When we talk about the customer experience, the customer journey or putting the customer at the centre of all efforts, we usually think of terms like omnichannel, digitisation of processes or personalisation; this often sounds more like a business idea than a real commitment to service. That in and of itself is the first big mistake. Profitability, the pursuit of profit and success are the goals of a company. But none of these are conceivable, let alone long-term achievable, unless the customer is at the centre of the process through a dedicated solution.

There’s a lot of data, but it’s disjointed, unrelated, and useless.

We find ourselves in the business ecosystem of many industries with a too fragmented perspective of information in an environment that is more VUCA (VUCA stands for Volatility, Uncertainty, Complexity, and Ambiguity) than ever. In other words, the data originates from a variety of sources, in a variety of formats, and is archived in a variety of locations, and thus “hides” much more information than it provides.

On the other hand, we come across a variety of clients. We are not dealing with a homogeneous division, but rather with a business architecture that is highly fragmented, heavily reliant on heritage, and lacks integration. We’re dealing with a lot of data, but it’s segregated among business lines, activities, projects, and even channels, making it more difficult to generate meaningful and coherent insights with vision and complete information. In addition, although having a great amount of data, they are incompatible with the formation of patterns of consumption of energy products and services.

In the B2C business, intelligent information is essential for maintaining, acquiring, and growing the value of customers. However, data intelligence that gives the genuine value the market needs to give back to the customer far more than they are searching for or expecting from the brand continues to be established. To that end, it’s critical to gather and consolidate all customer data before archiving it in a single repository. It must be a consolidated and comprehensive representation that allows for a better knowledge of who each consumer is and what actions they take at each point of contact. Personal data, information on websites visited, purchase and consumption behaviour, transactional or privacy data will all be relevant in this race if they are fed back to each other, shared, and enriched.



The integrated handling of customers, also known as a unique customer identifier, in the new approach to the market will allow for a more precise understanding of their true needs and the projection of a single vision of those needs, regardless of where they are in their life cycle.

Our goal: a data-owning based paradigm

This data-owning based paradigm will enable the rational and intelligent utilisation of information in an appropriate context, directing the organisation also towards market-demanded services. With all necessary information readily available and accessible, it will be easy to incorporate systems that enable the development of insights using business rules based on decision models. In a nutshell, personalise.

Artificial Intelligence is becoming more powerful in B2C business. Channels that are automated, semi-automated, or human-assisted allow organisations to anticipate client needs while also allowing them to provide highly distinctive services that reinforce brand identification and their value proposition to the customer. This encapsulates one of today’s most crucial competitive paradigms: the requirement to be in the right place, at the right time, in the right channel, and ready to service customers who expect to be served. We need to pay close attention to the growing number of customers who want more control over their experience, because this represents a significant challenge for the processes, structures, and systems that businesses put in place to communicate with their customers.

If we want to provide the omnichannel experience that our customers expect, we must avoid disconnecting customers who register online from those who register offline.

Not everyone is equipped to meet this problem. The retail of today and tomorrow deserves a solution that takes into consideration their characteristics in order to give a genuine, dependable, and error-free solution that includes grouping all the digital traces of customers from different channels and data sources. All sources of data will be integrated with their corresponding volumes of information; users are to be supplied with tools that enable fluid engagement with suppliers while also facilitating self-management of their contracts, services, and solutions via any channel.

A user case powered by Keepler: Meliá.

Meliá’s key goal was to retain the consumer identification throughout all of their contacts via the various communication channels (email, web, telephone, social networks, loyalty programs…), Keepler helped by creating an advanced solution that  would enable communication optimization by integrating with the customer’s suggestion and campaign personalisation system, which was in need of an update to accommodate this new capabilities.

To that end, a reference table/entity was created that had all of the identifiers that allowed for the client’s unique identity and was scalable. This table had the quickest response time possible.

The exploitation process was examined after the aggregate criteria for the various IDs were defined. This assessment gathered thorough information about the information aggregation process in ETLs and the integration of the data lake with Meliá’s various systems.

To accommodate the intake and processing of data sources at varying speeds, a three-layer Lambda architecture was built:

  • The Speed Layer managed the gathering of real-time events generated by the various websites.
  • The Batch Layer was charged with gathering data from systems whose data is incorporated into the platform at a slower rate, with frequency ranging from one to many times per day or even several days apart.
  • The Serving Layer is responsible for carrying out all of the logic for serving the information, which is the unique user for the various ids of the various origins.

If you know how, you can identify the consumer using a unique ID independently of the channel via which they connect with the brand. And we at Keepler know how to do it 😉

Image: Unsplash | @georgeprentzas


  • Ernesto Diezhandino

    Technical Writer at Keepler. "I've been a technical writer and instructional designer for different industries for a decade now and I still haven't stopped learning. When I'm not reading and writing about new methodologies you can find me writing science fiction."