Data Scientist 2.0: Who They Are – And How We Placed 50 In Scotland
Increasingly fast moving, transformational businesses require key staff with data science and analytics skills. Today, we are seeing this spread even more to traditional businesses. As we see new approaches to data enabled business and decision making, together with the increase in high quality insight and analytics tools, it is clear that data has a pivotal role to play in enhancing the prospects of commercial and organisational success.
However, this won’t happen in isolation and without the skills necessary to wrangle, govern and exploit the data, we’re left with little more than an IT enabled filing cabinet. But what of these skills?
Can we meet the data scientist 2.0 demand? The current shortage of skills globally demonstrates that business-as-usual strategies won’t satisfy the growing need. If we are to unlock the promise and potential of data and all the technologies that depend on it, employers and educators will have to transform. But so do candidates, and that’s where we at MBN come in!
Late last year, MBN Solutions competed in open tender and were selected as partners by Scotland’s Data Lab to deliver their MSc. Placement Project – a project specifically focused on looking at the intersection of the skills needed by business, by candidate data scientists and those taught by academia.
The Data Lab is a government funded Innovation Centre. Its core mission is to generate significant economic, social and scientific value from data. As part of this mission, the Data Lab funds eligible students through their MSc. courses at a variety of Scottish Universities. These courses are handpicked as representing the pinnacle of data science technical and related business skills to help prepare students for real world assignments. As part of these courses, students are offered the opportunity to participate in 10 to 12 week long Industrial Placements – giving them real-time and real life exposure to commercial Data Science challenges at a variety of Scottish businesses.
Having placed nearly 50 such data scientists, we’re now in a position to reflect on our experiences here and what that means for Employers, Data Scientists and for the Academic realm of Data Science.
Over the course of the last decade, Scotland has become a data science Tour de Force and with the Data Lab helping to push this even further, it should come as no surprise that businesses located in Scotland are benefitting from the availability of high quality, fit for purpose Data Scientists.
The purpose of this exercise, backed by the Data Lab, was to ensure that the next generation of Data Scientists would benefit Scotland even more, by being equipped with commercial knowledge and employability skills to complement their deep technical competencies.
MBN ensured that the 50 Data Scientists placed had the very best prospects of a great match, real opportunities to add value to their host organisations and the real prospect of developing soft and employability skills to enhance their future employment chances. In short, our goal was to create and place the Data Scientist 2.0 in the very best businesses in Scotland.
So… what did we find?
Well, we suspect there will be few surprises here, but what employers really want are skills, and not just qualifications. For these skills, many are actually in the area of soft skills such as interpersonal and team based competencies mixed with a commercial readiness to work in the labour market. Skills to facilitate greater collaboration and appreciation for business strategy and not just deep technical skills, are most frequently cited as key to the success of a data scientist in post. Perhaps version 2.0 of the Data Scientist is likely to be more rounded and ready to work alongside key data-enabled staff and management of the organisation in which they are employed?
But it’s not all one way here. What we found was that there was a need for businesses and other organisations to finesse or modify their approach to finding Data Scientist 2.0.
Of key importance here is that the organisation is able to provide great clarity of purpose. Not just a vague job description but signals and indicators for academics and data scientists alike to use to help them navigate towards a more effective curriculum and better equipped staff.
It’s also about understanding actual role descriptions. Data Science is a broad term – is this what you really need? Or do you actually need an Insight Analyst? Either way, building structure into your talent plan for any organisation should help you to understand what you actually need and how it might fit in with the rest of the workforce. Building a modern business and its associated workforce in a connected economy need real thought as to how the jigsaw must be pieced together.
Perhaps even more importantly here, is that it will help you to understand gaps which can be used to craft learning and development activities for data science staff when they join the organisation. Think about this not just as an opportunity to equip the staff with employability skills but as a mechanism to enhance organisational collaboration between those with deep data science skills and knowledge and those with business and domain expertise to deliver inclusive lifelong learning for all staff.
For the academic community and based upon our experience of placing 50 of the most suitably qualified individuals into leading organisations, we feel that Higher Education needs to continue to focus course development to help break down the data science silos and instead facilitate multidisciplinary strength.
Our view is that Scottish Universities perform particularly well here, but there is always more that can be done through the process of office and business simulation, business placements and internships such as those offered by the Data Lab and by encouraging greater collaboration between businesses and the academics running the courses.
For non-data science courses, it is key for universities and business schools to help champion robust data literacy for all students since all aspects of their future work and life are likely to be impacted by data science in some way, shape or form.
From our experience, some of the issues faced by organisations themselves getting the best from their data scientists boils down to their own data literacy… Perhaps schools and further education have a role here to play in enabling all students to become data literate and open to more non-traditional routes to data science – Once again, many in the world of academia could take a lead from the Scottish Universities as we felt that they, with the Data Lab, set clear intentions and outcomes for the MSc placement project that were a fantastic exemplar of what needs to become part of the fabric of higher education. In particular, building more commercial and professional ties with the business communities and strengthening alignment of the academic agenda to the unmet business need.
