360º Customer Vision

Meliá Hotels International has more than 380 hotels open or in the process of opening in more than 40 countries under the brands Gran Meliá Hotels & Resorts, Paradisus by Meliá, ME by Meliá, Meliá Hotels & Resorts, The Meliá Collection, INNSiDE by Meliá and Sol by Meliá, in addition to a broad portfolio of hotels under the Affiliated by Meliá brand.

Meliá Hotels International is one of the world’s leading companies in the leisure hotel segment and its experience in this field has allowed to consolidate its position in the growing market of leisure-inspired urban hotels. Its commitment to responsible tourism has earned it recognition as the most sustainable hotel company in Spain and Europe.

Identifying the customers throughout their interactions along the different communication channels is essential to optimize the relationship with them.

The challenge Meliá had to address was to bring together all the digital customer traces generated from different channels and data sources, integrating these different origins with their corresponding volumes of information.

As prerequisites when addressing the challenge that was proposed:

  • Build a solution based on Open Source tools.

  • Enable Real Time query of tracking sources (possibility to send to key-value databases).

  • Retain the change history (ID of a user over time) with the date of changes for complete upstream and downstream traceability.

  • Support queries from the analytical environment with HIVE and Athena.

Solution on Amazon Web Services

Keepler designed a reference entity/table containing all the identifiers that would allow a unique identification of the customer and that would be scalable. This table would have the shortest possible response time.

Once the aggregation criteria for the different IDs had been defined, the exploitation process was reviewed. In this review, detailed information was obtained about the information aggregation process in ETLs and the integration of the data lake with Meliã’s different systems.

A Lambda architecture with three layers was implemented to support the ingestion and processing of data sources at different speeds.

Speed Layer

This layer is in charge of collecting the events produced in real time from the different websites. Using the Kinesis service, which already exists in the company and which contains the events, the information is pre-processed in real time. At the same time, this information is also stored in the Batch layer in order to feed the processes that require the rest of the sources to perform calculations.

Batch Layer

This layer is responsible for collecting information from systems whose information is incorporated into the platform at a slower rate, with frequencies of one or several times a day or even several days apart.

The information is received in raw form in a Staging layer. After performing the necessary transformations to make the user matching of all sources through ETL processes, the information is stored in a Main layer, where it can be exploited by different analytical use cases or those of a more batch nature.

Serving Layer

This layer is in charge of performing all the logic of serving the information of which is the unique user for the different ids of the different origins. It has a key-value database to quickly return this information in scenarios where this information is needed in real time.

An access through Athena is also enabled in order to be able to explore the Data Lake data.

In the event that greater visibility is needed on how all the platform IDs relate to each other and what their interconnections are, a documentary database is made available with this information, so that it can be exploited in real time.

In this context, the AWS solutions used are the following

  • AWS Lambda: Computation of events.
  • AWS S3: Storage of information in previous stages and ingested events.
  • AWS EMR: Batch and historical loading processes.
  • AWS DynamoDB: Master reference table to find the client by secondary ID and obtain the ID360.
  • AWS DocumentDB: for object storage.
  • Meliá’s main objective is to have the customer identified throughout all their interactions through the different communication channels (email, web, telephone, social networks, loyalty programs…). This will allow the optimization of communications by integrating with the customer’s recommendation and campaign personalization system, which will have to be adapted to this new functionality.

  • All these technologies follow the serverless/managed services development paradigm:

    • Avoids server and application maintenance.
    • Allows horizontal scaling of the platform according to actual usage.
    • It only generates cost for the time of use of the process.
    • Facilitates easy integration with the rest of the services offered by the platform, such as logging, virtualization or endpoint creation.

Keepler is a boutique company of professional technology services specialized in design, construction, deployment and software solutions operations of Big Data and Machine Learning for big clients. They use Agile and Devops methodologies and native services of the public cloud to build sophisticated business applications focused in data and integrated with different sources in batch mode and real time. They have Advanced Consulting Partner level and have a technical workforce with 90% of their professionals certified in AWS. Keepler is currently working for big clients in different markets, such as financing services, industry, energy, telecommunications and media.

Let’s talk!

If you want to know more or if you want us to develop a proposal for your specific use, contact us and we’ll talk.