Capgemini Invent acquires French digital marketing consultancy

08 October 2018 Consultancy.eu

International professional services firm Capgemini has in home country France acquired June 21, a consulting firm that specialises in digital marketing. The agency will be integrated into Capgemini Invent, the group’s digital consulting, innovation and transformation division.

On 12 September, Capgemini unveiled its new business unit that bundles its services in the area of management consulting, innovation, design and digital transformation. The powerhouse of over 6,000 employees was formed through the joining of forces of six units: Capgemini Consulting (management consultancy), LiquidHub (customer engagement), Fahrenheit 212 (innovation consultancy) and three creative design agencies – Idean, Adaptive Lab and Backelite.

Under one month following the announcement, Capgemini Invent has now closed its first acquisition under the new banner, building on its two previous acquisitions this year, that of Adaptive Lab in June and LiquidHub in February. The bolt-on of June 21 sees around 30 employees join Capgemini Invent in France, and bolsters the firm’s digital marketing service portfolio.

“June 21’s truly entrepreneurial spirit and well-established expertise will enable us to offer our clients an enriched portfolio of advanced digital services. I am delighted to welcome them to the Group,” said Paul Hermelin, Chairman and CEO of Capgemini. Cyril Garcia, who is the top boss of Capgemini Invent, added “In June 21, we found an exceptional team able to deliver on the customer-led transformation challenges faced by our large clients.” 

Capgemini Invent acquires June 21

Founded in 2007, by Jean Pierre Villaret (former France & South Europe Regional Head at Young & Rubicam) and Jean Marc Benoit (formerly General Manager of French marketing agency DevarrieuxVillaret), June 21 advises its clients on marketing and communications in the digital era. The Paris-based company works for clients such as Carrefour, Orange, Kingfisher, SNCF, Veolia and AG2R La Mondiale.

Asked about the rationale behind joining Capgemini Invent, June 21 co-founder Villaret pointed at the possibility to provide an end-to-end service to its clients. He further pointed at the wealth of technology expertise present within the Capgemini group. With over 190,000 employees globally, generating revenues of €12.8 billion, Capgemini is one of the globe ten largest IT consulting firms, system integrators and technology outsourcing providers. The company also operates at the forefront of technology-led innovation, investing hugely in emerging technologies such as artificial intelligence, machine learning, robotics and the internet of things, among others.

“We founded June 21 to enable our clients to make the most of the digital revolution that would transform their marketing and even for some, their business models. The arrival of artificial intelligence in our daily lives is a breakthrough that triggers new challenges for our clients, all leaders in their markets. We were convinced that we could not remain effective both as a force for strategic propositions and as a creator of content without the support and power of a reference partner. Capgemini will allow us to develop new expertise while preserving our know-how and the originality of our approach. This is a very positive move for both our clients and employees.” 

Capgemini purchase in France comes two weeks after onepoint, one of the country’s larger home grown IT consultancies bought weave, adding 400 experts in France to its footprint. Earlier this year in France’s consulting industry, KPMG acquired Carewan and MAPP, Sia Partners purchased digital marketing agency Fove, while French operations consultancy Argon Consulting merged with UK's Crimson & Co.

Related: David Williams on why Capgemini regrouped to launch Capgemini Invent.

AI can improve operational risk management in banking

17 April 2019 Consultancy.eu

Risk management is an integral part of banking. By taking financial risks, banks are able to generate the profits that are necessary to survive. Risk management aims to control this process by making potential losses more predictable. This makes the bank more robust to external fluctuations.

Whereas profits can be made by accepting certain financial risks, operational risk is intrinsically different. Operational risks only cause losses – financially in terms of bottom-line impact and non-financially in the form of for instance reputational damage. The consequences of operational risk events can have a large impact on an organisation and the financial system as a whole as experienced during the last financial crisis. It is not surprising therefore that operational risk is receiving more attention within the financial sector, with banks trying to minimise the operational risks they take, given the resources available while keeping in mind the strategic goals of the organisation.

