I asked Maven Data Challenge winners for their top data viz advice. This is what they said.
Are you a new data analyst, looking to round out your portfolio? Or are you an experienced data analyst and need to brush up on your data viz skills? Either way, Maven Analytics Data Challenges are a great (and free!) opportunity to showcase your skills. In these monthly challenges, you work with real-life data to create a dashboard or report to explore trends and impact.
Each month, the Maven Analytics team chooses a challenge winner. I reached out to previous challenge winners to ask: “What are your top tips for data analysts entering the data challenges?”
Here’s what they said:
Gerard Duggan
Won 4 data challenges! Check out his portfolio for inspiration: Apple Carbon Emissions Progress Report, Telecoms Customer Churn Analysis, Unicorn Company Trends, Hotel Revenue Report
Gerard says:
- Establish your dashboard’s ultimate goal.
- Write a list of questions that you’ll set out to answer in your data visual. Check that you have suitable data to answer those questions.
- Consider your dashboard’s structure and develop initial wireframes. Keep in mind:
- Typically, people read left-to-right in a Z shape, and top to bottom.
- You may want to layer or create a hierarchy of information, starting with larger/macro numbers and drilling down into deeper analysis.
- Use white space, alignment, and other Gestalt principles to “guide” the end user through your dashboard.
4. Develop a theme/style.
- You may be provided with a style guide, or you may need to develop your own.
- Having a defined style helps you find appropriate colors, font styles, and sizes.
- Apply the color thoughtfully and consistently — less is usually more.
- Limit the font styles used.
- Use consistent sizes for titles, subtitles, text, and notes.
5. Use appropriate visualizations for the data being presented.
- Refer to the Financial Times Visual Vocabulary Guide for presenting data.
- Keep visuals as clean and easy to understand as possible. For example, most people won’t understand chord charts or Sankey diagrams. Try to use widely understood visuals unless there is a real need to go more complex.
6. Give your visuals sufficient labeling and context. This can be achieved through titles, axes, legends, and/or tooltips.
7. Always include units of measure! Otherwise, we won’t know if you’re displaying metrics about people, dollars, billions of dollars, bananas, or tons.
8. Do a quality control check: make sure that filters, slicers, and bookmarks all work correctly. If your dashboard shows incorrect numbers, your hard work will be in vain — end users won’t trust or use your dashboard. Never rush to publish without running quality checks!
Katie Welch
Won with “Electric Grid Outage Analysis”
Katie says:
- Define the dashboard’s scope and size based on your audience’s needs. Is a succinct one-page summary best for their needs, or a detailed multi-page report? What should the balance be between interactivity and narrative storytelling? Setting the scope early on helps you analyze and select content for your report.
- In the final report, focus only on the most relevant insights. Don’t overwhelm your audience with excessive data! It is tough to leave out parts of your analysis, but this gets easier with practice. For example, for the Power Outage Challenge, I showcased only the previous 10 years for most of my graphics, because that timeframe best demonstrated the key insights.
- Seek knowledge and inspiration from others. In this data challenge, I was guided by the words of Isaac Newton: “If I have seen further, it is by standing on the shoulders of giants.” Before designing my dashboard, I studied the dashboards of previous winners and finalists. This showed me what effective dashboards look like.
- Check out Maven Analytics’ YouTube channel. I especially recommend these videos:
- “Data Visualization and Storytelling Pro Tips”(Chris Dutton)
- “How to Build a Data Analyst Project Portfolio” (Enrique Ruiz)
Jasmin Jusufbegovic
2-time winner! Won the Maven Healthcare Challenge with “Pandemic Prescription: Are U.S. hospitals losing their recommended dose of trust?” and also won the Maven Space Challenge.
Jasmin says:
- Everything starts with the challenge prompt. Often, many participants overlook this step and don’t invest sufficient time in understanding the prompt, which is a fundamental part of preparing for the challenge. If it says you’re a data journalist reporting on Apple’s carbon emissions, then stick to that role. Avoid going off-track with something like a super detailed Power BI dashboard with 10 different filters/slicers that don’t match the brief.
- Learn graphic design and color theory. Also, before you even touch data analytics, spend a good three months getting the hang of graphic design and color theory. It makes a huge difference. You want your work to inform and captivate, not just bombard with colors and complex visuals.
- Explore data to discover a story. Maven Analytics challenges are diverse, ranging from creating infographics to conducting in-depth analytics. In analytical challenges, dig deeper than just presenting high-level metrics. Explore the data to uncover a story, continuously asking ‘why’ to understand the underlying reasons (e.g., investigating the drop in hospital satisfaction ratings during 2020 and linking it to the pandemic). Discovering the story makes it easier to ask relevant questions and effectively narrate the data.
Final words: The key to success in these challenges is a balance between understanding the data, the challenge’s requirements, and presenting it in an engaging, clear, and visually appealing manner.
Stefano Ciurlia
Stefano says:
- Read the challenge requirements and focus on its questions.
- Don’t put anything more than is needed in the dashboard. You may be tempted to showcase all the complex analyses you’ve performed, but for that, use the project page instead. In the report, only include what is needed to complete the task. In real life, reports are often streamlined into a one-pager, and everything else goes in the appendix (the equivalent of a project page here.)
- Beautify the report. A report should make stakeholders say “Wow!”, because it’s clean, visually pleasant, and summarizes the key concepts in an impactful way.
- Use consistent colors throughout your report.
- Use the challenge as an opportunity to try new things! This can mean practicing a new visual or a new coding language (e.g. HTML or Vega within Power BI, or creating customized visualizations using YouTube tutorials).
Luka Mihailo Jović
Won the Maven Family Leave Challenge
Luka says:
- Go for it! One can easily get intimidated by looking at the entries posted by more experienced data analysts. Take this as a learning opportunity.
- The devil is in the details, so read the challenge description carefully.
- More often than not, the challenges are based on real-world datasets, so you need to go beyond what is provided and explore a particular subject to put things in context. Without that, you’re stuck with showing stats that won’t tell much.
- Check out these sources to learn more about data visualizations:
- Book: “Storytelling with Data: A Data Visualization Guide for Business Professionals” by Cole Nussbaumer Knaflic (Amazon link here)
- Books by Edward Tufte
- Medium blog by data visualization expert Bill Shander
- Kerry Kolosko’s blog
Final Thoughts
Thank you so much to all the Maven Data Challenge winners who took the time to share their advice with me. If you’re planning to participate in the latest Maven Analytics data challenge, save this article for future reading!
Are there any data viz tips that stood out to you? Is there anything you would add? Comment below!
The contents of external submissions are not necessarily reflective of the opinions or work of Maven Analytics or any of its team members.
We believe in fostering lifelong learning and our intent is to provide a platform for the data community to share their work and seek feedback from the Maven Analytics data fam.
Happy learning!
-Team Maven
