Being an entrepreneur means making decisions, many of which involve choosing between two sides of an argument. For example, two vendors might be vying to provide a service or two employees might differ on how to approach a problem.

Smart decisions are difficult to make if you lack data, but even if you've got plenty of data, you can easily make bad decisions if you're swayed by arguments that seem plausible but are actually bogus.

Here are the five most common "reasons to make a certain decision" that surface in the business world:

1. Misleading Analogies

Analogies are useful and appropriate when they're used to simplify a complex concept. For example:

  • Investor: "Why is it so difficult to design semiconductors?"
  • Engineer: "Etching circuits on silicon is like drawing a blueprint with a piece of chalk."

However, analogies are misleading when they're trotted out to add emotional weight to one side of an argument. For example:

  • HR manager: "We need to spend money to reduce workplace stress."
  • CFO: "No we don't, because diamonds only form under pressure."

To avoid misleading analogies, ignore the emotional content and consider whether the analogy actually corresponds to facts. In the example above, for instance, the way diamonds are formed is utterly irrelevant to the ROI of a stress-reduction program.

2. Ad-Hominem Arguments

Many very bad decisions get made when decision-makers get caught up in who's making the argument rather than the argument itself. For example:

  • CEO: "Why should outsource customer service?"
  • Consultant: "Because I'm the expert."

In this case, the consultant's expertise (or "credibility" as it's often called) is irrelevant because important decisions demand evidence not just expertise.

Perhaps the biggest danger in believing ad-hominem arguments is that they can make you blind to good ideas, simply because they come from the "wrong" person.

  • CEO: "Joe told me that running two assembly lines is a waste of money."
  • CTO: "What does he know? Joe's only a technician!"

To avoid ad-hominem arguments, take into account who's making an argument (because it may be influencing his or her thinking) but assess the argument on it's own merits.

3. Irrelevant Anecdotes

Telling stories (anecdotes) is a wonderful way to illustrate a business point and make something abstract seem more personal and human.

  • CSO: "You claim to have the best customer support. How so?"
  • Vendor: "Well, we've won several awards, but let me give you an example. About two years ago..."

However, anecdotes are not evidence and can easily drive decisions that are questionable or even outright stupid.

  • CEO: "What's the priority for porting this software?"
  • CSO: "I was talking to the CIO at XYZ, and he told me that they'll find another vendor if we don't do Blackberry first."
  • CEO: "Okay, let's do Blackberry before Android."

To avoid irrelevant anecdotes, notice whether the anecdote is 1) illustrating an established fact or 2) attempting to establish a fact. If the latter, ignore the anecdote.

4. Unproven Causality

Just because two things at or around the same time, it doesn't mean that one of them caused the other. For example:

  • Fred: "I just read that 95% of all embezzlers drink coffee!"
  • Mary: "Wow! Let's throw out our coffee maker!"

While that particular example is obviously stupid, I have personally seen dozens of executives make decisions based upon the same flawed reasoning.

  • Vendor: "We gave our training at your Albuquerque office and sales went up by 40%!"
  • CEO: "Wow! Let's roll out the training nationwide!"

In the above example, there may be a connection between the training and the increase in sales, but the increase might be due to something else entirely, like a single big customer calling out of the blue and placing a big order.

Unproven causality is the most dangerous when the causal connection "seems right." To avoid it, set aside your preconceived notions of what ought to be true and instead look for the mechanism by which one event created the other.

For example, in the case of the sales training, you'd want to dig deeper to see what actually happened. For example, if the training was "how to close faster" and the average close rate rose after the training, it's fair to say that they're causally connected.

5. Foregone conclusions.

This takes place in business when a question makes assumptions in such a way that there's only one possible decision. For example:

  • CSO: "You said we need to increase sales by 10%. To do that, we'll need 5 more salespeople. Should I start interviewing?"
  • CEO: "OK."

The assumption in this case is that the only way to increase sales is to hire more salespeople. While this may indeed be true, the CEO is not really being offered a choice. Other approaches (like a change in sales process) aren't even being considered.

Another certain sign that you're being presented with a foregone conclusion is when you're presented with multiple options, but only one makes any sense. For example:

  • CSO: To increase sales we can 1) completely overhaul our product line, 2) reduce price below cost, or 3) hire more salespeople.
  • CEO: Let's go with option 3.

To avoid foregone conclusions, pay less attention to the question itself and more attention to the assumptions that lie behind the question. Don't make a decision until you're reasonably certain that the assumptions are objective and complete.