How to improve simple bar charts for a better story

How to improve simple bar charts for a better story

In this article I will discuss 3 methods you can use to tell a better story with bar chars. The data we will use in the graphs will be simple and relatable to everyday information you could be sharing. Here is the basic version of each with will enhance with our 3 methods.

In this article I will discuss 3 methods you can use to tell a better story with bar chars. The data we will use in the graphs will be simple and relatable to everyday information you could be sharing. Here is the basic version of each with will enhance with our 3 methods.

First, let’s identify three (mostly simple) methods you can use to beef up your story. 1) Sorting – With graphs this simple, there are 3 sorting options that will shape your story.


Alphabetical – This is almost never an option I choose. The only true benefit is the user does not have to work hard to search for a specific category name. When this is the main objective, it usually means there are a lot of categories and a tabular report is better way to deliver the information.


Descending/Ascending- If not dates are involved, I typically am choosing to sort the data descending or ascending by the measure value. This gives the user more information instantly. For example, what categories are first, last, and somewhere the middle.


Chronological– If you are trending a metric, this is the obvious choice. You will sort your dates ascending to see the change in your metric over time. Sometimes your data will happen in a business specific order. For example, you could have things bucketed stage 1, stage2, stage 3, stage 4 where it is important visually to keep the order of these stages as they happen in the business process.


2) Coloring – Using color can make the ingestion of information easier on your consumer. Done the wrong way, it can make the user work harder with no payoff. In general there are 3 obvious ways to color these graphs.


By Category Value or Date –While there are some situations to color by category value, I find it typically not the best option with graphs this simplistic. There needs to be a good business reason associated with the color to justify this move. For example, maybe your category is inherently a performance metric like meeting expectations and not meeting expectations. If you are simply coloring each bar a different color for the sake of having a legend of colors to look up the value, I’d say you probably have the wrong visual and should consider something else.


By Measure Value (Gradient) – This coloring scheme will give the same information as the height of the bars. This can be a great thing, so don’t be afraid of giving the info twice. People receive intelligence in different ways. Some people will appreciate how the color gradient allows easy identification of first and last places as well as tiebreakers on bars that are very similar in size.


By Performance Metric – This would be coloring your bars based on another metric that is related to the current metric. For example, coloring the bars based on being above or below the average instantly separates the bars into two groups in the consumers mind. Business thresholds like targets are another perfect implementation of this that executives really enjoy.


3) Transformation Additions – You can add something to the graph that gives more detail on a performance level. These take a little more skill depending on the platform you are creating graphs with, but knowing how to present these options in your graphs can really make you look like an expert! Here are some examples:


Trend Line – This is a line of best fit you would use with trending graphs. It tells the user instantly how performance is trending given the entire timeframe in view.


Percent Increase – Great for trending graphs. This info needs to be presented with care, as it can be very confusing if not positioned in the right way. There are tons of options like this, and they can get more and more complex but if you have the data chops to present them correctly, you can be a pro among your peers.


Performance Buckets – Business threshold expectations or average lines clearly shows the user what’s good, and what is not so good.

Examples of methods with trending graphs

Again, here is the baseline bar that we will start with. This graph does show the trend of sales and the bars can be compared to other bars close in proximity.

Color the bars by measure with gradient

When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars.Color the bars by measure with gradient


When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars.


Colors and Sorting: Colors are used to give the same information as the bar height. Sorted by month. Color the bars by measure with gradient


When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars.


Colors and Sorting: Colors are used to give the same information as the bar height. Sorted by month.


Pros: 1) Shows the trend over time of our metric
2) Colors make it easier to ingest comparisons between bars


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead.


Pros: 1) Shows the trend over time of our metric
2) Colors make it easier to ingest comparisons between bars


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead.

Above/Below Average (or other business threshold)

When to use it: Use this graph when you want to show a trend over time and separate performance into two groups “Above or Below Average”. Adding above or below average coloring to this graph gives the user a lot of additional insight with little effort on their part. You could substitute the average for the median or any arbitrary threshold that has business value.


Colors and Sorting: Colors are used to bucket time periods by performance. Only using two colors keeps the graph simple, yet informative letting the user decide if next steps or a deeper dive is needed. This is a trending graph, so sorting is done by time.


Pros: 1) Shows the trend over time of our metricColor the bars by measure with gradient


When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars.


Colors and Sorting: 1) Colors are used to give the same information as the bar height. Sorted by month.


Pros: 1) Shows the trend over time of our metric 2) Colors make it easier to ingest comparisons between bars


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead. 2) Buckets the time periods into groups by performance


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead. When not to use it:

Add a trendline

When to use it: Use this graph when you want to highlight the overall trend direction of a metric over time. If the trend line is tilted downward, overall the metric is headed down. If the trend line is tilted upward, overall theWhen not to use it: metric is headed up. This is very intuitive analysis for any user and even small directional trends can be detected.


Colors and Sorting: Sorted by month. No colors in this graph. Coloring could be added for additional performance details, but could get overwhelming.


Pros: 1) Shows the overall direction of the trend with the treColor the bars by measure with gradient 2) When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars. 3) Colors and Sorting: 4) Colors are used to give the same information as the bar height. Sorted by month.


