To Thine Own Data Be True

Marketers constantly look for better ways to use data—and we should. You could find, buried deep in a spreadsheet, data so clear that it suddenly solves your problem at hand.

But data, in and of itself, isn’t enough. You also have to interpret and communicate it logically.
That’s where the art of understanding logical fallacies comes in. Understanding logical fallacies won’t guarantee data success, but it sure helps.

Flawed Reasoning

Simply put, a logical fallacy is flawed reasoning on the path to an idea.
If you’re presenting information or a strategy whose data analysis is based on one or more logical fallacy, you’ll distort otherwise useful findings. Decisions on what appear to be rational analyses may actually be a step in the wrong direction. Incorporating a fallacy can lead to ideas based on unstable foundations, and that can cost you and your client dearly.
As a presentation participant, stay alert to logical fallacies. Be a healthy skeptic. Defend yourself from bad ideas based on bad evidence.
Scientific Theory Chart

Three Logical Fallacies

These three fallacies aren’t necessarily the worst or most common ones. But they’re the ones most closely tied with data representation.
Definitions are quoted directly from your logical fallacy is.1 Do yourself a favor and read up.

Texas Sharpshooter2

“This ‘false cause’ fallacy is coined after a marksman shooting randomly at barns and then painting bullseye targets around the spot where the most bullet holes appear, making it appear as if he’s a really good shot.”
This is essentially the act of establishing what data is relevant only considering what you need to support your hypothesis.
Target Board
For example: Say that your marketing agency is conducting an end-of-project report. Their analyst explains how great their digital efforts are performing. Sales are up 200% in the top three cities that saw their ads online.
Logical fallacy: The digital efforts must have made users in those cities behave in a unique way.
Why this is false: Their analysts forgot to consider that the entire company—and industry—is experiencing sales lifts. Other cities that never saw the ad performed even better, in fact. That’s correlation, not causation. Maybe your market agency’s display ad didn’t cause this.
Solution: Gather enough data to ensure that you aren’t confusing a random data cluster for something important.

Black or White3

“Also known as the ‘false dilemma,’ this insidious tactic has the appearance of forming a logical argument, but under closer scrutiny, it becomes evident that there are more possibilities than the either/or choice that is presented. Binary, black-or-white thinking doesn’t allow for the many different variables, conditions, and contexts in which there would exist more than just the two possibilities put forth.”
This often occurs when you’re attempting to move on from a looming decision by clarifying the available alternatives. A black or white logical fallacy takes you further away from the gray alternative that might be the best option or explanation.
Black and White Checker Board
For example: During a meeting to clarify your client’s Q3 marketing strategy, your client reiterates its two major revenue channels: online sales and phone sales.
Logical fallacy: Someone attempts to clarify that “people either shop online, or they call us. Our numbers are showing how much phone sales are increasing. We should move the budget away from our website and invest into more phone sales staff.”
Why this is false: What if both of those realities can coexist? What if sales are up via phone because of how the website supports it? And we didn’t even mention whether or not sites sales are increasing.
Solution: Is there more info we need to include or ask for? Yup! Layer in some context on data points to help explain their relationship to the whole story.


“Often when something is true for the part, it does also apply to the whole, or vice versa, but the crucial difference is whether there exists good evidence to show that this is the case. Because we observe consistencies in things, our thinking can become biased so that we presume consistency to exist where it does not.”
This logical fallacy claims that evidence about one aspect of something must be applied to other parts, or all, of it. If you’re working with customer segments, watch out for this one.
Division/Composition Venn Diagram
For example: Your team notices that visitors to your client’s website are overwhelmingly doing so from a Windows device.
Logical fallacy: Someone says, “Our client’s customer base loves Windows products!”
Solution: Include more data. Do only online customers love those products? What if your client offers a different way to buy that doesn’t attract any specific technology user? What if your client has loads of iPhone users who are forced to use Windows to purchase?

Three Penultimate Questions

Finally, I ask myself these questions before making a data claim.
Is my claim verifiable?
Is it even possible to prove wrong what I just said? If I said, “People 3000 years ago would have loved this app,” how could you prove me right or wrong? It’d be tough. Aim at making claims that could be proven.
Could something else be causing it?
Is my claim based on correlation or causation? What else could cause it? How could I confirm causality? Consider other alternatives; you might find beautiful concepts that you’d otherwise overlook.
What would it take for me to change my mind?
What evidence could convincingly counter my claim, forcing me to change my mind? Be skeptical of your own thoughts so others don’t have to carry that burden for you.

Understanding logical fallacies won’t guarantee correct or coherent claims, but it’ll help your conclusions better align with reality. To truly unlock the answers hiding inside unanalyzed data, be truthful to the evidence. Knowing where you’re wrong can guide you towards being right.

About eROI

eROI crafts compelling digital experiences across email, web, and social channels. Our work has been consistently successful in driving revenue and exceeding goals for our partners.