How do you create a data-driven culture in your marketing team?

Developing skills needed for data-driven marketers who can understand and apply metrics and analytics

Becoming a data-driven organisation doesn’t just rely on the right technology, structure and processes. The human element is essential, and without the right skills, qualities and roles, any effort to be successful at data-driven marketing is destined to struggle.

The kinds of skills that support a data-driven philosophy are rich and varied

'The team needs to be technically savvy,' says Jim Sterne, Founder of the Digital Analytics Association.

"They do not need to be technologists, but they need to understand how the Internet functions at the bits-and-bytes level, so they can make the most of it at the human level. The team needs to understand how data is gathered and where it may have weaknesses so they can make informed decisions about which numbers to trust in which situations."

Sterne explains further that: 'The marketing department does not need to have a statistician on board, but there must be a solid understanding of basic of higher maths so they can ask intelligent questions of those who profess to understand such things. Ideally, the marketing team can write javascript tags and can troubleshooting network outages and can debug a regression test and can hold their own in a strategy session about branding, but that is unrealistic.

Skills that make for a great marketing team

The data-driven marketing team is knowledgeable enough to converse freely with technical and statistical resources while staying laser-focused on getting the right message to the right person at the right time. But the most important quality needed in a modern marketing team is curiosity. Without that, I may as well outsource all of my data related work to a third party. Curiosity stimulates creativity and conversation, and aids decision-making.

Other skills that are invaluable for any data-driven organisation include an understanding of the relevant legislation, such as the Data Protection Act. And some of the disciplines that are required may benefit from dedicated resources.

'Having strong analytical skills is a must - this is a quality that most team members need to share across multiple disciplines, and it’s a mindset that needs to be reinforced every day. However, when it comes to number crunching and deep data analysis, any marketing team will be better off with a full time resource dedicated to making sense of data,' suggests Sofia Quintero, Head of Growth at Geckoboard.

'A data scientist or business intelligence specialist can add tremendous value to the overall marketing activity. If you can provide your team members with the data they need in a way they can easily digest, you can free their time to focus on what they’re good at instead of chasing after numbers.'

Gerry Brown, Senior Analyst for Customer Engagement and Marketing Technology, at Ovum, adds: 'The ability to play with numbers, determine data outliers, data pattern norms and determine relationships between data points is important. But you must make sure individuals do not create data / analyse data ‘for the sake of it’. For this reason, you should align them overtly with the customer - e.g. ‘customer acquisition or retention data manager’, ‘customer insights manager’, ‘audiences development manager’ - and measure their performance and remunerate them based on customer-related metrics'.

Towards a Chief Data Officer?

One role that has become increasingly popular as the drive towards more data-orientated business has gathered momentum, is that of ‘Chief Data Officer’. Recent research by Experian Data Quality revealed that while the role is still in its infancy, there is a growing appetite for such a position, with 92% of those surveyed reporting that a Chief Data Officer would be best placed to define data strategy and be the guardian of data quality within an organisation. 61% of respondents also told Experian that they wanted a CDO to be hired within the next year to absorb the pressure of increased data volumes and drive the formation of a long-term data strategy.

Indeed, Gartner has forecast that as many as 25% of enterprises could have implemented a Chief Data Officer within the next 12 months.

John Mitchison, Head of Preference Services, Compliance and Legal at The DMA, notes: 'Data is a powerful thing but it can also put a company at risk'.

One of the chief recommendations that came out of Which? Taskforce on behalf of the government to look into the use of data and how consent was gathered, was that somebody should be held accountable at board level.

They do this kind of thing for health and safety at a company and it should be the same with data, because quite often people at the top of an organisation are quite removed from the people at the coal face. The targets and the pressure that they put on people to perform often encourages them to maybe push the limits a little and source data that they wouldn’t normally. And that’s when you start to get complaints. I think it would be very useful to have somebody that was fully responsible for that kind of thing'.

But Sterne has a word of warning about the longevity of the CDO’s role. 'A Chief Data Officer can be a huge help to an organisation - for a while. Perhaps ‘chief data-driven change management officer’ (CD-DCMO) is a more edifying title. This person is given broad authority to help a large organisation change the way it works, culturally. This takes education, tools, reengineering of business processes and even changes in personnel. It is a long and arduous task - and it might not be possible in the long run. There will either come a time when the CD-DCMO is no longer needed because the job is done or corporate culture is so ingrained as to be impenetrable. Either way, this role will no longer have any use.'

It's all about process...

Quintero adds that it isn’t the role that is important, so much as it is the processes that they support.

