transform an overview slide

When you're tasked with putting together a presentation, it's often expected that you’ll kick things off with an overview slide. Think of it as a quick peek at what's coming up, giving everyone at the meeting the same expectations for what details you’ll eventually cover. But here's the catch: it's easy to get so caught up in building out the meat of your presentation, or to have a laundry list of other tasks that need attention, that the overview slide gets shortchanged.

If you leave that slide for the last minute, treat it as something you just have to get out of the way, or just update a few figures without really thinking about the bigger picture, you might start off on the wrong foot, with your audience scratching their heads over what you’re talking about and where you intended to go next. 

Even when you are pressed for time, your overview slide can be dramatically improved in a short period of time by simply (1) editing out unnecessary detail, (2) keeping your formatting consistent and organized, and (3) adding a line or two of text to summarize the overall message.

We recently worked with a client—we’ll call them MetroLink Urban Railways (MLUR) to protect their confidentiality—who created this overview slide as part of a larger presentation that was meant to show how MLUR was performing both against its internal targets and the overall region.

Before we leap into what we’d change, let’s give credit where credit is due for the things that are done well. Firstly, no distracting visual elements are competing for my attention. I’m sure you can imagine a slide with images or icons used to depict the metrics. (Showing delays with a red signal image comes immediately to mind.) It’s a relief these distractions aren’t present. I also appreciate the bold-facing that has been applied drawing my attention to the key metrics: train capacity; number of journeys; delay hours; and the forecasts. Our company, MLUR, is always clearly identified.

That being said, consider your experience when you first glanced at this slide. 

  • What information were you easily able to retrieve? 

  • Was it structured in a way that made sense to you? 

  • Were you confident you knew what the presentation was going to cover? 

  • Could you quickly identify what the key points were going to be?

Making a few quick changes allowed us to strengthen this slide for our client. Let’s go through the revision process together, and see how you might do the same thing in your own organization.

Edit out unnecessary detail

Ideally, you want the level of detail in your overview slide to hit that sweet spot between “too sparse” and “overwhelming.” As it stands, this slide tends more towards the latter. You likely need to spend a considerable amount of time understanding the level of the detail (overall market or MLUR specific), the type of variance (absolute or percentage), and the comparison being made (to previous years, or versus targets and thresholds).

How do you approach editing out some of this detail? There are so many metrics here that in the process of working on this slide, you may want to make yourself a data table, just to categorize what was being shown. Listing out your metrics in a table like this can provide some immediate clues about where to refine.

Out of the 17 different metrics mentioned in the slide, just 6 (highlighted in grey) are used for both the overall regional numbers and MLUR. What’s more, some of these metrics seem incidental—not important to the story at hand. Other, more important metrics, are partially (or completely) missing.

It’s a trap that’s easy to fall into, especially when rushed: we just include all of the data we have, without giving too much thought as to whether critical pieces are missing, or if some of what’s included is no more than distracting trivia. 

Let’s imagine that, together with the client, we revise what metrics are included so that we have a consistent and meaningful set of data. 

Once consistency across the metrics has been established, you can apply these measures to the slide.

Keep your formatting consistent and organized

Once we’ve settled on the information we’ll include in our overview, we move on to re-thinking its layout.  With the changes we’ve made so far, the slide feels less dense, but comparing MLUR to the region overall still requires considerable effort. For instance, how quickly can you determine whether MLUR is forecast to have more train capacity than the region overall?  This question still takes some searching to answer. 

As a general guideline, it is easiest to compare things that are physically close together. and consistently aligned with one another. Whatever comparisons we think should be easiest for our audience to make will inform how we lay out our information on the slide.

Currently, each metric is written out in text, as though it’s a sentence. This layout hampers our ability to easily scan the values of similar metrics. Instead, we can use a more grid-like structure, so that we have two vertical columns (one for overall region, one for MLUR), and then consistent horizontal rows for individual metrics and years. We use tab stops within individual lines, when multiple metrics appear, to make them easier to separate out from one another visually:

With this refresh, the slide is now much easier on the eyes, letting you scan through the info smoothly. Notice how much easier it is to see how things are shaping up not just in the area overall but also for MLUR, comparing their current performance against their goals for 2023 and even how they stack up to 2019. It's clear from this view that MLUR is on a roll, hitting all its targets and getting closer to the kind of success they saw before the pandemic hit. Unfortunately, the same can't be said for the region as a whole. 

