Organizations Striving To Close The Data Science Skills Gap

Big data is undoubtedly one of the hottest trends of our age, and the promise of enormous amounts of data to fundamentally transform how our organizations operate is considerable.  For many however, the promise remains just that, with numerous barriers holding them back, whether it’s a lack of board level buy-in or poor quality data.

Arguably the most substantial drag on our efforts however has been a lack of skills.  It’s a situation that is likely to see companies aim to triple the size of their data science teams in the next few years.  That’s the finding of a recent paper from ESADE researchers.

The researchers examined over 100 Spanish companies from across a range of sectors, most of which had over €200 million in turnover.  The results revealed the long way we still have to go before data is at the heart of organizational behavior.

Slow progress

Despite big data being technologically feasible for several years, over half of the organizations revealed that they are yet to have a culture of data-based decision making, whilst 40% admitted that they don’t have a specific leadership role for data.

This reticence is important, as the study found that companies with a more analytical culture performed better than those without.  This was reflected in both their financial performance and the perception of staff at the companies.  Indeed, some 78% of companies who were regarded as very analytical thought that this culture had a significant impact upon their performance.

The study found that data professionals tended to fall into one of two categories:

  1. Data scientists, who tend to perform advanced analyses.
  2. Data managers, who provide the business vision to connect these analyses to the strategy of the business.

The typical data team would have between 5 and 20 members, but pretty much every organization reported finding it difficult to find the talent they needed.  Despite these recruitment challenges, the majority of organizations wanted to considerably increase the size of their data teams in the next three years, with three times as many data scientists and 2.5 times te number of data managers.

Train or recruit?

The desire for data science skills is clear, but this study suggests that most companies want to hire in external talent, or in other words the finished article.  This strategy would be fine except by all accounts, that talent isn’t currently existing in the marketplace, so there appears to be an inherent hope that external bodies will train people for them.

I’ve written previously about a similar issue when it comes to artificial intelligence skills, and data science and AI are so intertwined that the same surely applies.

Rather than attempting to hire in the finished article in an increasingly barren marketplace, companies are surely better off investing in data-science training and therefore upgrading their existing talent pool.  This approach has numerous advantages, not least of which is raising data skills across the board at a time when a growing number of organizations are attempting to democratize data science capabilities across the workforce rather than concentrate it within a data science function.

Organizations can achieve quick initial results by identifying employees with existing programming, analytical and quantitative skills and augmenting them with both the latest data-science skills and access to powerful tools, such as Python and Hadoop.

Spreading the availability of data education across the business, into marketing, finance, engineering and various other functions provides data literacy to people from various backgrounds.  This in turn will help to spread the data-driven culture that data advocates so crave.

A good example of this in practice is the Data University that Airbnb have created to provide anyone who wants to learn about data an opportunity to do so.  Already the company has trained over 500 (or 1/8th of the workforce) employees, with dividends already being reaped in the shift towards data-based decision making.

There has never been a better time to invest in the skills and talents of your workforce, with data promising to transform functions and processes throughout organizations that are already experimenting with a range of data science and machine learning initiatives.  Expertise is the principle barrier holding these back, so now really is the time to invest in the training that will bridge that gap.

Related

Facebooktwitterredditpinterestlinkedinmail