Pros: 1) Shows the trend over time of our metric 2) Colors make it easier to ingest comparisons between bars


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead.nd line 2) Shows the trend between time periods of our metric


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead. 2) Large outliers – if your data has large outliers your trend line could be thrown off. Make sure the trend line makes sense visually. It should be relatively close to all data points. When not to use it:

Examples of methods with categorical graphs

Again, here is the baseline bar we will start with. This is sorted alphabetically. The only upside to this graph is that category values can be quickly looked up since they are in order. This graph is an example of providing data, not providing information. The user is able to compare the bars to other bars close in proximity, but the proximity is nothing more than alphabetical.Color the bars by measure with gradient


When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars.


Colors and Sorting: Colors are used to give the same information as the bar height. Sorted by month.


Pros: 1) Shows the trend over time of our metric
2) Colors make it easier to ingest comparisons between bars


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead.

Sort Bars Descending or Ascending By Metric

When to use it: Use this graph when you want to highlight the performance between the bars relative to the performance of the group. The user can easily see which bars are doing well, which bars are not doing well, and which bars are in the middle of the pack. Although the user must search for specific category values to compare them, the comparison is easy once they are found.


Colors and Sorting: Sorted by metric value either descending or ascending. Typically, you will position positive performers on the left, so if larger is better with your metric, sort descending, otherwise sort ascending. No color added to this visual.


Pros: 1) Shows the best, worst and middle performing bars
2) Visually compare performance between bars by position


When not to use it: Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead.

Sort Bars Descending or Ascending By Color the bars by measure with gradient

When to use it: Use this graph when you want to show a trend over time and also want to highlight similarities between time periods. The coloring gives the same information as the bar height, the higher the bar the darker the color. Providing the metric value in a second way is not wasteful. It makes it easier for the user to determine the top and bottom bars as well as similar bars.


Colors and Sorting: Colors are used to give the same information as the bar height. Sorted by month.


Pros: 1) Shows the trend over time of our metric 2) Colors make it easier to ingest comparisons between bars


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead.Metric and Color By Average


When to use it: Use this graph when you want to highlight the ranking performance of a bar when compared to the entire group and you want to separate the bars into two subgroups based on being above or below the average or other arbitrary threshold. The user can easily see which bars are doing well, which bars are not doing well, and which bars are in the middle of the pack. The user can also detect which bars are above or below the threshold given to color. Although the user must search for specific category values to compare them, the comparison is easy once they are found.


Colors and Sorting: Sorted by metric value either descending or ascending. Typically, you will position positive performers on the left, so if larger is better with your metric, sort descending, otherwise sort ascending. Color buckets the bars into two groups by performance. Average is commonly used, but a threshold like targets can work just as well.


Pros: 1) Shows the best, worst and middle performing bars 2) Buckets the bars into two subgroups by performance 3) Visually compare performance between bars by position


When not to use it: 1) Small differences – if your time periods have very small differences between them, it can be hard to see the trend with a bar graph. You will want to use a line graph instead. 2) Large Outliers – Don’t use average if there are large outliers.

Sort Bars Alphabetically and Color by Metric

When to use it: Use this graph when the ability to look up a category value quickly is most important. The coloring gives the same information as the bar height, the higher the bar the darker the color. This is not wasteful or redundant, it makes it even easier for the user to determine the top and bottom bars as well as similar bars. Coloring also helps the user determine approximately where the bar ranks among the group.


Colors and Sorting: Sorted alphabetically so values are easy to look up quickly. Colored by metric values so the coloring is giving the same information as the bar height. Typically, graphs that are needed to be sorted alphabetically have many category values, so this coloring gives additional insight into the approximate ranking of a bar when it is looked up.


Pros: 1) Category values are easy to look up 2) Approximate ranking can be deduced 3) Bar similarities bars stand out with color


When not to use it: When comparison between the bars is important or a more specific story needs to be told.

Sort Bars Alphabetically and Color by Median (Performance Bucket)

When to use it: Use this graph when the ability to look up a category value quickly is important, but you also need to highlight performance. Adding above or below average coloring to this graph gives the user a lot of additional insight with little effort on their part. You could substitute the average for the median or any arbitrary threshold that has business value. Providing this info with color is more difficult on the user with a line graph. In general, bar graphs will only be used for trending when we have significant differences between the time periods, as we do here.


Colors and Sorting: Sorted alphabetically so values are easy to look up quickly. Colors are used to bucket time periods by performance. Only using two colors keeps the graph simple, yet informative letting the user decide if next steps or a deeper dive is needed.


Pros: 1) Category values are easy to look up
2) Buckets category values into groups by performance


When not to use it: 1) When comparison between the bars is important or a more specific story needs to be told.

Our software creates dozens of pre-formatted graphs for you at once.

Try it free today!

Share This Post

Share on facebook
Share on linkedin
Share on twitter
Share on email

More To Explore

test_graph3

Trending Bar – Above or Below Average Graph objective: Show the trend over time and some extra insight into above or below average performance. When

test_graph2

Trending Bar – Above or Below Average Graph objective: Show the trend over time and some extra insight into above or below average performance. When

Get in touch.

I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Click edit button change this. I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Click edit button change this.

Get started today

©2020 Make Data Graphs