Human Resources a smart path to Intelligent Enterprise value

10 October 2018

Despite all the attention that goes to emerging technologies and disruptive tech, Hans-Petter Mellerud, the founder and CEO of Zalaris, is convinced that people will remain the prevailing factor within organisations.

As the CEO of a company exclusively focused on HR, human capital management and payroll solutions since I founded Zalaris nearly two decades ago, my passion begins with the people we help in the world – including many beyond the walls of businesses across Europe. Yes, it’s the human side of what we do that I’ve always considered most rewarding. This probably comes as no surprise, given our profession, of course. On the other hand, we can all thank technology for giving us new advantages (as well as more than a few headaches) over the years.

What’s particularly astounding right now is the rapid convergence that’s occurring between data, human intelligence and decision-support capabilities. We tend to forget that so-called “artificial intelligence” would never be possible without, well, people. In other words, all roads eventually circle back to the brains, talents and skills that walk in or log in remotely to the office each day. They naturally rely on technology like never before – yet may not fully recognise how much “smarter” or more impactful they’re becoming (or can become) in the process.

Intelligent Enterprise

The point is that the moulding of tech and cerebrum is what will truly soar to epic proportions in the years ahead. Unprecedented advances are already being made in practically every field, from medicine and manufacturing to travel and tourism, at paces that were simply unheard of even a few years ago. All this industry progress boils down to two words that are redefining the competitive landscape: Intelligent Enterprise.

Human Resources a smart path to Intelligent Enterprise value

Intelligent Enterprise is by no means an altogether new concept. The roots of this inexorable movement date back to the early 1990s when the approach to management came into existence – in words and theory as well as in practice via a more rudimentary sense of applying technology and emerging service paradigms to improve business performance.

The barriers to realising the Intelligent Enterprise back then were many. Doors to getting there are plentiful today, and the overall platform chosen will make a profound difference. With modern cloud-based Human Capital Management (HCM) solutions, HR can literally serve as the backbone to smarter, more nimble operations with better-trained personnel and sharper insight into what to do and not do at any given time.

Efficiency gains are more than a by-product of the Intelligent Enterprise. For each task that can be automated, people can devote more attention to the ones that truly need their intellect. Instead of draining mental acumen on mundane matters that previously required interaction with others or tedious HR paperwork, for instance, the proficiency essentially feeds on itself, regenerates and extends into other areas.

Where the Intelligent Enterprise ultimately leads to is anybody’s guess. Although Intelligent Enterprise is an end-to-end ambition that doesn’t really reach an “end state,” as it embodies the notion of continuous improvement, many dimensions are now well within reach – especially within HR’s reach and influence where it counts the most: in identifying, recruiting, hiring, compensating, assessing, developing, training, inspiring and empowering people.

Powering up the edge

People are what really fuel the Intelligent Enterprise. It’s how they use and evolve the technological edge that’s made available to them and instilled in the company’s culture. To this end, the use of HCM adds value, as it facilitates HR best practices and processes. The commitment to HCM solutions needs to be “top to bottom” and resolute, reinforced at every juncture, whether procedural or strategic in nature.

“While technology gives us new advantages, I am thoroughly convinced that people will remain the prevailing factor.”

In a new report developed by experts at Zalaris, titled ‘HR 2020’, it was found that the human resources profession is facing a number of key trends that will change the landscape. All of the developments correlate and contribute to the Intelligent Enterprise, nurturing it in a tangible way that can be more elusive in other areas of any organisation. HR leaders need to seize the opportunities now and turn their own knowledge into a more invaluable asset.

A number of takeaways that are intrinsic to the future of HR as well as the Intelligent Enterprise:

  • Shared Services will continue to grow, increasing efficiency and cutting cloud costs
  • Automation will create more strategic HR functions – not fewer overall jobs
  • Advances in analytics will drastically improve recruiting, onboarding, training and more
  • BPO value, design thinking, mobile, and security will drive HR decision making

Personally, I am thoroughly convinced that people will remain the prevailing factor. Without the right technology platform and solutions, however, most will never be as “intelligent” as possible in business and likely to fall more than a bit short in reaching their full potential.

AI can improve operational risk management in banking

17 April 2019

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.