By placing the forecast figures right beneath the current achievements, we've made it a breeze to visualize what MLUR is aiming for in the year ahead, and what they’ll need to do to get there.

Make the overall message clear

While we have a much cleaner slide than before, the volume of data still demands that our audience make some effort to process it all. We can make this task easier with the use of an observational takeaway, careful use of colour to highlight specific areas, and a clear call to action. At this stage, we can move MLUR to the more prominent position on the left-hand side, which suggests to a reader that it’s the MLUR data that is of most importance, and the regional data only meaningful as a comparison to MetroLink.

You’ll notice that since the consolidation of the metrics, no data has been removed from this view. Even so, a far more effective and action-provoking slide has been achieved via a clearer layout and careful use of words. Not all of our data is visualised with charts and graphs but even with an overview slide like this a positive impact can be seen. Check out the before and after below.

For more examples of visual transformations, check out the before-and-afters in our makeover gallery. Then, practice honing your data storytelling skills by undertaking an exercise in the SWD community.

from touchdowns to takeaways: a Super Bowl commercials makeover

in

More than 100 million people tuned in to watch the Kansas City Chiefs defeat the San Francisco 49ers in last weekend’s Super Bowl, which has evolved beyond a mere sporting match to something more like an unofficial American holiday. Historically, while many tune in for the football itself, a significant number of viewers are equally interested in the commercials. The team at SWD tracks along with both the football and the advertisements, but also a third aspect: any data visualizations associated with the game.

As a case in point, our friend and former colleague Elizabeth shared a graph with us that she discovered in an article published in the run-up to the game. We’ve recreated it below; it illustrates the most common Super Bowl commercials by industry over the last five years, providing a glimpse into a tradition that spans even further. 

The bold palette of this graph is certainly a scroll-stopper, which is a necessity when competing for attention in a week saturated with Super Bowl content. This rainbow color scheme does an excellent job of catching the eye.

However, as I scrutinize this chart through the lens of a data storyteller, I naturally start to think about the intent behind its creation. What are we meant to discern from this visualization?

A useful principle to remember is that the most straightforward comparison—that is to say, the easiest comparison to make visually—often reflects the creator's intended focus. In this instance, each year is represented by a different color in a stacked bar, and each commercial category has its own bar.  

Since the easiest thing to do here, visually, is to scan across the tops of each stack of bars and compare their heights against each other, my assumption is that the graph was created primarily so we could see which categories had the most advertisers over the past five Super Bowls.

The categories are sorted alphabetically (with Wellness & Insurance as the notable exception…if I had to guess, I’d suggest that the category was originally called Insurance & Wellness but was re-named at some point in the process without also being re-sorted). This arrangement is useful for long lists of categories, if the goal is to make it easy for someone to find a specific category quickly. However, if the aim is to highlight which categories are most and least prevalent, a more logical approach would be to sort them from most to least.

While the categories are now sorted meaningfully, the visual is still harder to interpret than it needs to be, owing to the diagonally placed category labels. Rotating this vertical bar chart to a horizontal bar orientation would allow the labels to be written in a single, easily readable line.

This adjustment makes the graph more navigable and the category names clearer, yet the vibrant color scheme still dominates the view. Now I have to wonder if the rainbow palette has gone past “attention grabbing” and into “overly distracting.”

Let’s think back to answering the question of the purpose of the visual. 

  • If the goal is to observe the fluctuation of commercials across categories over the five years, we could better achieve that by iterating to a different graph type. (Foreshadowing!)

  • On the other hand, if we’re meant simply to compare the overall category trends, toning down the color usage might be beneficial.

I may start by making every year an identical gray color…

…and then perhaps bring some color back in, just to highlight selected aspects of the data. Maybe we could color in only the year with the highest number of commercials in each category? 

