TAKING OVER AND SCALING UP A DATA INTEGRATION PLATFORM
EXPERTISES
Vinci Énergies France Infrastructures Telecoms, which operates primarily under the Axians brand, supports the sector’s major operators in the roll-out, operation and maintenance of their fibre and telecoms networks.
To manage these activities, Vinci handles large volumes of diverse data: service requests, orders, commercial files, administrative documents, technician data, progress reports, audit reports, etc. This data is highly heterogeneous, partly due to the multitude of sources (customers, industrial partners, subsidiaries) and partly due to the variety of formats (structured files, emails, documents) and access points (SOAP and REST APIs, FTP, internal systems, ERP and SaaS tools).
BeTomorrow is supporting Vinci Énergies France Infrastructures Telecoms in the structuring, standardisation, development and operation of a data integration platform at the heart of this application ecosystem. The collaboration began in response to a highly operational need: to take an existing integration system – historically built using ad-hoc methods – and transform it into a robust, maintainable, well-documented and standardisable solution.
Over time, this foundation has evolved into a central integration platform, progressively enhanced with multiple connectors, business applications, interfaces, data processing functions, real-time data flows, dashboards and operational services.
applications and connectors developed
A comprehensive suite of applications has been built around the integration platform, and has been enhanced through successive iterations since 2016.
year the collaboration began
Vinci and BeTomorrow have been working on the platform, its connectors, its development and its operation for nearly ten years.
of disaster recovery time
The cloud infrastructure’s disaster recovery plan (DRP) has been reduced from several days of downtime to just a few hours, thanks to the Azure cloud architecture and the automatic provisioning of the entire infrastructure
CHALLENGES
Tech
Revamping and scaling up an existing but ad-hoc system
The initial system was based on macros, configuration files and processes built up piecemeal, with no documentation. Before any further development could take place, it was necessary to understand what was already in place, address vulnerabilities, standardise the code and map out each process. This groundwork laid the foundations for subsequent projects.
Standardisation of heterogeneous data
Incoming data flows originate from a wide variety of sources: emails with PDF attachments, EDI files, partner APIs and FTP servers. Each partner has its own formats, and these formats are constantly evolving. The platform accommodates this variability without compromising downstream processing.
Orchestration of a multi-component application stack
Over the years, the platform has been enhanced with numerous connectors, backend services and interfaces. Evolving this system coherently without disrupting what works requires constant architectural rigour. Each new connector must integrate with the existing foundation, rather than being added in isolation.
Migration and operation on Azure
The applications were initially hosted on infrastructure operated by Vinci. We gradually took charge of the migration to Azure, followed by operational management: monitoring, alerting, secret management, disaster recovery and business continuity planning, updates and maintenance to ensure operational readiness.
Exceptional software performance
Making data usable for business units involves more than simply storing it. It must be structured, validated and made available in formats that can be used by the teams managing operations on the ground.
Processing and routing of operational data streams
When a new data stream arrives, it is automatically processed using a generic sequence (data verification, cleansing, storage, routing, audit trail), as well as specific sequences (dispatch, notifications, aggregations, etc.). Some data streams are critical and linked to time commitments to end customers. Processing must be automated and reliable to ensure no deadlines are missed
Data utilisation for management
Data from business applications, management tools and customer flows is extracted, consolidated and presented to feed activity dashboards. Vinci teams can monitor volumes, measure lead times and produce reports without relying on manual processing or isolated files.
Agility
A long-term partnership
The partnership with Vinci spans ten years of continuous collaboration. It combines specific projects, application maintenance, ongoing enhancements and the operation of the infrastructure. This duration ensures consistency in technical choices and the ability to maintain quality over time, without each new team having to relearn the existing systems.
Managing priorities at a steady pace
Some periods have been structured around projects broken down into successive work packages. Others have been organised around overall budgets covering a range of ongoing enhancements and business requirements. In both cases, our teams maintain a documented pace and meet their delivery commitments.
