TBI
Admiral Pay's e-Wallet: an innovative process for an innovative product
Chirale built one of the first complex national-scale systems with a microservices architecture, allowing the customer to dramatically cut the time-to-market for releasing new products and features. Find out how.
The market for electronic payments is one of the fastest-growing sectors in recent years.
According to a study by the Politecnico di Milano Observatory, already in 2020 — despite a decline in consumption — the volume of electronic transactions had grown by almost 5 percentage points, with a trend that is set to increasingly displace the use of cash.
It is therefore easy to imagine how high competitiveness is in this sector, and how challenging the task of Admiral Pay is in becoming one of the leading circuits at national scale.
A sector marked by very binding mandatory regulations, constant compliance requirements and frequent changes to the reference rules, high technology and competition based on service quality and customer focus.
It is in this scenario that the recent release of the e-Wallet system, commissioned by Admiral Pay to Chirale, takes place.
The result was an innovative, strongly customer-focused product, designed, built and continuously kept up to date through the use of innovative methods and technologies — one of the first major use cases of microservices architectures and DevOps development model in Italy.
Who is Admiral Pay, what is an e-Wallet and why is it important, and what does the innovation we just mentioned actually consist of?
In this article we will try to answer, telling you about our latest major project.
Admiral Pay is the Payment Institution of Novomatic Italia, the Italian subsidiary of the large Austrian group, world leader in gaming machines and systems, State Concessionaire for the telematic control and operation of AWP and VLT systems.
Controlling fiscal compliance in the legal-gaming business and fighting money laundering are two of the main goals of this kind of operator. Discouraging the use of cash in favour of electronic, traceable payment methods is one of the main tools to pursue these goals.
Admiral Pay is, to all intents and purposes, an Electronic Money Institution, subject to authorisation and supervision by the Banca d’Italia.
The offering of effective, easy-to-use services that meet consumers’ needs is a fundamental factor in differentiating and making one’s offering competitive. The e-Wallet tool was developed precisely in this sense.
E-wallet is short for “electronic wallet”, literally a digital wallet — a virtual instrument that can hold payment cards, loyalty cards and enable any kind of purchase or money movement through a smartphone or other electronic devices.
The e-wallet replaces the cash in the consumer’s physical wallet. Payments are made using the funds deposited on cards or current accounts linked to the e-wallet.
Transactions must be carried out in total security through dedicated apps designed to deliver high performance in compliance with all the standards required by law.
Admiral Pay’s e-wallet allows users to collect gaming winnings, make payments of any kind, withdrawals and money deposits without thresholds — like on a regular bank account — top up phone or other subscription services, pay bills, fines and taxes.
Through this new service, every POS (Point of Sales) on the Admiral Pay circuit becomes a one-stop shop, allowing the customer to carry out all the transactions they need.
The challenges of this kind of project are essentially two: ensuring maximum responsiveness to market needs by releasing improvements and new services very quickly, and ensuring high efficiency in transaction handling without compromising reliability and security.
To handle both challenges at once, you need not only suitable technologies but also the adoption of an innovative software-development-and-release method.
The classic organisation of software-production activities provides for a sharp separation between development activities (development) and operations management (operations).
This kind of methodology is also known as the silos model — that is, a compartmentalised approach.
One part of the organisation implements the development cycle, usually through an agile model that iterates among requirements analysis, design, software development and testing, until a stable, tested release of a set of application features is reached.
Once development and testing of a batch or release are complete, the software is handed over to the organisational unit that manages operations, or operations, and the software is rolled out to actual use through a dedicated “go-live” or roll-out procedure.
Any problem found during roll-out leads to a roll-back to the previous stable version and the return of the software to the development cycle.
This kind of organisation is in fact present in all large Italian companies. The two units — development and operations — are often constantly in conflict, leading to bureaucratisation and inefficiency in the process of updating and innovating the software.
In the field of Software Engineering, for projects releasing innovative products where time-to-market is a decisive factor for the company, a different model has been theorised, in which the two phases of development and operational release are managed as a continuous cycle.
This model is called DevOps, from the blend of the two terms Development and Operations.
The actual implementation of a DevOps model within a large company is not an easy goal to reach.
You have to overcome a series of obstacles and resistance on the part of the organisation, solve methodological, practical and technological problems. The risk is that of releasing products that are not valid in terms of features, efficiency and security, or of failing to actually implement the model, missing the goals tied to the speed of releasing new products.
In the case of the project commissioned by Admiral Pay, we managed to concretely achieve this goal. The first release of the new e-wallet was rolled out and deployed at the Points of Sale selected for the pilot phase in December 2021.
The pilot phase was passed brilliantly, and in January the rollout of the new service was launched across the entire national network, alongside the distribution of a new multifunction terminal model.
Since then the system has been the subject of numerous changes, requested by various players in the chain as the use of the new product and new terminals ramped up — with the resulting continuous releases demonstrating the organisation’s high responsiveness and the ability to release new features or improvements within hours rather than weeks or months, as was previously the case and as is still the case for most competitors.
This important result was achieved thanks to Chirale’s ability to handle complex, highly innovative projects.
Several factors contributed to the success. First, the involvement of the entire network of companies that work on a stable basis at the FabLab Ostiense was essential. This was, in fact, the first major project carried out at the Digital Innovation Hub of Rome.
This made it possible to draw on the best skills present in the Hub’s ecosystem and to industrially leverage the results of the continuous research process that has been taking place under an open innovation regime for several years now.
In addition to the skills held in the fields of value-added electronic services and cybersecurity, widely present at our Hub, specific Software Engineering technologies were leveraged — such as containerisation and orchestration — to solve the well-known issues that plague the DevOps model and usually cause it to fail.
But what are the technological innovations Chirale brought to the table?
First, the system was designed according to a software architecture known as “microservices”.
Compared to a traditional architecture, the system is broken down into a larger number of elementary features, each dedicated to a precise task — as elementary and reusable as possible in the implementation of more complex services.
Each microservice is then developed according to its own autonomous software-life-cycle.
The project must be led by experienced, highly specialised staff, capable of designing a correct decomposition into atomic, truly independent and reusable modules. Ideally — but also practically — each software component implementing a microservice must be releasable to operations without significant disruption to the rest of the system.
The integrated, continuous DevOps development-and-release cycle is strongly reinforced by the presence of a microservices architecture.
The main problem affecting the operational release of new software is compatibility and interactions with the current configuration of operating systems and middleware on the servers.
To address this issue, the technology of so-called containers was developed — environments comparable to virtual servers dedicated to individual modules or software systems, isolating them from the environment, ensuring immediate deployment and possible relocation to different physical servers.
In our case, the Docker container technology was used, which represents the state of the art in this field.
Using containers to deploy and operate the microservices that make up an application raises the second important problem: managing the complexity due to the proliferation of containers.
To solve this further issue, dedicated technologies known as “orchestration” have been developed, allowing operations managers to easily and securely govern the set of containers.
In our case, the orchestration technology chosen was Kubernetes, another state-of-the-art example in its field.
The e-Wallet system of Admiral Pay — to the best of our knowledge — is one of the first complex applications, distributed and used at national scale, that in practice implements a microservices architecture, containerised and orchestrated within a DevOps process.
The end result is a system that allows Admiral Pay to react in real time to market needs, always staying one step ahead of the competition.