Building a data science project team

Media release

“There is no ‘i’ in team,” as the well-worn saying goes. History has proven that it does indeed take a team, rather than an individual, to achieve incredible results.

Assembling the right team for a particular task is all about the perfect blend of skills and capabilities. From five-time FIFA World Cup champion team Brazil, to Black Widow, Iron Man and The Hulk saving the universe, it was initially down to Brazil’s national team managers or Nick Fury to sort the right talents into the right combinations.

8 June 2020 Posted in DataIQ

That’s easier said than done and success depends on tackling a number of challenges. Typically, siloed teams in modern businesses mean managers are often unaware of what their existing teams are capable of, or what new skills are needed to plug the gaps. When it comes to data and analytics, building the winning team is all about aligning business objectives, the ability to demonstrate value and ROI back to the organisation, high performance at the task, and then retaining that team talent to fight another day, on another project.

Like football or cricket, data science is a team sport and depends on an understanding of the skills and capabilities of the whole team, plus a common language to bring a variety of skills and expertise together to understand and talk about what needs to be done. It’s this thread that helps to build a commonality of purpose and means that what is presented to the outside world is more powerful.

You are not going to recruit three magical people with all the expertise.

But how do you go about assessing what skills you already have internally and what skills you need to recruit/upskill for?

What’s important is that you determine what you need at your current phase of business and then build a team with the right data and analytic skill to deliver on your objectives. For example, if your business wants to save costs through automating some processes, that would require a specific set of programming skills, as well as statistics and business knowledge.

This does not mean you’re going to recruit three magical people with expertise in all those areas, but rather three different people, each of whom have a key strength in each of these things. There are software tools available that can help to identify who in the team may, for example, have the best technical skills to deliver projects and add business value, or the best communication skills to communicate complex technical details for non-technical audiences.

The key priority when taking this approach to building data science project teams is constantly to attract, develop and retain key skills and capabilities. Most of us have a strong desire continually to grow ourselves, keen to engage in professional development wherever opportunity presents. In some cases, this happens as a natural consequence of engaging in a wide variety of project work.

Six core capabilities define data science competencies.

But in others, it happens due to development planning aligned with our career framework. In these instances, using software solutions designed to align development needs with key data science traits can support teams in growing capabilities in the areas that are of most interest to them. Here are six core capabilities that we believe can help define the existing data science competencies, align strategic approaches to learning and development and resource projects to maximise value:

1. Communicators: Data doesn’t sell itself – it needs a communicator to guide the way. Because of this, many great data scientists are master communicators, able to lead key business decision-makers into an ongoing conversation about the right questions to answer with data and communicate analytic results in a meaningful way.

2. Data-wranglers: All great analysis starts with data, or rather starts with data in multiple locations, in different formats, languages and time zones. Data-wranglers understand that defining the question and the approach to creating insight stems from getting the data into a useable format.

3. Programmers: These cool, rational individuals are masters of multiple technical languages, excel at combining constructed analysis workflow, but also enjoy building applications from scratch. Scripting and programming are two very different, activities!

4. Technologists: Never satisfied with good enough, they find the best tool to aid with every challenge, constantly exploring how evolving tools and techniques can add value to the data science workflow. They use their technical knowledge to understand how best to deploy data science on scalable infrastructure.

5. Modellers: By creating quantitative descriptions of data, they create insight that is a key deliverable for the data science project team. Modellers are the ultimate investigator – when they’re on the team, if there is information that can be gleaned from a system, they will find it.

6. Visualisers: They are experts at converting information into a landscape that can be explored with the eyes to create an information map. This skillset is absolutely indispensable for organisations that are lost in information.

Having a thorough understanding of capabilities and skill level mapped against traits like these for the team can help guide and shape the data science project team best suited to the task. The result is a significantly more engaged workforce with a set of skills that the business understands and needs to deliver data-driven value.

Written by Rich Pugh