In the past decade, there has been major progress in the development of artificial intelligence (AI). AI algorithms excel at data analysis and have evolved to the point where they surpass human performance for a wide variety of tasks. More and more businesses exploit these technological advances to optimise different kinds of processes such as marketing, sales and e-commerce, manufacturing and logistics. In today’s growing data-driven world, this trend is expected to continue on the back of widening opportunities for use cases, including in the area of operational risk management.AI can improve operational risk management in banking

Challenges in operational risk management

In June 2011 the Basel committee published the Principles for the Sound Management of Operational Risk (BCBS), which provides a framework for the development of proper operational risk management. Three years later, a survey was conducted to measure to what extent banks complied with these principles. One of these principles states that banks should write a risk appetite and tolerance statement. Banks reported that this is more challenging for operational risk than for other risk categories and attributed this to the nature and pervasiveness of operational risk. The banks that did comply with this principle often reported the use of backward-looking metrics of operational risk, such as operational losses as a percentage of gross revenue.

The above example underlines the challenges which banks face in the management and measurement of operational risk. Compared to financial risk, operational risk is a more qualitative field of study. Whereas financial risk management has been the main priority of banking for a longer time, operational risk management is much younger resulting in less extensive historical data. Predictive modeling becomes more of a challenge in this situation.

On top of that, the events in operational risk are much more diverse in scope. The Basel committee defines operational risk as the “risk of loss resulting from inadequate or failed internal processes, people and systems or from external events”. Internal fraud, data leakage and reputational damage are very different problems, yet can be very closely related as well. A lot of the data in operational risk consists of textual input which contain qualitative information. The qualitative nature of operational risk is reflected in the Basel framework, which encompasses guidelines for organisational structures, culture and awareness, and qualitative reporting.

The computer as a reader

Artificial intelligence could play a valuable role in upgrading operational risk practices, with in particular machine learning – the field of self-learning computer algorithms – showing promise. Machine learning algorithms can make predictions based on data fed to the algorithm. Recently, major progress was made in the field of natural language processing (NLP). NLP focuses on using textual data for predictions. As an example, an algorithm can learn to rate hotel reviews. By processing large amounts of reviews together with their given ratings, the algorithm can learn to give a rating to a new review it has never seen before.

“There are many opportunities for operational risk management to exploit AI and other related technological advances.”
– Lars de Ruiter and Matthias Geerse, Solid Professionals

In operational risk, many textual reports are written regarding specific risks or risk appetite of the organisation as a whole. These reports are generally written by risk managers – experts in their field. Assuming that risk reports hold information which is absent in historical loss data, this information could be extracted and used for predictive purposes. So why should computers perform this task instead of risk experts? The answer lies within the physical limitations of human being – as well as the natural biases they have to cope with. While humans may struggle to remember a piece of text, a computer algorithm easily processes thousands of books and finds structure within using statistical methods with near-perfect precision.

Towards prediction of operational risk

The performance of NLP algorithms using word embeddings has increased tremendously of late. Using the embedding as a starting point, these algorithms learn the meaning of full sentences and use this information for further predictions. This is more generally called sentiment analysis. A popular application in the financial setting is predicting stock price movement from news articles, or twitter feeds. A recent paper by Denmark’s national bank predicts corporate distress of firms by analysing their annual reports. Taking textual data under scrutiny can make a contribution to quantitative predictive modelling, even in the financial sector where the use of words is very different to day-to-day language.

In summary, there are many opportunities for operational risk management to exploit AI and other related technological advances. One possibility is the classification of risk events. As an example, a computer algorithm can read risk descriptions written by risk managers and classify them according to their impact and frequency. Combining loss data with risk reports, improved prediction of risk events might lead to more accurate prediction of future losses. A different application is the measurement of more abstract concepts such as the financial health or maybe the cultural aspects of a company.

An article by Lars de Ruiter and Matthias Geerse. Both are finance & risk consultants at Solid Professionals, a consultancy from the Netherlands.