Forget graduates, only retraining can tackle skills crisis

British businesses cannot rely on graduates to combat the major data science skills shortage but must instead take urgent action to reskill and train up their own workforce to prevent the UK sleepwalking into a crisis.

That is the stark conclusion of a new study by SAS, following a survey of decision makers from major firms in the UK and Ireland, with an average of 27,000 employees.

According to the “How To Solve The Data Science Skills Shortage” report, 44% of firms plan to invest in AI technology, but 63% said their staff did not have the skills necessary and 61% did not have enough staff to deliver the benefits of AI.

Compounding this problem, 53% of respondents were unsure what AI qualifications and skills were needed.

Many firms said the pandemic had worsened the shortage of skilled staff, with just 8% stating staffing levels were better post pandemic.

Nearly 9 in 10 (88%) business leaders believe the staff shortage is preventing them from meeting customer demands, while three-quarters (75%) insists this is also hitting morale among the workforce.

According to a separate UK Government’s Data Skills Gap report, published last year, there were 215,000 vacancies for “hard data skills” that need to be filled.

However, the SAS study maintains that this is unlikely to be met by graduates, as there are only 10,000 potential graduates in “data science” each year.

In addition, the UK tops the list among countries in Europe for investment in AI firms and projects but the report warns that value from this significant investment will not be realised if there are not the skills available to deliver on it.

SAS senior director of education services EMEA Glyn Townsend reckons the report’s findings are a major concern for UK productivity and innovation and that businesses need to act now before things get worse.

He added: “Record employment, soaring inflation, the productivity conundrum and a permanent shift in consumer behaviour and expectations due to the pandemic, means demand for data science talent has never been higher.

“It will have wider repercussions for the economy, especially in the UK where there has been a tendency to outsource the need for analytics and AI skills to Asia Pacific and Latin America, compounding the problem in the global race for these skills.”

Townsend believes firms should consolidate diverse AI and analytics tools around modern, open, multi-language tools which will increase data science productivity and empower end users to do basic analytics tasks, allowing data scientists to focus on core tasks.

He added that businesses should grow the UK and Ireland data science talent pool by reskilling existing staff and creating more university data science graduates.

“Encourage a diverse range of certifications including training courses from software tools vendors. And create attractive data science employee networks, career structures and employment benefits.”

AI expert Dr Sally Eaves, who contributed to the report, said: “Businesses cannot rely solely on graduates or continue the poaching merry-go-round. The good news is employers have already begun to recognise the value of on-the-job training and other certifications as stated in the report.

“There is no single approach – but a combination of expanding mid-career training including to those currently in non-technology roles, equipping people with the right tools for the job and growing the data science community will start to see that skills gap narrow.

“Together, they could significantly increase the supply of talent, and create good quality satisfying jobs that benefit individuals, organisations and the wider economy.”

Dr Eaves also believes that there must be a broader focus on so-called STEAM [STEM plus the Arts] learning approaches which place an equal value on skills such as emotional intelligence, empathy, curiosity, problem solving and communication.

It is claimed these skills help give people enhanced agility to change both depth and diversity of experience and thinking.

Dr Eaves concluded: “Building data science capabilities doesn’t happen overnight but with the right learning pathways, and investment in modern analytics tools, it’s getting easier to upskill and reskill people, from both a tech and non-tech background. This can help build a pipeline of talent that’s going to be so vital to the UK and Ireland.”