analysts, so that they enjoy the understanding and sufficient autonomy to carry out their work, without having to understand the mathematical and algorithmic background of the Models.