Photo by fabio on Unsplash

Data Analysis: A Powerful Tool to Navigate Logical Fallacies in Strategic Decision-Making

Marko Joensuu
8 min readDec 14, 2023


Strategy-making involves an array of complex decisions affecting the future of an organization or a business. Whether comparing potential targets for cross-team collaboration or reviewing multiple solutions to a product issue, it’s crucial to avoid mistakes.

But it’s not always easy, as decision-making can be jeopardized by logical fallacies. These are erroneous beliefs or points of view that sound logical or plausible but, upon closer examination, reveal flaws or inconsistencies. Logical fallacies, such as confirmation bias, sunk cost fallacy, survivorship bias, and false dilemma, can lead to poor decisions and actions. This is where data analysis can be a huge help.

Confirmation Bias

Confirmation bias happens when decision-makers favor information that confirms their existing beliefs while ignoring contrary information. They sift through masses of data, choosing only the bits that support their assumptions. Consequently, the decision ultimately made might not be the best, most strategic one.

Data analysis can mitigate confirmation bias. It offers an objective review of the situation, waving aside preconceived notions and assumptions. By taking advantage of data visualization tools, such as pie charts, histograms, and scatter plots, decision-makers can see correlations, trends, and outliers. This information can help broaden perspectives and expose any potential biases. Furthermore, using algorithms or machine learning removes personal bias from data interpretation. Let’s dive deeper into confirmation bias with a few practical examples.

One example of confirmation bias in the business world could be seen in the product development stage. Suppose a company is committed to developing an innovative product they believe would disrupt the market. They conduct surveys and focus groups to gather feedback. Even if the data suggests that the market doesn’t need such a product or it might not be as successful as they believe, they may ignore this data and focus only on the positive feedback they’ve received. This can lead them to continue investing in a product that might not succeed in the market.

Another example can be seen in strategic decision-making. Let’s consider a company brainstorming ways to improve its sales performance. Management already believes that offering discounts and promotional deals is the best way forward, despite mixed responses to these tactics in the past. To validate this approach, they might overlook data showing that customers value improved service and quality over price reductions and focus only on instances where discounts have spurred sales. This could lead to a strategy that doesn’t fully maximize sales performance, as it’s based on confirmation bias.

Using data analysis can help in both these scenarios. In the first example, analyzing data could reveal what the market actually needs, helping the company to pivot or modify their product accordingly. In the second example, comprehensive data analysis might highlight other possible solutions to improve sales performance — solutions that might have been overlooked due to confirmation bias. From the data, they might discover that customers indeed value service and product quality over reductions in price.

Overcoming confirmation bias isn’t always easy, but it’s crucial for well-informed decision-making. Taking time to fully analyze and understand the data, rather than focusing on what confirms pre-existing beliefs, ensures a more accurate view of reality, resulting in better strategic decisions.

Sunk Cost Fallacy

The sunk cost fallacy is another common trap in strategic decision-making. It occurs when decision-makers are reluctant to abandon a project due to the invested resources, such as money, time, or effort, irrespective of the project’s future viability.

Data analysis can provide a clear picture of the return on investment (ROI). By comparing the sunk costs to the potential rewards, one can apply rational decision-making rather than sticking with an unprofitable project. Let’s deep dive into the sunk cost fallacy with some practical examples.

Businesses often fall prey to the sunk cost fallacy in the realm of project management. For instance, imagine a company that has been working on a software development project for two years. They’ve invested a significant amount of time, human resources, and capital into it. However, due to unforeseen challenges, the project is now over budget, delayed, and still far from completion. Despite a newer, more cost-effective solution being available, the company continues to pour resources into the failing project, driven by the sunk cost fallacy. They reason that they’ve invested too much to abandon it now, resulting in the prolongation of a clearly unprofitable venture.

Another example comes from the world of marketing. Suppose a company has invested heavily in a particular marketing campaign over the span of several months. Even though the campaign is not yielding the expected returns or increasing the brand’s market share, the company might continue to put money into it. This decision is based on the belief that sooner or later, the strategy must start to pay off due to the resources already sunk into it, rather than recognizing it as an unsuccessful endeavor and pivoting to a more effective approach.

In both these cases, data analysis could be vital in eliminating the sunk cost fallacy. Running predictive models could show the future profitability of the software project or marketing campaign based on current trends, helping businesses to make more informed decisions. By doing so, they could decide to cut their losses and invest in more viable projects or strategies, instead of being blinded by sunk costs.

