Many organisations do have data leaders, and many more have plans to recruit one. This is great to see. The value a data leader can bring is something we’ll cover in a future blog, for the moment let’s focus on the recruitment itself, not just to get a data leader, but a great data leader.
Whether you’re a CEO, COO, CFO, recruiting manager, recruiter, or candidate please read on…
In the past two months I have had discussions with a handful of organisations who decided they wanted a data leader and asked for help. I was intrigued to understand their thinking. Why they wanted one, what sort of person they were looking for, and how they were going about the process?
SHAPING THE THINKING OF THE RECRUITING MANAGER
I discovered that the journey starts when someone has been given, or has asked for, the right to recruit a data leader. Whatever their background, as a COO, CEO, CFO or CIO, none of them were data experts so they were starting out a little bit ‘in the dark’. This creates an interesting situation. Should they ask for advice, who do they ask, and how good will that advice be? Indeed what does a great data leader look like? Their ideas of what they needed were based on the last data event or problem that happened. For example, they had a problem with a set of reports where the quality was poor, or couldn’t find anyone to own the stream of data issues they had.
Rightly or wrongly the event(s) shaped their thinking. What they really needed was a bit more context to the challenges they were facing and those that they would soon face. In fact what they really needed was a data strategy to ask the leader to deliver. Even if it was a simple strategy written by the business leadership as a starter for ten.
WHAT SORT OF LEADER DID THEY WANT?
The organisations were in different places in their data journey. Some had greenfield data sites, some had settled data teams, another an underperforming team, and one a high performing competent team.
What became obvious is that the type of leader you need depends on where your organisation is on its data journey. In my mind there are three different data leadership types:
Aspire and mature: Beat the door down. The key skills at this stage are those of a challenger. It is usual to inherit work half-finished or failed, and encounter disappointment and even resistance. The people who went before have tried, and failed, to fix the problems. So at this stage, our data leader must be prepared to evangelise and win friends, make a new business case from the ground up, create a coherent vision, and make hard decisions – break down silos and even kill off projects that can never succeed.
Industrialise: Consolidate. This wave is about creating repeatable gains, driving up quality, and scaling solutions. The leader still has to face challenges, but the key skills are attention to detail and process, integration into business processes, automation, and team building. The key achievement will be to create a stable culture of improvement that achieves the business goals promised at the outset.
Realise and differentiate: Innovate. With a toolkit of data products, with management’s attention and a track record of success, this requires a deep thinker to create innovative solutions to problems that will give the business an advantage over its competitors, change existing markets and create new ones. Innovators no longer need to convince executives to buy into the data transition, because it defines the strategic direction of the entire organisation. They do not need to break down silos or plead for investment. But as the size of the bets placed on transformative projects increases, so must the leader’s ability to think creatively and subtly to stay ahead of the curve.
Very few data leaders combine all three attributes. So it’s important to think about what sort of leader you need. You may want to bring in an interim leader to knock down a few doors, followed by a permanent person who can consolidate the data capabilities a year later.
MOTIVATION OF THE CANDIDATE
Some of these firms has been stung, and I’m sure the candidates felt stung too, by bringing in someone to do a completely different type of role. They soon left.
You need to consider the motivation of the candidate. If someone is a steady state leader don’t recruit them to knock down doors!
IT’S ALL IN THE NAME - “LEADER”
The candidates they were considering come from a variety of backgrounds – data science, business operations, data management, IT. They were unsure where was the best place to recruit from.
Honestly that doesn’t matter. What mattered was the common thing they were all looking for - a leader. Where they come from brings another set of skills as well as some potential bias in the way they manage. But leadership can bring so many capabilities so make sure that you bring someone in who is capable of the following:
Building a balanced team. As long as they recognise their strengths and weaknesses and build a team that complements them we’re off to a good start. This includes data governance, data science, requirements analysis, storytelling, data analysis, and data engineering.
Allow diversity of thinking. A data leader needs a combination of different skills for different roles. This means a combination of people who are task / project focussed, people focussed, code focussed, or solution focussed. How they get this comes down to diversity of thinking, diversity of culture, race, age and gender.
Delegation. The temptation for leaders is to hold on to power when they should delegate. So, to repeat a final piece of advice: ensure they understand that a data transformation is a team game.
Influence and inspire. Data is a relatively new discipline for most businesses so you need a leader who can influence those outside their control, and by inspiring people to follow them on the journey. You need to recruit a flag bearer.
If you find yourself recruiting a data leader with specific technical skills, such as Python, then the chances are you’re not looking for a data leader but a technical specialist.
THE PERSON WHO’S AVAILBLE INTERNALLY
The starting point for several of the discussions was “We have an internal candidate in mind and we think they’d be a good fit.” This was either driven by a person who’d put themselves forward and had few, if any, prerequisites skills, just wanted a change of role or driven by a manager looking after a member of their team.
Ok, this is an option, but it is just that. An option. The danger of filling a role with who you know without the clarity of what the role entails or without them having any of the prerequisite skills outlined above is high risk, yet commonplace.
Generally the best people are the busiest people not the ones that are available. Experience says that some of these leaders who get the experience from the role, don’t drive much value and go on to another firm before it all goes wrong. This is one of the drivers of high turnover of data leaders and leads firms to follow up by bringing in a hard hitting person to knock down doors.
Should you wish to bring in a candidate who is known and trusted within your organisation then you should consider helping them to be successful by getting them a data mentor or a coach. This has the benefit of developing a good employee into a new skill set but also de-risking the work by bringing in someone who can act as a trusted data advisor. It can also prove cost effective.
Businesses are built on success and so are data teams. Failure to recruit a great data leader will result in failure to realise the potential of the value of data, reduced investment and lack of opportunity for those in the data teams.
If you want to recruit a great data leader then I hope this helps your thinking and will help you develop your own ideas too. Please do share them in the comments. Here’s the summary:
It was rewarding for me having the opportunity to shape some roles but it also made me realise why the turnover of data leaders is so high today.
We recently published a related blog on why you need the right sort of data leader
Other DataTick blogs are available on a range of data and analytics subjects
If you need help with the issues raised in this blog or any other data & analytics advice please do contact us
This article is written by Simon Asplen-Taylor of DataTick. More detail is available in his recent book entitled "Data and Analytics Strategy for Business"
If you need help understanding how to get the most value out of your data, get in touch with our experts by email or give us a call. We’re happy to help!