What are some of the best practices for data visualization and storytelling?

Knowing your audience is probably the single most important component when presenting findings. Different audiences care about different insights and this should be taken into account when presenting. Data scientists and ML engineers might be more interested in model performance (AUC, R2, etc.) or clever feature engineering approaches. Business managers and product teams will probably care more about the ROI implications and revenue impact of improved ML models or exploratory data analyses. By knowing our audience we can better align the key takeaways to suit different needs and convey the right message.

To help me know my audience before my presentation, I often ask these 3 questions which are very helpful to help me understand what the audience needs and how I can help them by the end of the presentation: 1. Who am I communicating to? 2. What do I want my audience to know or do? 3. How can I use data to help make my point? Hope that's helpful to you too!

Here are some guidelines you can use to select a suitable chart for effective data visualization: 1. Bar Charts: Compare categories or show changes over time. 2. Line Charts: Display trends; compare multiple data series. 3. Pie Charts: Illustrate proportions; limit slices for clarity. 4. Scatter Plots: Reveal relationships between two variables. 5. Heatmaps: Visualize data matrices or correlations. 6. Histograms: Show the frequency distribution of continuous data. 7. Box Plots: Summarize distribution; compare categories. 8. TreeMap: Represent hierarchical data with nested rectangles. Keep in mind that these recommendations are only a starting point for your choice of the correct visual. You could easily write a book on this subject.

Top reasons why creating the right visuals to tell a story is essential for effectively communicating your message: 1) Enhance comprehension: Visuals can help simplify complex information and make it easier for the audience to understand the main points 2) Capture attention: Well-designed visuals can grab the audience's attention and draw them into the story, making it more engaging and memorable 3)Convey emotions: Visuals have the power to evoke emotions and create a deeper connection with the audience, enhancing the impact of the story 4) Support retention: People remember visual information better than text alone. By incorporating visuals, you increase the likelihood that your audience will remember the story and its key points

It is highly important to understand the the design practices while building dashboards or visualizations. Highlighting crucial information and not cramming a lot of information into one makes it easier for the audience to grasp the actionable information by reducing focus on the noise.

3.Law of Closure: When we see an incomplete or fragmented object, we tend to mentally fill in the missing information to perceive it as a complete object. This principle suggests that we have a tendency to perceive whole objects even when parts are missing.

6.Law of Symmetry: We tend to perceive symmetrical shapes and forms as a whole rather than individual parts. This principle suggests that we look for balance and harmony in visual stimuli.

5.Law of Figure-Ground: We naturally separate a visual scene into a foreground (figure) and a background (ground). The figure stands out and is perceived as the main object of focus, while the ground is the background against which the figure appears.

4. Law of Continuity: We tend to perceive continuous lines or patterns as belonging to the same object. This principle suggests that we prefer to see smooth and continuous paths rather than abrupt changes in direction.

I adhere to Matt Abrahams' three-part storytelling approach. 1.“What” - outlines the core message and sets expectations. 2.“So What” - delves into the relevance, urging the audience to care. 3.“Now What” - is the call to action, prompting engagement or decision-making. It also sets the stage for an elaboration or opportunity to take the story further in a specific direction. I meticulously tailor each story to resonate with its audience. The language, jargon, and context are adjusted to ensure the narrative is not just heard but felt, evoking a response. This strategy has proven effective across diverse audiences, from business leaders to students. Meet your audience where they are! P.S. Even the above article has 3 parts, review!

"Know your audience" helps you to select a story that resonates with your audience. Within the first 2 minutes of your presentation, your audience has to decide if they should pay attention to you, or look at their phones. This opening story must clearly identity the problem of interest to your audience. The details of your solution will be revealed...

I have found the McCandless method of data presentation extremely helpful for effective data visualizations. It is a five-step method. - Introduce the graphic by name - Answer obvious concerns before they're asked - Give your audience an insight into your graphic before diving in - Call out data to support this insight - Close and foreshadow the next point Following this step-by-step, or at least as a guideline, would set you up to effectively present your visualizations.