'I think about this less from a role perspective and more from a process perspective,' she explains. 'Businesses need a process that facilitates the access to data for every individual so that they can use that information to make better decisions. It’s data visibility that makes a difference, not job titles. Making data available across multiple teams can be done by any internal leader who wants to build a more data-driven culture and drive those efforts regardless of their business title. I believe that businesses can do a better job encouraging all levels in their organisation to be more data-driven instead of relying on a single individual to make it happen. This is not an easy process and requires cultural buy-in from a large number of people in order to make it work. A data-driven culture is the result of a company-wide effort, there are no shortcuts.'

Creating a Data-driven culture

So how else can you create and nurture a data-driven culture in your marketing team? The Experts we interview recommend these three practices

• 1. Encourage a mentality focus on hard financial metrics. Historically, many marketing departments have had a bias towards ‘soft metrics’, but these have led to a focus on frameworks that aren’t geared towards fact-based decisions. This should be replaced with an emphasis on facts and numbers.'Instil evidence-driven decision-making,; recommends Brown. 'Margaret Thatcher used to start her cabinet meeting discussions with the phrase 'what are the facts?' A similar questioning approach is required of modern marketers i.e. what and where is the data? How robust is it? How is it changing / what are the trends? What do we do to commercialise this opportunity? What ROI could it deliver over what timescale? Don’t just ‘do marketing’ but think about the data and what it means. Not a bad idea to put the slogan ‘Think’ on every desk like IBM did to such great effect in the 1940s and 1950s.'
• 2. Encourage accountability. Soft metrics aren’t conducive to accountability, but harder targets enable this. After all, if you’re going to set hard targets you should hold someone accountable for meeting them.
• 3. Celebrate small wins and highlight lessons. 'Celebrating small wins and highlighting lessons learned is the best way to get a marketing team to understand the importance of analytics,' says Sterne. 'When a test points to a higher lift result and that leads to a demonstrably successful change in marketing messaging or targeting, its party time! At the start, it may be necessary to research case studies to provide examples of what you want your team to achieve. Clear goals and objectives along with sufficient resources are the best tools for directing a team. Conversely, when a hypothesis is proven wrong, a constructive review of what was discovered during the process is even more valuable. It signals the team that making decisions based on data is always applauded and lessons learned are far more valuable than blind luck.'
• 4. Ensure there is transparency. 'Being able to communicate successes and failures freely can dramatically contribute to a culture that uses data to improve performance and make better decisions,' says Quintero.

Quintero adds: 'Building a data-driven culture is not an overnight process. It takes time. To me, a data-driven culture means building a safe environment where experimentation is encouraged and mistakes are tolerated. It’s less about having all the right tools in place – although that’s a critical part of the process – and definitely more about cultivating excitement around discovery and objectivity. Being data-driven is exciting and people should be encouraged to enjoy the process as much as making things happen.'

But of course instilling the culture is also a by-product of identifying and hiring the right staff in the first place. And this is no easy task.

'There is a lot of movement around in the industry at the moment - a lot of people changing jobs,' says Mitchison. 'A lot of companies put a freeze on recruitment a few years ago when things were quiet and now of course they want to recruit and there is only a limited number of people out there with the skills and experience.'

Finding the right talent

So how do you identify the best candidates?

'Do like Alan Turing did in the Imitation Game – give candidates a maths test and see how they perform. Then ask them what the numbers mean,' advises Brown. 'You need interpreters as well as statisticians, as Turing understood well. Stats graduates are an obvious first call, but being good at maths doesn’t mean you can think on your feet and determine what the numbers mean to the brand and / for the customer in terms of deliverables and value.'

Interview questions for candidates

Nick Cicero recommends preparing written and verbal questions to measure a candidate’s tech savvy and analytical IQ. He recommends asking candidates the following questions

• Can you explain the difference between metrics and analytics?
• What are some tools you’ve used for measurement and analysis?
• Describe a time where you had to use analytics to support a marketing decisions.

Sterne says: 'You identify talented data-driven marketers by the light in their eyes and the questions they ask. When given a data-related task, the truly data driven are inclined to ask, 'Why?' more than once. They want to understand the purpose of the report, the value of the dashboard and the sort of business decisions that are going to be made based on the results.

He adds: 'Hiring them is a combination of money and the promise of really cool problems to solve with really cool data, along with the assurance that they will have the tools and autonomy to make magic.'

And once you’ve got the right team in place, then of course there’s the matter of keeping them - something that not easy when those with the right skills are very much in demand.

'Retention is a matter of keeping them engaged and living up to those promises you made during the recruiting process,' notes Sterne. 'Like Sherlock Holmes suffering from a lack of cases, the data-driven marketer is desperate to test new methods and blend new data sources. The more you allow them to experiment, the longer they will stay in the laboratory.'

from Smart Insights http://ift.tt/1dXtlXq
via IFTTT