“Data science is all about arming people with more information to make more informed decisions,” said David Pardoe, group head of data science at Hays.
Pardoe is a data science and analytics professional with over 20 years’ experience in the field, previously working with several consulting firms, including Accenture and Capgemini.
He has applied advanced analytics and machine-learning techniques in the CRM and marketing arena.
Prior to joining Hays, Pardoe was science business partner for EMEA at AIG.
Hays is a London stock exchange-listed company providing recruitment and HR services, with a legacy that extends back to 1867.
The company operates in 33 countries worldwide.
As head of data science at Hays, what does your role involve?
Looking across our business to identify where and how we can use data science methods and techniques to enhance decision-making across our business; making it more effective or more efficient.
In order to deliver data science across our business, I have been establishing a data science practice from scratch; developing and building the core capabilities required to make data science successful.
How is data science contributing to the digital transformation of companies?
Data science enables key insights and action triggers to be derived from the vast and growing quantities of the data that the digital world is creating.
The emergence of methods and technologies that can handle this data is unlocking the ability to meet customers, clients, and employees’ expectations in the digital world.
Can you cite good examples of companies that are being transformed through data science?
AIG has gone through a significant data science journey; in 2012, the CEO hired a chief science officer and had him report directly to him (rather than to a CIO or some other functional areas).
A science practice was rapidly established and grew to nearly 200 people. The majority of these people have now migrated into key business and operational roles across AIG.
This means the AIG has many data science practitioners in key roles and hence, it is central to the strategy and decisions the company is making.
Is data science going through a hype curve or is it the real deal for business in the 21st century?
There is no question there is a hype curve – data science has existed for longer than my working life, and I have been working for over 20 years.
The main hype is that the tools and technologies can help anyone do it; this is simply not true.
You need experience and expertise, but further[more], you need the right blend of different skills. Too often, individuals have skills in only one aspect of data science and it is essential to blend these skills.
How should companies be preparing their staff for this new business reality?
Working on teaching all types of decision-makers that data science can make a difference, but that it is not about replacing human decision-making.
It is all about arming people with more information to make more informed decisions.
In addition, staff should be encouraged to recognise that capturing data and information that represents what actually happens in an organisation is the fuel that drives data science. Commonly, many members of staff do not see the benefit of entering data and information accurately and comprehensively; they see it as a burdensome overhead.
How should organisations be managing their legacy data or structuring existing data for better real-time insights?
First, identify [if] you really need real-time insight – if the business process or workflow is not affected by real-time data, then your data science and insight does not need to be real time.
Second, it is critical to understand the structure of the data around the company and how that structure relates to how it is used in real business processes. 99pc of companies operate with systems that run on data – this data has a natural structure and order that maps to business processes. Understand this and utilise it, and you are one step ahead.
How must we prepare future workers to excel in this area?
Again, focus on teaching and training about what data science actually is; not how to do it, but how to interpret and use the outputs.
What thoughts do you have on the trends and big shifts in the technology world and what forecasts do you have for the evolution of data?
The data and data science world simply evolves – I see many examples of the challenges and issues and hype around data and technology that I saw in my first job back in the 1990s.
The biggest shift in recent years is firstly towards infrastructure that is not on-premises, and therefore does not have significant capital outlay.
Secondly, the evolution of open-source tools and technologies also unlocks the ability to deliver great data science outputs with minimal capital outlay. Future evolution will be all about embracing the other challenges that these two shifts create – primarily skills and talent to utilise them effectively!