Setting up a high-performing analytics team that can address and manage diverse business expectations is not an easy task. At most times, when data shows any deviation from pre-conceived notions, in the business team’s mind, questions abound:
“My sales benchmark is below average?”
“Site visits cannot go down drastically – the data must be wrong…”
“It’s human error.”
All of our practitioners have heard this comment from time to time.
Some questions we often get from business stakeholders are:
- How do I get the momentum going for analytics in my organization?
- What kind of team do I need? How do I build this team?
- What about apps? Should I build one big app or several small ones?
- How do I deal with difficult stakeholders?
- How is your architecture set up? Because we have a data warehouse.
- Are we creating a platform as a service OR insights as a service?
- When should we progress from operational to descriptive to perspective to advance data science teams?
Essentially business stakeholders want to know how to build a high performing analytics team at every step of the organization’s data maturity journey.
The answer to this question would vary depending on the type of organization involved. However, the essential recipe for success remains the same for all.
Here are some key steps to building a high-performing Analytics team:
1. Building the team:
Four important personality types are ideally included in any analytics team –
The “Analytics Champion”, fearless analytics leader:
First things first, you’ll need someone to head your analytics organization. It is the Chief Data Officer’s or Chief Analytics Officer’s job to set an analytics vision for the business. This person will need to ensure that analytics is a well-respected function with a strategic voice and ongoing participation in execution in context of organizational objectives. So, make sure this person is ready to take on internal business leaders and external leaders if required before setting up the rest of the team.
The “Data Guys”, the data extraction extraordinaire:
These are Systems Architects, Data Engineers, and Data Architects who are the masters of data extraction, manipulation, and governance. They aren’t the builders of the apps themselves but instead deal with the logistics surrounding them, including source data governance, acquisition, management, aggregation, security and scalability. They will be the first layer of validation experts, ensuring data gets populated in the right cadence, frequency, and quality. This team is also responsible for streamlining server architecture for on-premise solutions or managing administrative capabilities for cloud-based BI solutions. They know how to establish a flawless data foundation and have a keen eye for creating the best possible development environment.
The BI team:
These team members can help with requirements gathering and project management for analytics projects, produce static and dynamic BI, third-party in-tool reporting and base level analysis. They are your foundation for establishing subject matter expertise on the team and your first tier of support for analytics requests. Value and nurture them as much as possible.
The APP builders:
Builders of reporting apps and BI solutions, they create easy to use applications for the BI and business teams. As the skillsets for data providers and the BI folks vary, the app builders help bridge this gap.
We miss this aspect more often than not. The storytellers are the folks who put the narratives behind the data. Start building this small team after the first year itself.
2. Plan for Investments upfront:
Building an analytics team requires payroll set up, technology and hardware investments, and initial seed money. All this requires investment and two approaches could work here:
Building a business recovery model:
As seen in our experience working with multiple customers, building on seed money by pitching to internal stakeholders (business stakeholders) for additional funds always helps. This model creates flexibility in scaling up programs and supporting team members when the time arises, provides enough cash to manage business downturns, and creates a buffer for rework on any analytics deliverable.
Going with a central committee for financial approvals:
This would be another way of looking at financing your teams’ investments but this approach normally takes time and throws up bureaucratic hurdles from time to time.
Organizations today have data amalgamation and reporting processes that typically run in silos. The larger the organization, the bigger the silos. And that is why the role of the internal analytics team becomes so important—after a few pilots run, their role is to champion the cause of analytics across teams from time to time. Eventually, once the team matures, they can define future roadmaps in collaboration with the business teams.
4. Skill set upgrade:
With the advent of new technology with increased ability to compute terabytes of data today, along with beautiful visuals created by tools like Tableau, PowerBI and DOMO, it is imperative to train your analysts and other functional folks in new technology and methodologies. This is even more essential in case of the advanced data science folks.
5. Setting expectations:
As business owners, we want data and analysis to be available yesterday. However, a lot of data processing needs to happen before we can see the dashboard. Hence it is imperative to set expectations on delivery of the insights and data among the business stakeholders, else it does not take long for small issues to snowball into larger problems.
6. Managing Culture:
Building an analytics team within the organizational context is a cultural change and such teams are usually the harbingers of change. As we onboard and start building internal teams, there will be reluctance from old timers and existing practitioners resisting change, so it is important to communicate the value that will be added by the new teams, and ensure ongoing communication across teams.
Talking about culture, building the Analytics team’s culture itself needs focus. Here are a few guidelines that have helped our clients in the past:
Provide a challenging and exciting environment:
Data analysts by nature seek out challenging and high-performing environments, and colleagues with whom they can bounce off and build on new ideas. They need to be able to engage with others in solving significant problems. The Analytics team champion should be able to assemble a dynamic group of people who are not afraid to disagree with or entertain the most inventive of solutions.
Invest in retention but do not splurge:
Continuity of the data analytics team is very important. It takes time to build up intellectual trust in this field and to understand the many unique ways colleagues might approach problems. If a team member leaves, sure you can hire a new analyst with skills, but it will be hard to quickly re-create the team knowledge-base and trust that has been built up. Serious investment in team retention is far less costly than turnover. At the same time, it is essential to create parity in the team and have minimal instances of team members comparing salaries to maintain an overall productive team environment.
Data analysts and data scientists like to get their hands dirty with data and usually work hard. Make sure they have some built-in downtime and flexibility on the job to pursue new ideas and research. Analysts crave time to invest in their own processes and productivity. Allowing this could pay off later in the form of on-the-job satisfaction, retention and innovative work that leads to breakthroughs.
Have analysts hang out together:
It’s a simple thing, but in today’s networked organizations, not always obvious. Team members should sit close to each other, and not be spread across a building or organization. There are many nuances to working with data and hence huddling together frequently is needed. Holding events like monthly breakfast meets or quarterly all-hands meetings are great to break the ice and get moving on program priorities.
Building the right team culture and getting upfront investments (seed money) are key and should be planned early in the journey. While team culture prepares the ground for success, seed money is akin to buying mortgage insurance—when we are unable to pay 20% upfront payment, pay 5% upfront using the seed money and then build it up from there.