How to Secure Data Science Talent in 2020
Uncategorized

How to Secure Data Science Talent in 2020?

March 1, 2020 By 1 Comment

How to Secure Data Science Talent in 2020?

The United Kingdom is fast turning out to be a data science hub.  But the question remains “How to Secure Data Science Talent in 2020″?

London is already home to the Alan Turing Institute — a prestigious national research centre for data science and artificial intelligence (AI). There are at least six government-backed data science accelerators[https://www.gov.uk/government/publications/data-science-accelerator-programme/introduction-to-the-data-science-accelerator] spread out across the U.K.

It goes without saying that data science, data analyst, and data specialist jobs have also been on the rise. According to BeSeen [https://www.beseen.com/blog/talent/5-most-in-demand-skills-for-uk-data-scientists/], data scientist job postings in the United Kingdom have increased by 114% since March 2016. And in today’s market, data scientists can expect to earn between £55,170 and £76,199, according to Indeed data.

For organisations, there are really only a few ways to secure data science talent:

Data science specialists are part problem-solvers, part programmers, part-scientists, and part-analysts. Special skill sets such as these are now in huge demand for organizations that rely on data science to solve problems, make key business decisions, to cut costs, or to stay ahead.  So How to Secure Data Science Talent in 2020?

  1. Create data science positions and then hire experienced data specialists or train existing analytics staff to fill up those positions.
  2. Spreading Data Science skills on the existing analytics team
  3. Outsource advanced analytics to specialist vendors.

How do you decide the best option for your business?

Consider various factors such as the costs involved, the time required to build or hire capability, whether you have the ability to train and deploy staff, the impact on project cycle time, and the sustainability of the approach to decide.

Let’s discuss the options in detail:

Upskill the entire team with data science skills 

The usual approach that most businesses take is to spread data science skills among existing teams believing that the whole team needs to be capable.

  • Technical specialists work on servers, software, the processes, and the workflows needed for gathering, cleaning, and integrating data.
  • Quantitative specialists are responsible for analysing and modelling data.
  • Business domain specialists use the data for making decisions, finding solutions to common business problems. They are ultimately responsible for defining the business problem, providing business context, and communicating findings persuasively.

Companies that take this approach of teamwide upskilling might start with general training for everyone and then narrow training efforts into silos.

Technical specialists might be further trained on additional (but relevant) technology streams such as distributed computing platforms.

Quantitative specialists might take up additional training on Machine Learning.

Finally, business domain specialists could get further trained in business management, etc.

Advantages: 

  • Limited impact on the company’s data science capability
  • Easier to build targeted skills than train a team member on all aspects of data science
  • Easier and more cost-effective to hire someone with expertise in one area, than hiring versatile data scientists
  • Low risk, since in-house talent is used

Disadvantages: 

  • No distinct specialization of any kind. Deep expertise capability is hard to achieve.
  • Unwillingness or inability of existing team members to learn new skills could be a potential issue.
  • Effectiveness of training on multiple skills is already a diluted, half-hearted approach to upskilling.
  • A lack of collaboration between team members can disrupt project timelines

Empower Specific Individuals with Data Science Skills 

Given the complexity and diversity of the skills required, data scientists are hard to find. Further, data professionals are rebranding themselves as data scientists, even though they do not have the requisite skills.

An alternative approach available to businesses looking to develop the skill sets of team members when it comes to data science.

Businesses can look for candidates with an expected baseline level of expertise and knowledge (in technical, quantitative, and business domain skills for instance). Companies can then seek to provide additional training and upskilling in specialisation areas such as Machine Learning, IoT, or others.

For this to happen, businesses need a robust hiring process to test candidates for data scientist roles through a combination of interviews and practical tests.

For data scientists, testing for applied skills is the key —not just theoretical knowledge—and a high degree of intellectual curiosity.

Practical tests involving tasks such as integrating disconnected datasets and analysing them to identify hidden patterns are key to testing applied skills and curiosity.

Advantages 

  • Reduced cycle time, since the same resource has skills across the workflow
  • Greater likelihood of discovering hidden patterns, since they work with raw data
  • Low risk, since the data science capability, remains in-house.

Disadvantages 

  • Negative impact on the company’s capability if a key data scientist leaves
  • High cost, since data scientists are in short supply
  • Significant time investment in finding data scientists externally
  • Difficulty in upskilling team members into versatile data scientists

Outsource Data Science 

Finding it hard to hire, train, and retain data science specialists? Another option available to businesses globally is to outsource advanced analytics to specialist vendors (i.e., consulting firms, pure-play analytics companies, and specialty consulting companies with expertise in Data Science, Big Data, Artificial Intelligence, etc.).

Vendor-built solutions often do not fit well with company resources and existing skillsets on the analytics team.

Using a vendor is only appropriate if a long-term, dedicated relationship exists to the extent that the vendor can adequately understand internal data sources and customize and develop solutions based on changing needs and continued feedback.

If you think that outsourcing data science capabilities and advanced analytics, the onus is on you to explain internal data. You’d have to provide business context to the vendors to ensure the accuracy of the solution. Finally, you must be able to understand, use, and benefit from the vendor developed model to ensure the sustainability of the data science efforts.

Advantages 

  • Ability to tap into instant teams with high-level expertise and deep knowledge.
  • Faster and quicker to hire the capability than build it in house
  • Easier to manage (since you deal with professional vendors)

Disadvantages 

  • There’s a higher risk involved since the capability does not exist in-house and companies must depend on other vendors or specialists to meet project demands and deadlines.
  • There must be an optimal match in interests for both the businesses and for the vendors since both parties are businesses in reality.

How are you going to seek, hire, and train talent for Data Science, Big Data, and more?

If you need help with an established and experienced vendor to help with your data science, data analytics, and problem-solving for your business, do get in touch with us today  [https://globaliotsolutions.co.uk/contact/].



1 Comment on "How to Secure Data Science Talent in 2020?"

  1. verbiage
    December 16, 2020 Reply

    Sаved as ɑ favorite, I lіke your website!


Would you like to share your thoughts?

Your email address will not be published. Required fields are marked *

Leave a Reply