Ugh, no. This results in a visually chaotic and demanding graph. If you can stand to look at it long enough, this view reveals some interesting trends, like the predominance of entertainment and alcohol commercials in 2023 and the beginnings of sports betting ads in 2022. More importantly, though, it underscores the need for a clearer visualization method for depicting changes over time.

Whenever I hear the phrase “over time” in my head in relation to data visualization, it’s a cue for me to at least try a line graph. For showing continuous data such as we have here, it seemed a promising alternative. 

Unfortunately, as is painfully obvious here, a standard line graph quickly proved unsuitable with this data set. It wound up looking like a "spaghetti graph"—an overly complex visualization with numerous overlapping data series. 

There are alternatives, however, when faced with this situation. Small multiple charts, which break down the data into individual series for easier comparison, offer a cleaner, more comprehensible format.

This visualization type allows us to create a grid of smaller, identically-scaled versions of the same axes for each data series independently. Scanning across all of them makes it easier to pick out variations in trends, peaks, valleys, and other anomalies. While they could be sorted alphabetically, here they are sorted from left to right, top to bottom, in order of total number of commercials across all five years of data.

Adopting small multiple charts for line graphs clarified the trends—we can see the dominance of Food & Beverage, the one-year peak of Entertainment in 2023, the rise and fall of Fintech—but makes it more challenging to see the volume of advertisements.

More so than line graphs, bar charts imply that the data they represent can be counted, or measured. Line graphs are excellent choices for showing rates, ratios, position on an arbitrary scale (like temperature)—anything where there’s not necessarily a meaningful relationship to zero. Bar charts, on the other hand, by their visual nature, imply that there is an amount of something. If one bar is twice as big as the next one, we think that there’s twice as much of whatever it is we’re graphing. Conversely, 100 degrees isn’t “twice as hot” as 50 degrees. 

The number of commercial advertisers in each category, in each year, is a countable, measurable value. If we use bar charts instead of line graphs, we can intentionally emphasize that aspect of our data. 

This bar chart version of a small multiple graph presents a more intuitive representation of volume, but now the trend is more challenging to see, since there’s now a "stair-step" effect to the change over time.

At this point, I’m torn. I appreciate aspects of both line graphs and bar charts here, but each one seems to be a little bit lacking. Ultimately, I found myself drawn to a graph I rarely reach for: the area graph. In most cases I find them too easy to misinterpret, and often overly complex. However, in a small multiple format, this approach effectively balances the visualization of trends over time with the representation of volume, fulfilling both objectives without overwhelming the viewer.

If tasked with sharing a visualization of this on social media, I would likely opt for the area graph small multiple chart. It maintains visual interest while facilitating more straightforward comparisons across categories over several years. Although the colors used do not carry inherent meaning, this compromise is often necessary when engaging a general audience, as opposed to the more focused use of color in business communications.

As we’ve iterated through various visual formats, we’ve also explored our data from a few different perspectives. This has yielded insightful observations, particularly regarding the dominance of the Food & Beverage category in Super Bowl commercials, with notable fluctuations over the years and a brief overtaking by the entertainment category in 2023. We also noted that the Fintech category also displayed interesting peaks in 2021 and 2022, particularly influenced by cryptocurrency and mortgage broker ads, which vanished by 2024.

In a business context, with a captive audience more interested in key takeaways, I would likely discard the small multiple view entirely. Instead, I’d employ a combination of line graphs with descriptive captions to convey these insights more clearly.

Ultimately, there is no singularly correct approach to data visualization. The key is to consider the audience's needs, the context of the presentation, and the intended message. Visualizing data is as much an art as it is a science, requiring experimentation, iteration, and feedback, rather than adherence to a strict set of rules.

Just as teams like the Kansas City Chiefs demonstrate through their repeated victories and strategic gameplay, excellence in data storytelling also requires continuous practice, experimentation, and adaptation. Each attempt at visualizing data—whether through bar graphs, line charts, or small multiples—serves as a learning opportunity, guiding us towards clearer, more impactful communication.