Why don't your Data Stewards care?
The words “Data Steward” appear 11 times in the DMBOK, “Data Stewardship” appears 16 times, and “Data Stewards” appears 96 times, in 628 pages of content.
It would appear then, that Data Stewardship is well defined in the literature and we all know what to do and how to do it.
If that’s the case, why do you struggle so much to get the engagement of your business peers in the activities around data stewardship?
Most of the people I talk to tell me “I’m banging my head against a brick wall trying to get them to listen to me”…
What's in it for ME?
Motivating Data Stewards means we need Change Management. Change Management is the art of getting the organisation to change their behaviour. When we talk about changing Data Culture, this is usually the missing piece of the puzzle.
The DMBOK has a lot to say on this subject too, with “Change Management” appearing 95 times, and there being a whole chapter devoted to “Data Management and Organisational Change Management” (Chapter 17, page 573, for those that didn’t read that far before falling asleep).
Beyond DMBOK - motivating Data Stewards in the real-world
If we know WHAT to do, and we know HOW to do it, then why don’t our Data Stewards pick up the baton and get to work? Why is it so hard to get them to WANT TO do the work we ask them to do?
How can we actually get them to change behaviour?
He who pays the Piper calls the Tune
The problem in a lot of organisations is the Data Steward is NOT a full time resource, dedicated to Data Management activities.
Instead, it’s common to “borrow” a resource from a department, and we usually want to borrow the Subject Matter Expert.
Unfortunately, so does everyone else.
Nobody wants to talk to the new hire, or the guy that doesn’t really know what he’s doing. We want to go to the expert in the room. But their Manager doesn’t want precious time being wasted on meetings about Data, they want to get the outcome the SME was paid to do in the first place.
What do the Data Stewards want?
If the Manager of the Data Steward is not bought into the change, the Data Steward themselves will be very unlikely to buy into the change, too.
This is where our good friend Mr Maslow comes in.
The basic/physiological level of needs for our Data Steward are met by the Safety needs – employment and resources being important here.
Like every other worker, our first task when showing up to work is to keep our job – make sure you do what you’re paid to do so your company will pay your salary.
Some of us are content at that level, others see their work as more important – transcending the task at hand and becoming “the most that one can be“.
The art, then, is to identify the different context(s) your Data Stewards are in and to design their role around what’s most important to them.
How does Maslow relate to Data Stewardship?
Unless you’ve tied the KPIs of Data Stewardship to the benefit and job description of the Data Steward, their first order of business – keeping my job – is not going to be achieved by populating a Data Glossary or investigating a Data Quality issue.
If the Data Steward is struggling to hit their existing targets, or their workload is high because of another important project, they will drop your work like a hot potato.
To fix that, make the work relevant to their KPIs and their outcomes.
Pay the Piper
Typically there’s a Senior Data Steward or “Data Owner” that you are borrowing the Data Steward from in the first place. As noted above, “he who pays the Piper calls the Tune”.
Whilst your Data Strategy should have a clear link to business benefit, often these are too “high level” to matter to the individual Data Owner or line manager of the Data Steward.
How we fix this
When rolling out any Data Strategy and Governance initiative, the benefit cannot solely be related to the C-Suite. Unless you’re able to connect the dots between the value of the Governance work to the Data Owner and Data Steward, you’re never going to motivate them to WANT TO DO THE WORK.
You need to understand what’s going on in THEIR CONTEXT when you ask them to become a Data Steward, as well as when you ask them to perform Stewardship tasks and activities.
Here’s the bad news – getting this right is hard, it requires effort, and it requires learning new skills that are often foreign to data and technology professionals. To do it well, you need the following skills:
- Emotional Empathy – the ability to see the world through the eyes of your stakeholders, and to truly understand their perspective
- Uncovering Demand – where is there demand for better data within the team that are being asked to perform the work? If you scratch their back, they’ll scratch yours
- System Thinking – don’t just look at the work you have asked the Data Stewards to perform, understand the context around their role and the other pushes and pulls that impact their job and role
- Experimental Mindset – understand that you don’t know what works best. Create different support and incentive structures for different stewardship teams and iterate to learn what works best for your business
- Balance Priorities – you cannot have it all. We need to manage expectations and tradeoffs to get the right balance based on the time, money and capabilities you have to invest
Putting the Capabilities into Practice
What do we do with these skills? First, we investigate the experience of your existing Data Stewardship team. Typically we want to interview between 8 and 12 Data Stewards or Owners, aiming to talk to a broad audience, depending on your unique business context:
- Those that are reluctant to perform the work vs those that enthusiastically embrace their role
- Stewards with high capabilities vs those with lower skills
- People from high regulation markets vs those in small markets
- Those in growing business units vs those in stagnant markets
- Those with supportive Data Owners vs those with unsupportive Data Owners
To do this you need to evaluate your existing data stewardship landscape and identify the right traits for your business. The idea is to interview a broad range of Data Stewardship contexts in order to gain a 360 degree perspective of what motivates your Stewards and what holds them back.
Understanding the Stewardship Job
And lets be clear – not the job you’ve asked them to do, but the Job (capital “J”) that they have to be done in a data context. You’ll find there are four forces that impact motivation:
- Push: in their context, what happens that “pushes” the Data Steward to want to perform the role.
