The recent ‘2016 ANZ Teradata Index’ found when it comes to employing a data scientist to assist with analytics projects, 10% of respondents plan to hire one data scientist and 11% plan to hire an entire team of data scientists. 13% will use an external partner, while 6% will develop in-house skills. This demonstrates a strong understanding of the need to have data science skills available to the organisation, regardless of whether they are in-house or outsourced.
Australian organisations clearly understand the value of data but many are yet to fully realise its benefits and evolve to the point of becoming data centric. The sentient enterprise as an ideal state is likely to be some way into the future for most Australian organisations. These businesses need to focus first on implementing appropriate people and technology to fully leverage the data they collect. They also need to broaden their data source for a more complete view of the landscape in which their customers operate.
As organisations increasingly rely on the insights gleaned from big data to make critical business decisions, the role of the data scientist has become crucial.
An experienced data scientist or effective data science team can turn data into actionable insights, which can make the difference between overtaking competitors and lagging behind.
Experienced data scientists are a rare commodity and organisations should snap them up if they can find them. Creating a data science team is an important initiative for many organisations. However, it isn’t simply enough to employ a team of data scientists and leave them to it. To build a successful team, specific elements must be in place. You need to consider what you want your team to focus on, how you want them to perform, and how to get the most out of them.
The role of a data scientist is unique and can be valuable for all industries. As data becomes a core part of everyday business processes, having a data scientist will be a must in order to drive the future success of a company.
Data scientists can play a vital role in the business because of their ability to:
- Identify business problems
- Clarify how the business will need to solve problems
- Identify the right data to help solve a problem
- Communicate a problem and its solution back to the executive team.
Each of these tasks requires a different area of expertise. The perfect data scientist will possess well-honed skills in mathematics, statistics, programming, creativity and communication.
For example, it is quite common for someone to be a maths genius but not the best communicator. It can be quite difficult for an organisation to find someone with a combination of these talents, which makes them highly-valued and much sought after.
As data continues to evolve, it is vital that organisations understand how changing data can help improve business decisions and processes. The data scientist is the key to facilitating this.
5 steps to help organisations build a great data science team:
1. Stop hunting unicorns
Businesses are unlikely to find a single person with the requisite development, mathematics, statistics and business domain expertise. Instead, they should assemble a best-of-breed team and empower them to work together.
Some of the key types of people that should be included in the team are: data engineers; project managers; machine learning experts, and data modellers. If possible it is advisable to fill some of the roles with people already in the organisation, delivering some team members with existing business and domain knowledge.
2. If you build it …
As well as a team of smart people, organisations need a solid data infrastructure. Having the infrastructure in place while you assemble a team means they can get started right away.
3. Have a compass
Analysing data indiscriminately is ineffective and costly. The challenge to the data science team should not be to simply ‘find something interesting’. Their efforts must be aligned to business goals. It is vital that the business provides a question or hypothesis for investigation.
In addition, data scientists often have favourite tools and techniques. Make sure they are not so tied to a particular toolset or algorithm that they lack the flexibility to work on the organisation’s mission instead of their own research interests.
It is also critical to avoid getting caught up in problems that are interesting but have little bearing on the organisation’s main goals and priorities.
4. Have a timetable
Unlike pure research where publication is the benchmark for success, business demands iteration and delivery. It is therefore crucial to populate the data science team with people that know how to get things done, and can track project efforts and deliver at a steady pace.
5. Learn to spot success
When there is clear collaboration among business analysts, data engineers and data scientists, then the data science operation is on the right path.
Data science is a big field and a cross-functional team is better prepared than an individual to handle real-world challenges and goals. Hiring smart people who like learning and collaborating with others on interesting problems is the best way to create great data science teams.