Customer Intelligence

problem

Intelligent recommendation.

By offering the same products to all customers, in most cases we do not meet their needs. Customers are disappointed when they receive a mismatched offer.

solution

In communication with the customer, we want to offer them a product that suits their needs. Upsaily recommendation engine is not based only on remarketing. Deciding whether to generate a proposal for the customer, we analyze the history of transactions searching for repetitive patterns of customer behaviors. Finding a pattern corresponding to a given customer, we can conclude about the probability of making a purchase of a specific product and then offer him a product that best suits his needs in a correct moment in time after previous purchase.

Recommendation engine can work as:

1. Frame on a webpage.
2. Automatically generated message delivered by a specific channel (mail, push notification, text message).
3. Banner displayed in social media.

problem

Customers segmentation.

We do not know who our customers are and how to satisfy their needs. We would like to define a different communication approach to each segment of customers, but we do not know how to divide them into groups.

solution

If we divide our customers by the value orders, frequency of shopping, their preferences etc into groups of people behaving in a similar way we can precisely tailor offers to the needs of clients, and thus build long-lasting and long-term relationships.

The automatic clustering mechanism using advanced machine learning techniques will indicate the optimal number of segments as well as the criteria for allocating customers to individual segments.

Another example of the segmentation is the division of customers into those who are mainly motivated by the price and the discount received, and those who prefer to buy a fashionable products where price does not matter. Of course, there is a number of intermediate segments that need to be looked after differently.