- Python Snacks
- Posts
- Understanding Biases Within Data Visualization: Cognitive Biases
Understanding Biases Within Data Visualization: Cognitive Biases
See how you can improve your plots by understanding visual and cognitive biases.
Cognitive biases are mental shortcuts or tendencies that can lead people to think or act in certain ways that are not always logical or rational. They affect how we perceive, interpret, and remember information.
Biases can lead to misinterpretations and potentially flawed business decisions. A great example of this is a camera company that we were familiar with at one point: Kodak.
The downfall of Kodak is a perfect example of a confirmation bias, where the leadership believed that traditional film photography would remain the dominant form of photography despite trends showing digital photography would become the dominant player in the space.
The company’s executives lacked vision and consistency, not adapting to the change in the market due to the mentality of “perfect products” (source).
» To put things into perspective, Kodak, at the time, controlled around 70% of the market.
So bridging into data visualization, cognitive biases such as confirmation bias can appear in charts and plots, leading to misinterpretation of the data, which could propagate upward to leadership.
Cognitive biases in data visualizations
It’s best to explain 3 different biases with an example. Suppose that you are presenting data on a new software product's performance metrics and its market potential.
Your goal is to inform senior management and guide their decision on whether to increase investment, pivot, or discontinue the product.
Each bias presented below contains a short explanation of what the bias is, an example with and without the bias, and how management can interpret both situations.
Confirmation Bias
Confirmation bias is the tendency to focus on information that confirms our existing beliefs and ignore information that contradicts them.
So, for example, if you believe your new software product is successful, you might only show data that highlights its positive performance while ignoring any negative trends.
Visualization of choice: A line graph showing only the first month of high usage.

Confirmation Bias - Top graph showing bias, bottom graph without bias.
Management interpretation with bias: Management may interpret the product as highly successful based on the initial high usage, leading to decisions to continue or increase investment.
Management interpretation without bias: The full trend is shown, including the decline. This leads management to make more informed decisions, potentially reallocating resources or adjusting the strategy.
What this graph fails to communicate: We may want to communicate the success of the software upon launch, but what we fail to communicate is the steady decline in users. Instead, we fixate on the initial peak.
Anchoring Bias
Anchoring bias is the tendency to rely too heavily on the first piece of information (the "anchor") when making decisions.
For example, if the initial month of user engagement on your software was abnormally high due to something such as a promotional campaign, you may use this as a benchmark, making subsequent months' performance seem worse by comparison.
Visualization of choice: A bar chart comparing initial launch month vs. average of following months.

Management interpretation with bias: Management might undervalue the product's steady performance after the initial peak, potentially deciding to discontinue or reduce investment prematurely.
Management interpretation without bias: Management can assess performance trends more accurately, recognizing a decline in usage. This balanced view can lead to a more informed decision.
What this graph fails to communicate: Similar to above, we miss the general trend of declining users within the software. In the biased graph, we average them out over 4 months compared to 1 month, making it seem like the software isn’t trending downwards in users as bad.
Framing Bias
Framing bias is the tendency to be influenced by how information is presented rather than the actual information itself.
For example, you want to show an increase in user engagement despite the software declining in its user base.
Visualization of choice: A bar chart showing an increase in engagement.

Framing bias - User Base and Engagement
Management interpretation with bias: Management might perceive significant improvement based on the increase in user engagement and decide to increase investment.
Management interpretation without bias: Management gets a clearer picture of the actual engagement level relative to the overall user base, leading to more informed decisions about resource allocation and strategy adjustments.
What this graph fails to communicate: While we are increasing in user engagement, this could be causation = correlation scenario, where the reason more users are engaged with the software is because of the way we’re calculating it: total users / active users * 100
. But it’s not clear.
📧 Join the Python Snacks Newsletter! 🐍
Want even more Python-related content that’s useful? Here’s 3 reasons why you should subscribe the Python Snacks newsletter:
Get Ahead in Python with bite-sized Python tips and tricks delivered straight to your inbox, like the one above.
Exclusive Subscriber Perks: Receive a curated selection of up to 6 high-impact Python resources, tips, and exclusive insights with each email.
Get Smarter with Python in under 5 minutes. Your next Python breakthrough could just an email away.
You can unsubscribe at any time.
Interested in starting a newsletter or a blog?
Do you have a wealth of knowledge and insights to share with the world? Starting your own newsletter or blog is an excellent way to establish yourself as an authority in your field, connect with a like-minded community, and open up new opportunities.
If TikTok, Twitter, Facebook, or other social media platforms were to get banned, you’d lose all your followers. This is why you should start a newsletter: you own your audience.
This article may contain affiliate links. Affiliate links come at no cost to you and support the costs of this blog. Should you purchase a product/service from an affiliate link, it will come at no additional cost to you.
Reply