If you’re like most people, when you hear the term ‘consultant’ you think about hiring a subject matter expert to come into your company, execute on a project in a relatively small silo, make a set of recommendations, and move on. It’s generally time-consuming (and costly) to give a consultant access to your data and systems and get them onboarded. Even then they aren’t completely ready as there’s still a learning curve for them to understand what the business desires.

But what if you and your data team acted more like an internal consultant at your organization? This team would already be up to speed, have access to much more data on day one, and be able to apply business insights on top of their technical subject matter expertise. Even better, they could begin to define a standardized way of capturing and sharing insights within and across your lines of business. For example, you could build a data process, automate it, and then architect it to fit the solution that was being designed in a UI and hand it over to the business users once it was working. Anytime users or executive leads had questions the business line can’t answer, they’d know exactly where to find you.

Taking on a more consultative data role reflects my evolution in the last two organizations where I’ve worked. This represents a shift from responding to traditional calls to deliver more and more data to a mature posture where the data and analytics team is delivering key business insights based on that data. 

Layering a consultative mindset on top of technical and visualization skills can help any data leader to advance. You can deliver data all day, in other words, but that doesn’t mean your users will know what to focus on. I like to think of our team’s consulting role as the difference between dropping off a long printout of data and letting the user decide what to highlight (at which point the data is almost certainly out of date) and giving them fresh data where all the relevant points are automatically highlighted. 

Peeling back the onion

When you take on the insights orientation of a data consultant you begin to realize that by time a request has reached you, it probably has gone through several individuals from the top of the house on down and changed a little bit at each stop. Probing on these requests with methodologies like the Five Whys is a bit like peeling back the layers of an onion to get to the original ask. In my experience the primary request always has more layers and nuance, and allows you to build a scalable solution that answers many questions instead of just one. In our business, for example, these data tools are often aimed at understanding our clients better, and that can never be addressed by answering one monolithic question. 

Having a clear line of sight to the original request, as well as any relevant additions to it, is also helpful when considering KPIs. As any data analytics consultant looking into a new client will tell you, everyone’s KPIs are different because every department looks at things very differently, a difference often traceable to varying levels of data literacy. This lack of alignment often leads to KPI redundancy based on multiple definitions of the same metric. 

From pie charts to insights 

As any executive organization becomes more data literate, their appetite for actionable insights rises as well. They also want those insights at their fingertips. As a data team builds and designs functionality for its executives, trust starts to grow. Ultimately, if you’ve established your consultative credentials, they will let you build a solution however it makes sense because they trust what you’re going to give them. More importantly, they trust that what you’re going to build is intuitive enough for their use and will provide the insights they want. Any data leader should want to get to a place where their executives are saying, “Here’s my question on X” versus “I need a pie chart to do Y.” 

Another evolutionary shift to work toward is moving away from reports built by business titles in one silo fed with data from technical staff in another silo to one data-fluent SME or team building both the front and the back end of a solution. This hugely accelerates speed to market, which of course feeds competitive advantage. 

Thinking five steps ahead 

Where most organizations want to be is having the ability to think five or six steps ahead in their market. The big hurdle to getting there is that as humans we’re more accustomed to taking orders and having a list of requirements that we can’t operate without. That’s why it’s important for data leaders and their teams to do that question laddering, get to the Five Whys behind each request, and understand the direction where your users want to move. Otherwise you might be on the path to what one of my favorite books calls a zombie dashboard. 

You know that you’re generating the right insights for users when your solutions are constantly getting updated and enhanced. This tends to be a function of building something that answers more than one question. When you’re blindered by a long list of requirements, you can’t see anything at the margins of your perspective. 

I always advise taking a more consultative approach using a framework like MAD, which stands for Monitor Analyze Drill. In a solution built to MAD principles, a simplified UI allows senior-level users to monitor the situation at a 30,000-foot level, a set of more detailed views allows managers to Analyze where the numbers originated, while the most highly Drill (or Discovery) level is typically tables of transactional data for technical users and data teams.  

MAD is an invaluable tool because it helps you to think in terms of delivering insights to a broader audience versus a flood of data an audience of one. It’s this more consultative insights perspective that ensures you as a data leader continue to be relevant, and that your organization continues to thrive.