DEVELOPMENT, DATA INTEGRATION, CLOUD: A COMPREHENSIVE SCOPE
The relationship between Vinci Énergies France Infrastructures Telecoms and BeTomorrow is part of a long-term partnership. It is not limited to a single project, but covers a coherent set of tasks: redesigning and ensuring the long-term viability of an integration framework, developing connectors, creating SaaS applications, interconnecting with partner systems, document management, automatically generating administrative documents, order integration, feeding BI tools, migration to Azure, maintaining cloud infrastructure in operational condition, monitoring, alerting, secret management and ongoing support for business changes.
Our projects combine both technical expertise and a deep understanding of the business. The systems involved are directly linked to Vinci’s field operations: processing service requests, meeting deadline commitments, routing workflows to the correct regional entities, utilising customer data, monitoring activity and providing reliable data to the business units.
BeTomorrow therefore acts as a partner in development, industrialisation, operations and consultancy.
GILLES MANTEL
Delivery Director
The strength of this project lies in the fact that it confronted us with the full reality of a production platform: delivering rapid development and regular releases, whilst also ensuring stability, monitoring, maintainability and the platform’s continued operational readiness. This tension between speed and reliability lies at the heart of a truly mature and well-equipped DevOps approach, which we have been able to implement with Vinci over the long term.
Our Client
Vinci Énergies France Infrastructures Telecoms, which operates under the Axians brand, supports the sector’s major operators in the roll-out, operation and maintenance of their fibre and telecoms networks.
A subsidiary of the Vinci Energies Group, the company draws on a network of branches spread across the whole of France. It works for operators such as Orange, Bouygues, Free, SFR and TDF on projects of national scope, with field teams covering everything from in-building deployments to long-distance networks.
LAURENT COULAUD
Head of Information Systems Security at Vinci.
Over time, we have been strengthening our collaboration through new developments and connectors. For example, we have developed software modules to convert service request emails into tasks within the service management tool. This saves the teams a great deal of time. We have also automated the integration of orders received in EDI format into the ERP system
FREQUENTLY ASKED QUESTIONS – INDUSTRIALISATION OF A TELECOMS DATA PLATFORM
How can we improve the reliability of telecoms data streams when the formats hardly ever change ?
In the telecoms sector, complexity does not always stem from rapidly changing data flows. It often arises from legacy formats – some of which are obsolete – that evolve slowly but remain critical to operations. The platform must therefore cope with heterogeneous files, historical rules, business exceptions and data errors that can take several months, or even a year, to be corrected at source. The challenge is to build a foundation capable of controlling, standardising and ensuring the reliability of these data flows without waiting for the entire upstream ecosystem to be perfectly clean.
Why can the obsolescence of formats become an operational risk ?
An obsolete format is not necessarily a problem as long as it is under control. It becomes a risk when it remains essential to operations but is no longer adequately documented, controlled or monitored. In such cases, any data anomaly can spread through business systems, disrupt processing or slow down teams. Standardising a data platform enables you to regain control: detecting errors earlier, tracing processing flows, securing data exchanges and maintaining service continuity despite the limitations of existing systems.
Do you always need to start from scratch to make an existing system reliable?
Not necessarily. In many complex projects, the existing system already contains valuable business knowledge. The real question is what should be kept, clarified, standardized or rebuilt. On VEFIT, the approach was to deeply understand the platform, secure its foundations, standardize the code and progressively build a more robust ecosystem around it. This limits disruption for teams while preparing the platform for future business needs.
What makes a business connector valuable in a data architecture?
A business connector is valuable only if it truly simplifies teams’ daily work and strengthens the reliability of exchanges. It is not just about making two tools communicate, but also about detecting errors in data flows, avoiding duplicate data entry, automating recurring tasks and securing processing between systems. Its value is also measured by its operability, meaning the auditability of its activity and its recovery time after failure. In a critical ecosystem, short disaster recovery times are essential to contain the risks of service disruption
How can a data platform remain valuable beyond its first release?
A data platform remains useful only if it continues to evolve in line with usage patterns, security constraints, business tools and technical standards. This requires a genuine ‘run’ approach: monitoring, access management, roll-outs, version upgrades, continuous improvement and regular prioritisation. Success therefore depends not only on the quality of the initial development, but on the ability to establish a sustainable rhythm of collaboration between business, technical and project teams.
