Avoiding Confirmation Bias When Conducting Research


Browsing around your Facebook feed or doing a few searches on Google, you may notice that a lot of the posts and search results appear to be the same. The algorithms used by big tech companies are able to filter out the noise you don’t want to hear and really focus in on the topics and opinions that interest you. Not the worst thing, right? You know what you like and you’re happy to get more of it.

Maybe not.
When you’re only hearing one side of an argument or seeing endless new stories written on one angle, you can start to fall victim to a phenomenon called confirmation bias. If you’ve never heard the term, confirmation bias is a distortion of the truth that occurs when you over or underestimate certain information based on previously-held opinions. In short, if you see a whole lot of something you’re more likely to believe it as fact, whether it is or not. For example, those touting climate change as a fad often quote the scientific minority or even those outside of the scientific community. They also seem to somehow miss all the leading experts on the subject and discount any research pointing toward climate change.

Recently, I conducted some research about how research contributes to better decision making. As I expected, I was able to find several articles that confirmed this and I was ecstatic. However, soon I encountered an academic article with

Confirmation bias is something that affects all of us but until now, few people have considered what effect it may have on content marketing specifically. Are you thinking critically about your audience and what they’re looking for? What about the data you’re getting back – are you cherry picking data to fit a previously held belief about content success?

Writing in Forbes Magazine, content marketing mogul Jayson DeMers told us how this trend affects our efforts, and what we can do to overcome it.

Content targeting

When planning or writing content, you may make assumptions about the target audience that are wide of the mark. This can result in important information being ignored, or buyer behaviors assumed, leading to less relevant content.

Raw data analysis

If you already have an assumption for how well a marketing campaign should be doing, you’re likely to pick out figures from raw analytics data that back up this perspective rather than ones that provide a true overview.

Company branding

Taking a snapshot of your company from social media (including hyperbolas negative or positive comments that align with your opinion) isn’t likely to paint an accurate picture, resulting in you taking the wrong action.

DeMers recommends adopting a more scientific approach to overcome confirmation bias. In his article, he suggests treating everything as an experiment, wiping your mind clean of any prior hypotheses, seeking to disprove yourself and collating data/opinion from multiple sources.