By Shirin Oreizy, founder of Next Step.

By now you're probably familiar with nudges and biases. From New York Times bestseller Predictably Irrational by Dan Ariely to Britain's Ministry of Nudges, it seems like insight about human decision making is constantly making the news. And no wonder -- it's fascinating to think that you can predict (and potentially drive) human behavior.

In my design and marketing firm, we look to the latest research in behavioral science to create sites and marketing strategies that help companies connect with and motivate their users. But before you memorize Wikipedia's extensive list of different cognitive, social and memory biases, you need to realize that these biases and interventions are not foolproof. In fact, if you're not careful, their misuse can backfire.

Don't Get Damaged by Defaults

Default settings can be very helpful because as humans, we're notoriously lazy. That's not a judgment; it's evolutionary. We simply don't like to expend more effort than necessary.

In the 10-plus years that we've been helping clients drive user behavior, we've learned that the easier you make it for people to sail through your checkout process, the more people are more likely to finish. Defaults are so powerful that they've even been shown to increase organ donations in Europe by 56 percent.

In my own marketing firm, we don't deal with livers or hearts, of course. But a good default makes it easier for clients to select a shipping option and takes some of the pain out of scheduling a demonstration. Although this landmark study got a lot of people excited about decision architecture, misusing defaults can be highly damaging in the long run.

Remember how much trouble Ryanair, the low-cost Irish airline, got into by making customers opt-out of additional fees? This defaults abuse made customers sign up for unwanted insurances, hotels and rental cars unless they opted out. Eventually, the EU got involved and passed a law making these type of nudges illegal.

Takeaway: Nudge wisely. Yes, we're lazy. But we also resent being tricked. Burn us once with a malicious nudge, and it's the last time we'll trust you.

Use Social Proof to Erase Worries

If you're like 70 percent of Americans, before you buy online, you check out what other people think. As soon as you see that that product has earned thousands of five-star reviews, you become more comfortable with the notion of sending an unknown vendor your money. That's social proof at its best.

Now compare how you feel seeing the same gadget with only a couple of social media mentions. Still ready to whip out your credit card? The lesson here: If you don't have a significant amount of social proof, skip it. Otherwise, you could be actively raising doubts.

Look at the Ripple Effect of Comparison

Did you have a favorite teacher in school? Do you remember what happened when he or she praised someone for a specific behavior? It triggered others to follow suit, hoping for the same good feeling.

Savvy companies use this ripple effect of social norming as well, persuading others to go along with the crowd's good choices. Now, watch this social norming nudge backfire: an energy conservation program in California sent reports that compared each household's energy use with its neighbors. Nudge psychology would predict that social norming would inspire each household to aim for the "above average" category. Instead, it turns out that politically conservative households rebelled and increased their energy use.

Don't try to nudge anyone unless you've carefully assessed both the people and the situation. And don't release your nudge into the greater world unless you've tested it out on a select group alongside a control group. When my firm uses research to drive user behavior, we always recommend running a few experiments for each behavioral science intervention.

We've discovered that carefully crafting and testing each hypothesis is the only "trick" to preventing nudges from backfiring. When it comes to driving human behavior, you either step carefully, or you step in it. Instead of following the intervention of the week, put time into figuring out what works best for your unique user.

Shirin Oreizy is the founder of Next Step, an award-winning marketing agency that uses behavioral science to drive user behavior.