Above all, it’s important to remember the principle “do not throw good money after bad.” This means that the costs already incurred should not affect the rational consideration of future costs and benefits. The ability to make such balanced decisions often differentiates between successful businesses and their less successful counterparts.

Survivorship Bias

Survivorship bias occurs when decision-makers focus only on the successes while overlooking the failures. It leads to an optimistic and often unrealistic perception of reality. This fallacy can be a pitfall in industries such as tech startups, where success stories may incorrectly be regarded as standard.

Data analysis can reveal the entire landscape, not just the success stories. By dissecting the data of both failures and successes, decision-makers can gain a more authentic understanding of the probabilities, risks, and potential ROI.

For a concrete instance of survivorship bias, consider the tech startup industry, where companies like Facebook, Amazon, Google, and Apple are often hailed as typical examples of success. Many entrepreneurs and investors may look at these companies and assume that all it takes is a good idea and some determination to achieve similar results. However, this overlooks the many tech startups that fail, which according to many statistics, is around 90%. These failures are frequently not given much attention or analysis, leading to an overly optimistic and unrealistic view of the industry.

An example of this might be found in an e-Commerce business trying to increase their conversion rates. Let’s say the business only looks at the data from successful sales, focusing on customers who made purchases and analyzing their behavior to shape their marketing and sales strategies. This would overlook a potentially large group of customers who added items to their cart but never completed the purchase.

An effective data analysis would also look at the behaviour of the customers who didn’t convert. Why did they abandon their shopping cart? At what point during the shopping process did they decide to leave? Were they repeat or first-time visitors? By considering these questions and analyzing all the data, not just the successes, the business can gain a more comprehensive understanding of their customer behaviour and make more informed decisions to possibly improve their overall conversion rate.

False Dilemma

The false dilemma fallacy involves viewing the situation as having only two options: it’s either A or B. This simplification infringes on the exploration of other solutions and possibilities.

A well-designed data analysis can help dispel this fallacy by uncovering other plausible options. This is done by collecting and analyzing a variety of data points to identify patterns, trends, and relations not initially obvious.
A false dilemma, or false dichotomy, occurs when only two choices are presented as the only options, while in reality, there are more options available. For example, in the world of Software Development, a manager might say to a team, “We either meet the deadline by working extra hours, or we lose the project.” This statement presents a false dilemma. There are other options that the team could consider, such as re-evaluating and reprioritizing the tasks, increasing the resources, or renegotiating the deadline with the client. The false dilemma can lead to unnecessary stress, demotivation, or burnout within the team, hindering their performance and productivity.

In the context of data analysis, a false dilemma might present itself in a situation where a business analyst says, “To increase profits, we either have to increase prices or cut costs.” However, a thorough data analysis might reveal alternatives. For example, analyzing customer’s buying patterns may suggest that offering bundled products could increase sales, or looking at seasonal trends might reveal that running promotions during certain times could increase income. Thus, comprehensive data analysis can clear up false dilemmas by providing more nuanced and realistic options.

Here’s What Else to Consider

Avoiding these fallacies through data analysis means leveraging a set of tools, techniques, and statistical models that enable you to convert raw data into insightful information. It’s about establishing KPIs, using predictive analysis, forecasting trends, testing hypotheses, and linking data analytics to your strategic decision-making process.

Let’s take the example of an online video streaming service. Using data analysis, they can establish KPIs such as viewer engagement, satisfaction rate, or average length of viewing sessions. Predictive analysis can be used to determine which type of content is likely to be more popular in the future, and A/B testing can determine which user interface or algorithm recommendation could bring more viewer satisfaction and engagement.

Composable architecture and product-led growth are approaches that can augment this process. For instance, an e-commerce company, using composable architecture, can experiment with implementing different components like changing their payment system or recommendation engine, based on what works best over time. They can scale and adapt their IT systems based on customer behavior and market demands without the risk of complete system failure.

Product-led growth, on the other hand, is best demonstrated by companies like Slack or Dropbox, who focus on optimizing their product as the main driver of customer acquisition, conversion, and expansion. These companies invest heavily in delivering a great user experience to drive usage and adoption. They align their teams around the product, use their product as a key distribution channel and constantly use data to improve and iterate the product experience.

Data analysis serves not only to validate but also fine-tune these strategies. By providing tangible evidence, it enables decision-makers to validate or contest their preconceptions, assess risks, rewards, and alternatives creatively and effectively, and make more informed, unbiased decisions.



Marko Joensuu

I'm a digital nomad and involded in fostering digital ventures and steering composable businesses towards success, creating unmatched value in the tech sphere.