The A-Team


A force field of analytical power; QVARTZ Analytics is definitely the data whizz-kid of the QVARTZ family. With a knowledge base ranging from computational engineering to mathematics, digital marketing, data science and algorithm development – just to mention a few – the combined skills of the QVARTZ Analytics team are applied in everything from major digital transformations to artificial intelligence. Pair this with the eight different nationalities represented in the +10 strong team and ‘Diversity’ comes out as the given team screen saver. Read on as the people behind QVARTZ Analytics unravel their take on the three key elements that rock their world: Data, Analytics and People.

A connected future

After four decades of exponential increases in computing power, the world’s immense amount of processing power is now doubling every two years, which is leading to astonishing leaps forward in technological capabilities. As technology is continually becoming cheaper, the demand is met at lower price points, fuelling an explosion of devices with endless connections. Sophisticated artificial intelligence devices are now mass-market and better known as personal assistants by the names of Amazon Alexa, Apple Siri and Google Assistant.

Earlier this year, we were paid a visit by Peter Schwartz, one of the world’s premier futurists, innovators, authors and business strategists. According to him, we can expect technology to have a vast effect on mobility, connectivity, speed of change, intelligence and productivity in the future.

The combined effects of new technologies – mobile, cloud, artificial intelligence, sensors and analytics, among others – are accelerating progress exponentially. And as soon as we overcome the physical and chemical limitations that are inhibiting exponential gains in mass-market technologies such as battery storage and wireless charging, it’s likely that the pace of change will accelerate even faster. In our increasingly digital world, says Mr. Schwartz, data is the new oil, and we are the wells. However, a vast majority of the data we gather today is not useful – and 97% of the data that is indeed useful has yet to be analysed. The rise in data and the reduced cost of accessing and processing it has fuelled organisations to utilise even more data to better understand their customers, create new products, offer new services and optimise their operations. All with the aim of creating a new competitive edge.

‘A’ for Analytics

Say you want to understand the performance of a specific brand. The classic approach would be to do a survey or to poll a group of customers. An alternative way, however, would be to use social media data and sentiment analysis to map brand perception by consumers across all channels. By combining such insights with the sales data, a new kind of transparency is created, allowing a deep understanding of the brand and its impact on revenue. That’s the power of analytics.

For the QVARTZ Analytics team, focus always revolves around one decisive question: how can data be utilised and analytics applied in order to help clients implement a strategy and/or solve a business problem? When looking at a business strategy, the team first articulates what is happening in the specific business today; then they predict the future of the industry and lastly, they make critical choices as to where the scarce resources of the business should be placed in order to maximise value in the predicted future.

To scan what’s going on in a specific business – instead of collecting samples and extrapolating the information – QVARTZ Analytics uses a large volume of structured and non-structured data from a variety of sources across the organisation for processing and analysis. Why? Because this approach reveals much more insight than what can be deduced from the traditional approach of analysing several data sources independently and drawing cross-data conclusions afterwards.

In the great majority of businesses, a fair amount of energy is spent on considering, evaluating and sometimes worrying about what lies ahead. In order to predict the future, statistical tools have traditionally been used to create trend lines – by looking at one dimension at a time. For example, a 12-month rolling sales forecast has been predicted by looking at the historical trend of actual sales. More complex predictions such as market trends, brand development and competition, which all include multiple internal and external variables, have, however, been considered too time-consuming and costly, inhibiting organisations to even think in this direction. But with the growth in technology and the use of machine learning, the cost of prediction has been reduced in such a way that they are becoming part of new disciplines that were not even on the horizon a couple of years ago.

In QVARTZ Analytics, machine learning is employed to predict the critical outcome of various business parameters. As an example, by identifying patterns between sensor data, machine utilisation, operators, last maintenance cycle, weather conditions, etc., critical-asset breakdowns can be predicted. Another application is designing recommendation engines, which can predict the products that customers are going to buy next. Customer churn, future demand curves, optimal pricing for a new product and store performance are other examples of how businesses can be enabled to make better decisions driven by data.

On top of machine learning, artificial intelligence (AI) is the new analytics frontier; in addition to exploring the strengths of machine learning, AI interprets the context of user needs and predicts what might be the most suitable response to address those needs. This entails a quantum leap from reactive marketing to prescriptive marketing: it’s no longer about understanding which product a costumer will purchase next, but about pushing a new product into the market in the most effective way. QVARTZ Analytics uses such technologies to simulate various complex scenarios for businesses to better understand how they can allocate their resources as well as which risks and rewards are associated with each move.

There is no doubt that machine learning and AI are here to stay. However, maximising the value from analytics, AI and machine learning isn’t just about technology, but just as much about what drives all technology: people.

The human factor

QVARTZ Analytics focuses on three key aspects to extract maximum value from advanced analytics, AI and machine learning: people, skills and change management. First and foremost, the area of application and usage of machine learning and AI are still driven by people. While machine learning can automate and make decisions that are performed by human beings today, there are still many strategic and at times ethical choices, which cannot – or should not – be made by machines.

Situations that require ethical judgement or emotional intelligence are not easy to articulate by algorithms and hence difficult, if not impossible, to teach a machine. As an example, AI can predict the probability of a patient to wake up from a coma, but the actions and medical decisions following such a prediction can’t be made by machines. A holistic and human view has to be adopted. Machine learning and AI are continuous processes. The machines learn from the data they receive, including human responses to their predictions, so that AI can improve and understand the wide complexity of the choices a human mind makes. To achieve the best results, human beings and machines need to work together.

Machine learning and AI are continuous processes. The machines learn from the data they receive, including human responses to their predictions, so that AI can improve and understand the wide complexity of the choices a human mind makes. To achieve the best results, human beings and machines need to work together.

The second key element to maximising the value from analytics is to upgrade the skills of people working in a human + machine environment. As an example, organisations might traditionally have maintenance staff who understand the mechanics of the assets they are maintaining. It’s not a trivial task to train them how to utilise machine learning-based predictive maintenance. The engineers need to unlearn the old ways of doing maintenance by just following a maintenance cycle and acting upon the predictions from the machine learning models. The same is applicable across almost all divisions of an organisation, be it Marketing, Sales, Forecasting, HR or others.

The third and most important aspect is to understand that machine learning and AI not only represent technological evolution, but also require change from a human standpoint and thus require change management. However easy it is to calculate, identify and communicate how machines will improve productivity, equal attention needs to be given when defining the new roles of the teams involved. If the change aspect of the workplace is neglected, the misconception of humans vs. machines will be increased when in fact, a new way of collaboration of human beings and machines is what is required.