Robin Huggins who managed the programme for MBN, talks extensively on the diversity of the data scientists leaving higher education, but it is key that such diversity is still not quite where it needs to be. Higher Education needs to take a collaborative approach with industry to ensure that their plans for recruitment of both students and staff materially contributes to a diverse analytics workforce for the future.
We found that for many, this seemingly daunting activity of being placed with an organisation was compounded by the gap between academia and business. More evident to us as we work on the cusp of both, for candidate Data Scientists, language, timings, terminologies and techniques are different for those working and studying within Universities and those working in industry. With the right will and collaborative minds, this can be readily overcome. We found hundreds of people – students, academics, HR practitioners, technology professionals, data people within Industry and founders of businesses with the collaborative trait and this made the process much easier. But if you’re not so lucky, it could result in the square peg in a round hole syndrome.
How can employers find their own Data Scientist 2.0?
The starting point is to think about your requirements and how you will frame these. Don’t just think about academic qualifications, think about the soft or interpersonal skills you need. This is about how these are pieced together across the organisation’s entire talent management plan and identify skill levels required as ‘must have’ or ‘desirable’.
Think laterally, many companies experience great success with candidates with some exposure to the arts and humanities as it often teaches students to approach complex challenges using critical thinking and problem solving, creativity, communication, and collaboration. These creative thinkers have what management say they will want long after some of the coding and data science tasks we know today are automated through available tools, artificial intelligence and machine learning.
Be open minded to the social and cultural backgrounds of candidates. It’s all too easy to be unintentionally biased to recruiting the usual suspects who look, sound and act like the interviewer. Avoiding bias to gender, ethnicity, social background and possibly even the country of study will undoubtedly help find a wider gene pool of talented, more diverse candidates.
Hackathons, coding competitions and trial data model builds fall into the category of competition and these can be used effectively to identify the very best candidates. But don’t stop there. Why not hold situational analysis or critical thinking exercises as part of the competition? Use these to find those elusive, more rounded individuals with deep coding, data science and analytical skills with a thick veneer of soft skills and employment competencies to wrap their technical expertise.
How about working with an expert recruiter? Finding those rare recruiters who can sit across the board between business and academia is a useful starting point, but think on… Look for opportunities for the very best employers to ‘try before they buy’ with the very best candidates by building programmes such as that MBN built and operated for the Data Lab. Run an in-house version and find suitable talented people to join your team. The benefits are huge, as you get the opportunity to mentor and shape the individual’s approach to entering the labour marketplace. You also have the opportunity to collaborate and determine the extent to which you can build a learning and development wrap to enhance employability before making the critical employment decision. This is a classic, win-win as the candidate data scientist receives feedback and guidance on developing necessary commercial, interpersonal and soft skills and the business has the opportunity to find people where there is a complementary fit on a low risk model.
Finally here, what about meetups and data science events? These are regular sessions run across the country, designed to allow people to learn together and socialise to understand each other’s perspectives on what they have to offer and how there may be a strong and complementary fit.
I’m studying Data Science at University, how do I become and Data Scientist 2.0?
Well aside from the obvious attendance at meetups and data science events to ensure you are networking and socialising with the right kind of future employers, there is much that can be pursued here. MOOCs (massive open online courses) from the likes of Edx, FutureLearn and Coursera are a great source of soft skills and employment competency training. They are flexible and cheap/free and can be wrapped around other activities. Courses here allow for rapid acquisition of business and domain expertise and knowledge that will at least facilitate you holding your own.
Volunteering and internships also offer great opportunities for you to learn more and acquire employability skills, it’s also a great way to enhance your own network but real traction here comes by searching out internships and opportunities to spend time with future employers.
Finally here, one of the most innovative methods of showing it’s not just about your data science skills, is a recently experienced method of the candidate concerned positioning themselves as someone who could communicate in the same lexicon used by the business. A short thought leadership piece together with a schema for how predictive analytics may help with understanding some of the new country entry issues faced by the target business was enough to secure the candidate an interview with a key business team leader.
The fact that the thought leadership, once drafted, was in the same type of language used by the business itself helped illustrate the employability of the candidate concerned. Then, one assessment centre later, the Data Scientist concerned secured a role in the new ventures team at this particular FTSE 250 business without it ever being advertised!
The Need For Mentoring
Much of what we are describing here came about as a result of the mentoring provided to the students. MBN’s role was to figure out, determine any gaps and help build the bridges we so frequently talk about. The company helped candidates understand the specific unmet organisational needs of the target businesses. We also provided training on the soft skills and employability work done by the Universities to help the candidates to become sufficiently ready for their roles.
Facilitating such deep and long term relationships deliver tremendous benefits all round to clients and candidates alike.
Well, it was a great learning experience. It is clear to us that employers are seeking technical and domain expertise as traditionally expected. But – and it’s an important but – they want these in the context of the ability to apply the skills to real commercial and organisational issues. Backed by hands-on experience, evidenced through apprenticeships, internships and placements such as those delivered on this project. They also want stronger, more obvious collaboration and influence by the commercial sector on the academic agenda, and where the resultant candidate workforce has more diversity, inclusiveness and is non-exclusive.