Tips and Tricks
3 Ways Communication Can Help Analysts Be More Successful
There is nothing more satisfying than creating the perfect dashboard. You’ve found the data, created the right charts, your colors look great, and the dash is easy to use for you and business users alike. These dashboards are frequently created at the request of a someone else who needs the data to understand their own department and make the best decisions for the business. Marketing, finance and product look to the analytics team to give them the data they need to be successful.
Unfortunately, the dashboards are often only used a few times in reference to a specific situation, even though they could provide real lasting value. This isn't the fault of the analyst or the business partner, mind you. Shifting priorities and day-to-day needs may change what is being thought about over time.
Over the years, I have created dozens of dashboards I’ve loved, but not all of them have been adopted effectively. One thing I've learned is that being a good analyst is more than SQL, Python, or reporting—good analytics is a human exercise, a communications skill and way to foster a data-driven culture.
Here are three simple techniques that will help take your dashboards from simple intrigues to impactful business necessities, and they can be summarized as simply as "Make relevant content," "Let people know about it," and "Remind them why it’s valuable."
Make Relevant Content
Start with a business goal, get investment, and tell a story.
Before you begin writing a line of code, make sure your dashboard will be able to tell a story that is answering questions relevant to your business and the needs of the requestor. Every dashboard should be made with the business goal firmly in mind.
Once you and your requester are aligned around a common understanding of the objective, it will be much easier to create truly insightful reports. A focus on meeting the business need will also help generate buy-in from stakeholders, and keep you interested in the project. It is easy to make the requested chart, but context is king!
Dashboards should aim to be more than a collection of numbers, but rather an eloquent collection of insights and business answers. While KPIs are important for measuring business health, a great dashboard can help users make decisions by providing the context and structure to make that data actionable. When creating dashboards, "what" and "when" are easy to answer, but real value comes when you can also say "why." Some ideas to consider before you begin.
- Have a clear idea of what decisions your business partner is going to make before you build a dashboard. Understand what actions need to be taken and how the data can help make decisions effectively. Before you begin writing SQL, get your stakeholders to use a whiteboard to lay out their vision. It will help them and it will help you.
- Get everyone involved, invested and excited. Early buy-in, a sense of contribution, and continual updates will all helps your business partners feel some connection to the process—and ultimately will guarantee success down the road.
- Investigate what data you have, and use other numbers to provide context. For instance, when looking at leads generated by marketing channel, can you include channel spend on the same dashboard? This helps users can see the connections that drive success.
Always understand your dashboard’s and stakeholder’s goals, keep them aligned, and stay focused on keeping them front and center before you begin.
News organizations often can do a great job telling stories with data, and some are truly exceptional. Think about your favorite data blogs and platforms and try to emulate their successes and techniques. The New York Times does excellent data visualizations. FiveThirtyEight is a great example of data-driven reporting. Even browsing simple collections of good visualizations such as Reddit’s DataIsBeautiful or FlowingData can help spark creative design that will take your dashboard to the next level, from reporting to insight.
Let People Know About it
As a data analyst you are intimately familiar with your data, your creation process, and your chosen user interface when building a dashboard. This close familiarity creates a risk that you overlook information that could be helpful for casual users. Formally introducing your final product goes a long way to encourage adoption.
This should be more than a simple email or message on Slack. A proper introduction means setting up a meeting (and perhaps training) with key stakeholders and data consumers. If the dashboard was created as part of a reporting request, this means the original stakeholder and the analyst, but you should include the broader teams as well.
A good introductory meeting will seek to answer a range of important questions.
- What data is being used to create the report? How is it sourced, and how credible is it?
- How frequently is the data refreshed?
- Are there any exceptions or caveats that should be understood by business users when consuming the information? Is there a margin of error (and will discussing it be enlightening or confusing)?
- What questions does your dashboard seek to answer? What decisions can it help you make?
- Who else would get value from the dashboard?
A meeting covering these topics shouldn’t be overly technical, but rather strive to connect the dashboard to real value and instruct how to use it in the most effective way possible. Proper reporting is more than just creating dashboards, it is an opportunity to mentor and to drive a data culture at your company.
Remind Them Why it's Valuable
With modern business intelligence tools we have dramatically increased our access to information and our ability to create reports to visualize data. This is a blessing and a curse. Fast, actionable data means better business decisions, but it also means that the amount of data to look at can be overwhelming, and stakeholders are often focused on a single problem at a time.
This can quickly lead to dashboards being underused. Perhaps they are hard to find, or only need to be checked periodically, but even great dashboards can quickly fall to the side as the business moves forward.
If you have created and maintain a dashboard you know is critical to the business, it’s important to promote it to others who will find it valuable. No one individual can keep track of every datapoint, and leveling up the information you know to be valuable helps the business maintain a healthy culture of data-driven thinking.
- Set up meetings with stakeholders at regular intervals to cover the data and changes made to the dashboard. The frequency depends on the data; perhaps weekly for leads and acquisitions, and monthly for NPS or social media acquisitions.
- Highlight any changes made, new filters implemented, or new interactivity added. Take feedback on opportunities to improve the dashboard, and use meetings to highlight new work and keep the dashboard fresh in the stakeholder’s minds.
- Find connections and build bridges between data. Speak up when someone mentions a business challenge that you can help provide context for. If the value prop is strong enough, consider modifying your dashboard to combine the data.
- Make your dashboard easy to find! In the run of business, hundreds or thousands of dashboards can be created, but the easiest to find will be the ones that get used. Tag, star, or highlight valuable work in some way. Create a portal to your data that is easy to navigate, and provide instructions so new users can begin using your valuable work as quickly as possible. Consider a weekly digest or emailed report that will maintain interest in the dashboard and the valuable insights it provides. Reduce the effort needed to have business partners see your work as much as humanly possible.
If you prepare a story, introduce your dashboard, and continue to actively promote its value you will find your ability to drive business decisions to be far more powerful, your effectiveness as an analyst will be higher, and your business will grow much, much faster. Remember, analytics is about communication as much as engineering.
Good luck, and happy reporting!