How and why to develop as a Data Scientist

Continuing Professional Development (CPD) is a term that will already be familiar to many professionals. Your dentist, amongst many others, will be expected to show that she is keeping up with research and best practice.

Until recently, we heard little about this when placing data scientists or data engineers within businesses. However, that seems to be changing. Perhaps this is because of the increasing “professionalisation” of analytics. Either way, emphasising continual learning could be a very good thing for all involved.

Did you think it was just about studying hard at university to land that great first job? Are you hearing more about continuing to develop your skills and knowledge? The importance of this focus has been brought home to MBN during our conversations with 3 communities. In this post, I’ll share our perspective as a partner of all three.

Employers value Data Scientists who keep developing

One of the privileges of working on recruiting the right candidate for businesses is the opportunity to learn what they need. Our ongoing conversations, with a wide variety of data leaders, always give us insights.

Many people have spoken about the war for talent. With our strong links with universities and the wider analytics community, we also see this. With so many businesses waking up to their need of data scientists, demand can outstrip supply.

Many times, this means recruiting managers need to think outside the box. Their ideal wish list, of an experienced data scientist who is happy to be paid below the norm, may not be available. As we work with them, to help identify what is essential, learning is often key.

This chimes with the stories we hear from data scientists and their managers, years down the road. Those who settle well and who are most valued by their employer are those who keep on learning. It sounds like programming languages, data and business problems change. Those best able to adapt can cope with this reality.

For that reason, it is more important for data scientists (or data engineers) to be able to learn fast. Those who can quickly acquire new knowledge, or master new skills, thrive. CPD can really help them keep up-to-date with what may be needed next.

Only the other day, I saw a post on LinkedIn where a job applicant was complaining about all the programming languages he is expected to know. As he stated, he would need to spend many more years in university to graduate being fluent in all of them. Wiser employers realise it's best to look for candidates who can show their ability to learn new languages.

Data Scientists value workplaces where they can grow

Some employers are also experiencing high turnover of data scientists, data engineers or analysts. After all the effort put into finding a viable candidate and bringing them up-to-speed on your business, this can be painful. Leaders I've talked with sometimes feel like resourcing their teams is a full-time job in itself.

However, there are positive stories too. We do also hear from candidates who are still happy with their employer and managers who are managing to retain a growing team. What distinguishes them?

Both our experience and several surveys have stressed the importance of personal development. Realistic salary levels matter, but it has been true for years that analysts stay where they can grow. With an employer who invests in their CPD, providing access to learning resources and often the time required to complete this.

Such an investment also opens up the potential to grow your own talent. To help existing employees, with an aptitude, to gain the skills needed to embark on a data science career.

A recent trend seems to be the establishing of internal academies. These provide a structured programme covering topics relevant for data scientists. It is a testament to the breadth of data science roles, the number of subjects these academies can cover.

We have heard academies training in:

·        Programming languages (R, Python, etc)

·        Statistical packages (SAS and others)

·        Data (internal databases and metadata)

·        Presenting analysis (storytelling and presentation)

·        Data Visualisation

·        Data preparation (inc. feature engineering)

·        Customer Experience applications

·        Marketing applications

·        Pricing applications

·        Machine Learning and AI

·        Developing packages and APIs

·        Stakeholder Management

·        Report writing and style guides

·        Database and Blockchain technologies

·        Use of Big Data and Social Media sentiment

·        Use of Market Research and qualitative data

·        Statistical Significance and experimental design

·        Becoming a manager

·        Team work and Agile working

Phew, I'm tired just reading that list. It's interesting to see how it combines a mix of technical and softer skills. There also appears to be a need for general best practice and knowledge specific to that organisation. If you lead one of these teams, how well are you covering all those bases?

Don't worry, I've also been looking out for resources available.

Other Data Scientists are making use of the resources available

We've shared before on our collaboration with Data Lab Scotland. This includes both collaborating to put on events like DataFest and the placing of MSc students. This has been a great opportunity to see first-hand how powerful a partnership between universities and businesses can be.

In fact, the progress being made in Scotland is being recognised across the UK. Perhaps it can be a model for progress elsewhere.So, which resources is it worth you being aware of? First, let me caveat that I am no Data Science expert. I share the following list just on the basis of those I have heard others praise. I hope it helps as just a small contribution to encouraging CPD in this community.

Firstly, it is worth considering partnerships with universities. Here are some of the UK organisations focused on bridging the traditional gap between academia and commerce:

·        The Data Lab Scotland

·        The Alan Turing Institute

·        BBC Data Science Research Partnership

Beyond traditional universities, there has also been a rise in Massive Open Online Courses (MOOCs). Increasingly, we are hearing of candidates who have gained data science qualifications in this way.

Here are just some of the major providers of MOOCs, with Data Science courses:

·        EdX

·        Coursera

·        iTunes U

·        Udacity

·        Udemy

Finally, without completing a full qualification, considerable knowledge can be gained online. There are key global hubs, libraries of resources and active communities. It's worth having an up to date search, but a few I'm aware of include:

·        Data Science Central

·        GitHub

·        KD Nuggets

·        Kaggle

·        Data Camp

·        Dataversity

·        Data Kind

·        Data Science 101

·        EARL conferences (R)

·        PyCon conferences (Python)

How do you keep developing as a Data Scientist?

I hope these thoughts have helped you. At MBN we want to encourage the wider Data Science community to keep developing.

 We'd also love to hear your experience. Do you recognise this CPD challenge for Data Scientists? If so, please do share the resources that have helped you, or that you offer to help others.

 Let's all keep developing together, until Data Science truly becomes a profession.