We all know that we can’t believe everything we hear and read, don’t we? Yet many of us are naturally trusting, leaving us open to fraud in the most unlikely of places. Let’s look at a real-life work scenario in which a bit of skepticism would serve us well.

Imagine that your boss has asked you to do some online research to find arguments supporting a new project or change of strategy. Naturally, you are thrilled when you come across a recent study on a respected business outlet that showcases exactly what you were looking for. You send the article to your boss right away, and before long, your entire initiative hinges on the claims in this single piece.

Trouble is, what if the article isn’t entirely accurate? I’m not saying that the author set out to mislead you, but let’s face it, most websites do minimal fact checking and editing. Unfortunately, it’s likely that the results were at least somewhat editorialized so the author could make his or her point. It’s also possible that the study on which the article was based was not scientifically sound or relevant in the first place. For instance, perhaps the sample was only a handful of people, or the respondents were only located in the U.S. If you are looking for support for a significant global trend, results like this aren’t going to cut it.

So, if you find a piece of research on the Internet and want to use it for something important, don’t risk relying on false conclusions and having to start over either before or after a scandal occurs. Do your homework upfront. Locate the original source of the research (journal article, etc.) and read the whole paper. If there are parts you don’t understand, tap a colleague or academic friend who can help you decipher them. Make sure the study’s overall conclusions enhance your argument, not just one piece taken out of context. Keep the original document on hand so that you can formally cite it later.

If, despite your best efforts, you can’t locate the original source, that’s a red flag. It could still be legitimate, but you simply don’t have enough information to risk it. Take a little extra time and move onto a stronger selection.