- Typically this is framed with a “When I…” statement, such as “When I have been manually fixing bad data for my team for years and have learned that data governance can help us solve this properly”
- Pull: more commonly “what’s in it for me?” – what will the Data Steward gain from the work?
- Typically this is framed with a “So I can…” statement, such as “So I can learn new skills in a role with growing demand”
- Anxieties: everyone has some degree of anxiety when trying something new, these are the questions people ask themselves in their mind that prevent progress
- These are usually posed as questions, such as “Can I really learn complex data skills at my age?” or “Will this mean I have to work overtime?“
- Habits: what are the habits of their current day job that prevent them taking on new work?
- This could be “I drop everything to focus on my boss’ next pet project whenever she calls me“
Unpacking the Past
To unpack these four forces, your interview process has to ask about the past:
- When did they first learn about data stewardship, or have the first thought that they would be involved in this?
- What was happening in the company and their role at that time?
- How did they react when brought into the stewardship team? Were they excited or anxious?
- What was their experience like when performing the role? How well prepared did they feel, and what held them back?
- When did they avoid Stewardship tasks and activities, and what took precedence over the Stewardship work?
- Would they want to take up a Stewardship role in the future, or would they prefer to be left out of the data team?
By understanding the past feelings and experiences we learn about the forces that impact progress and either propel the team forward to work on your tasks or hold them back.
Building the Jobs
The 8-12 interviews you’ve run should identify a range of different pushes and pulls, anxieties and habits that were impacting your team in different ways. Document these for each interview, classified as Push (past event pushing you toward change), Pull (future benefit luring you to change), Anxiety (expected challenge or negative energy around performing the task) and Habit (existing behaviour or approach that prevents change).
Now it’s time to compare the different interviews, to begin the work of coding and analysing what we have discovered.
Making Qualitative Data Quantitative
The forces of progress you have identified to this point are all qualitative data. This means they are interpretation based, descriptive and based on language. To use this data we need to quantify it:
- Cluster together the Push/Pull statements that are similar in meaning
- e.g. the push “When I’m fed up with my existing role” from one interview could also be clustered with “When I feel demotivated in my job“
- Once you’ve clustered multiple statements, take a step back and give them a group label – e.g. “When I no longer see a future in my current role“
- Go back to each interview and ask yourself “Was this Push/Pull label true for this interviewee?” – the goal here is to tag each interviewee with a one or a zero depending on whether they experienced this Push or Pull
- Once you’ve tagged all the interviews, it’s time to leverage mathematics – use clustering to identify groups of interviewees that all shared similar Pushes and Pulls (you may find 3, 4 or 5 different clusters, depending on the number of interviews)
- Review the clusters – what did those interviewees share in common? Do the groups make sense based on your interviews?
- This is typically where you learn the most, as different people that you initially wouldn’t put together may appear as having the same “Job”
- Once you are happy, create a generic Job statement for each cluster:
- e.g. “When I’ve been trying to fix bad data for years and my firm launches a Data Governance program, help me learn mastery of the skills I need so I can become a data expert that is respected and admired” or;
- “When I’m told I need to be a Data Steward, help me quickly and effectively complete my stewardship obligations so I can get back to my day-job“
- This is where the Anxieties and Habits come in – you can now review those elements for all interviewees that share a common Job, to understand what holds each group back, what tradeoffs they are willing to make and the functional, social and emotional motivations that drive or prevent progress
What do I do with all this information?
Here’s the fun part – now you can identify the Job your specific Data Stewards have, and what they themselves see as progress, you can design different KPIs, support structures, and training plans to help each group make progress:
- Someone with the Job “When I’ve been trying to fix bad data for years and my firm launches a Data Governance program, help me learn mastery of the skills I need so I can become a data expert that is respected and admired” could benefit from data management training in order to improve their skills and make them feel like a valued data expert, whereas;
- Someone with the Job “When I’m told I need to be a Data Steward, help me quickly and effectively complete my stewardship obligations so I can get back to my day-job” may not want to engage in Data Training beyond the absolute minimum, and may need support from the group Data Management team to handle less critical tasks – only bringing them into meetings when it’s absolutely necessary
Yes, this creates a multi-tier Stewardship program. You are treating some Stewards differently to others. But far from creating division, this brings the team together.
You’re tailoring the role, support, training and tasks to the Job your Stewards have to get done. You’re helping each of them make progress whilst they help you make your own progress.
Most importantly, you avoid creating a “One Size fits No-One” Data Stewardship position where you have ineffective data teams making progress or holding each other back, which leads to more dissatisfaction, disengagement and difficulty for your Data Management program.
If you want to put this into practice, you’re going to need to develop your skills and learn how to effectively interview people. I’d suggest reading the following books:
- Competing against Luck, by Clayton Christensen – to understand more about the Jobs framework
- Never Split the Difference, by Chris Voss – to learn how to develop “Tactical Empathy” and improve your listening and interviewing skills
- Learning to Build, by Bob Moesta – to learn how to combine these skills and create innovative solutions to your Data Stewardship problem
If you’re too busy to study this inside out and learn these skills yourself, we can help. Get in touch using the form below and we’ll arrange time to share more: