Relevant case studies

At Deko Data Analytics we have worked with different sectors, obtaining specialised solutions.

Data Architecture


Deko Data has collaborated in the analysis, design and implementation of the architecture of the new Cloud Data platform for one of the main insurance companies in Spain, end-to-end, following a Data Mesh model and the best market practices (DataOps, MLOps, FinOps...).The Data platform offers a complete framework that allows the company to be more efficient and profitable through the use of information and AI.

Data Architecture - Insurance

Optimisation of production and maintenance times for analytical cases. Use cases have been put into production in record time (half the time it took before the platform was available), and maintenance costs have decreased significantly, in some analytical cases by up to 10 times, saving hundreds of thousands of euros.

The platform's focus on security has enabled us to comply with insurance industry regulations worldwide in an efficient and meaningful way.

Data Advisory & Implementation


The Deko Data team helped the client realize an ambitious digital transformation program by creating a new Data platform as a driver of this. The program wanted to promote the automation of operations to improve efficiency and effectiveness. Apply a "cloud-only" approach to the platform, and democratize AI in the company. One of the most successful points was the implementation of a complete Data lifecycle model.

Data Advisory & Implementation - Energy

Improved Time2Market: The client has accelerated the productivity of analytical models through MLOps, reducing the time from ideation to implementation by 30 to 40%.

Significant cost savings: Estimated cost savings of 30% to 50% in analytics cases implementation, attributed to rapid productivity gains, component reuse, and efficient FinOps management in the cloud.

Operational enhancement: The data platform has standardized model productivity, improving analytical case efficiency by 20-40%, scalability, and team management within the client.

Data Factory


Implementation of a Data Engineering Factory, developing the client's different Data & AI use cases, following the best practices of the market and the client, with Agile methodologies. To carry out this task, a self-managed service has been implemented that covers the management and execution of the development of Data use cases, their testing and delivery. And the entire service is formalized with a continuous improvement plan.

Data Engineering Factory - Industry

Operational Efficiency: By standardising, centralising and productivising at scale the Data cases by Deko Data Analytics, we have reduced development times by around 20%, and increased the number of use cases that meet the needs of the Business.

Quality in deliverables: The number of changes and incidents have been reduced by more than 40%, improving customer satisfaction.

Data & AI Platform


Design and implementation of the client's new information and analytical platform on public cloud on which different information and analytical use cases have been developed to improve recycling rates and operational efficiency: optimisation of campaigns, optimisation of collection resources, productivity per container, collection patterns, etc.


Relevant use cases they solved for a Recycling Company 

  • Data & AI Architecture: The Deko team defined the new information and analytical architecture on Azure to develop information and analytical use cases such as operational optimisation to improve waste collection costs and recycling rates. 
  • AI-driven operational improvement: Deko has developed different algorithms to optimise the allocation of resources for waste collection. Thanks to the application of artificial intelligence, efficiencies of more than 10% have been achieved.
  • Improved recycling rate: thanks to the application of machine learning, we have managed to improve recycling rates by more than 5% through the optimisation of campaigns in the Horeca/residential market, optimisation of the container strategy, etc.
Expansion Strategy


Deko worked with client to bring Data & AI to their core. Deko designed the architecture to capture and analyse all supermarketers' tickets in real time. Different business problems, such as assortment, type and number PoS, discount coupon optimisation, human resource planning, etc., were developed based on the purchase tickets processing.


Relevant use cases they solved for Carrefour 

  • Store layout optimization: Deko collected all available data, combined it with new external sources, and designed and developed ML / AI models capable of optimizing the SKU store mix and the number of cashiers. Results yielded a decrease in nº of cash registers (€120 K annual saving  per unit), improved staffing, and customer satisfaction  
  • AI-driven expansion strategy: Deko designed and successfully deployed models that selected the best location for a new opening based on all the data captured (+1,500 data sources). The model forecasted the store’s revenue with a 95% success, which allowed the client to prioritize openings 
Maintenance Optimisation


The client wanted to optimise preventive maintenance tasks to minimise costs in the different projects and workshops. The resolution is based on applying genetic algorithms under business constraints such as the number of available resources, working day duration, type of tasks that can be chained, fleet availability and Service Level Agreements.


Relevant use cases they solved for rail company:  

  • The customer's objective was to define a solver that would allow them to minimise the cost of maintenance tasks. Previously, different approaches, such as commercial solutions or ad hoc developments, had been tried without success.
  • The Deko team proposed using different open-source algorithms, such as genetic or swarm algorithms (PSO), to find the optimal solution. Furthermore, the solution had to be economically and computationally optimal to solve the algorithm at the right time for its application.
  • The project result represented a saving of more than 5% in maintenance tasks.