Share on

LinkedIn Facebook Twitter

Imagine you’re at a vintage store. After a minute of window shopping, you’re drawn to a fancy silk shirt. You check its price tag, and since you see it’s quite expensive – $150 – you put it back on the rack. However, determined to find something similar and cheaper, you continue with your search, only to find that other shirt prices range between $500-850. What do you do next? You go back to your first find and buy it. Know why? Because of a cognitive bias called anchoring.

Anchoring is the fact that people tend to cling on to the first piece of information (or anchor) they encounter, and let their subsequent actions, such as estimates, arguments, and conclusions, be made in relation to it. Think of it as a shortcut or heuristic our brain takes to speed up the decision-making process.

In the above example, the leading anchor was the $150 price tag, which helped you make a price comparison and conclude that the $150 shirt’s a steal!

Origin of Anchoring Bias

Back in 1974, Kahneman and Tversky conducted a study in which one group of high school students was asked to estimate the result of 1x2x3x4x5x6x7x8, and the other group was asked to calculate 8x7x6x5x4x3x2x1. They had only five seconds to answer.

Now, although the answer to both questions is 40,320, the groups gave different answers. The first group’s median estimate was 512, while the second group’s median estimate was 2,250. What’s the reason behind this big a difference?

Basically, faced with a time limit, participants quickly anchored to the product of the first few numbers of the mathematical expression (1x2x3=6 & 8x7x6=336), which then influenced their estimates: a smaller number for an ascending sequence and a bigger answer for a descending sequence. Ultimately, this experiment proved Kahneman and Tversky’s prediction that people rely on the initial cue to make estimates, even though it may result in an error in judgment.

Examples of Anchoring Bias in Action

Let’s look at how some brands use the Anchoring Bias to appear affordable and increase the perceived value of their products and services.

#1: Display Original and Discounted Prices Next to Each Other

We often rely on the price of a product to determine its worth. Expensive products, for instance, are assumed to be excellent quality, stylish, long-lasting, and hence worth buying. And this is exactly the reason why we’re tempted to buy them for much less on sale.

Here’s an example from the Columbia website where the original price is struck off, anchoring users to consider the discounted price as a good bargain.

Note: The Anchoring Effect can backfire if there’s a ridiculous gap between the original and discounted price. For example, wouldn’t you be suspicious if I propose selling you a brand new BMW whose actual price is $45,000, for $8,000?

#2: Offer a Suggestion

Ever noticed transparent donation boxes for various charities at checkout counters in stores and eating joints? How do you decide how much to give? My guess is you’ll look at the most popular bills in the box. After all, with people from behind the counter looking at you, you want to make sure you come across as a person with a big heart and an equally big wallet!

But unlike in the example above, it’s a good idea to take cognitive load off a customer’s mind. Notice how the folks at The Guardian manage that. The default contribution value works as an anchor, hinting it’s a popular option, thus making people conform to it. Whether people donate the recommended amount or not is a different matter, but, at least they have a number to adjust to.

#3: List the Most Expensive Plan First

Agreed it’s not quite the trend, but this Conversion XL study might convince you to change your tactics. Not only did they find out that ordering pricing plans from high to low results in higher revenue, but another of their studies revealed how highlighting a recommended plan results in even greater results.

MailChimp does both. Straight off the bat people anchor to the expensive plan, which makes other options more appealing in comparison.

Another reason this strategy works is the operative rejection-then-retreat technique. Robert Cialdini talks at length about it in his book, Influence. Essentially, you make a larger-then-smaller request so that even if someone rejects it at first, guided by the reciprocity principle, they’ll give in to the next request.

To prove it works, he cites a report wherein billiard table dealers were able to increase the average table sales by 50% by first showing the expensive models and then allowing customers to shop the rest of the collection. Suffice it to say that by organizing your plans from high to low, your prospects are bound to see the mid-price options more favorably.

#4: Leverage Multiple Unit Pricing

It’s a pricing technique where you offer a lower price per unit if someone buys a product in bulk. For example, a set of three mason jars for $5 instead of $2 for each.

The main purpose is to get people to anchor on the money-saving aspect. It makes the offer tempting and convinces them to buy more units even when they don’t actually need them. Supermarket giants have been capitalizing on this for years. An example from the E.Leclerc website below:

Other Cognitive Biases Used In Conversion Rate Optimization

Bandwagon Effect

People’s tendency to do something merely because others are doing it, irrespective of whether they agree or not, is what psychologists called the Bandwagon Effect.

One of the best ways you can harness its conversion optimization benefit is by using social proof on your website. Display how many businesses you cater to, your client names, ratings, reviews, and testimonials to establish credibility and influence prospects to ‘hop onto the bandwagon’. Take a cue from the company Box on how to drop numbers and names.

Barnum Effect

It’s a psychological phenomenon whereby people accept vague and generic statements as accurate descriptions of their own personalities, without thinking they could be applicable to other people, too.

You, too, can trigger this bias by personalizing user experience. Your prospects will feel unique and heard, when, in reality, you’re appealing to a whole segment, not one person. Think about personal recommendations based on visitor segmentation.

At AB Tasty, we ran a campaign for one of our clients, Sephora, that involved running a personalized, promotional banner exclusively for loyalty card holders. And guess what? The banner led to a 16% increase in transaction.

Framing Effect

The Framing Effect is a cognitive bias where people make decisions based on how information is presented to them as either a gain or a loss.

Generally speaking, people avoid risk when the option is positively framed, but become more risk-averse when a negative frame is brought into the picture. Why? Since the pain of losing something is felt much more than the happiness upon gaining something.

To capitalize on this psychological quirk, use urgency to frame the experience for your visitors. You can shine the spotlight on real-time stock levels, the time ticker, or competition. The intention is for visitors to notice the inevitable loss and avoid it, come what may!

Peak-End Rule Effect

Peak-End Rule Effect is a convenient cognitive shortcut our brain takes to recall an experience as good or bad based on how we feel at the peak (when emotions are intense) and at the end of the event.

From a CRO point of view, this means the focus should be on creating a user journey that concludes on a high note. For instance, at the checkout stage, don’t force registration, rather give visitors the guest checkout option, to lessen any potential for frustration.

Get Ready To Trigger Anchoring Bias

It’s an excellent psychological tactic to boost your product’s perceived value and influence buyers’ decision-making journey. But first, you’ll need to create anchors that spark a favorable response from your consumers. You may not perfect it in one go, which is why it’s recommended you never stop A/B testing.

Share on

LinkedIn Facebook Twitter

Share on

LinkedIn Facebook Twitter

Did you know that it takes about 50 milliseconds for users to form an opinion about your website? Or that 88% of online consumers are less likely to return to a website after a bad experience?

Well, now you do. And what these statistics imply is that you can no longer push user experience design to the back burner. Rather, know who your target audience is and involve them in the design process. Ask them questions to understand their struggles and desires. Eventually, you’ll begin to notice patterns in their answers that you can then use to identify design solutions.

But it isn’t as easy or straightforward a process as it seems.

When you test your website with users, a cognitive bias called the Framing Effect can interfere. If you remember, we’d talked about it at length with respect to loss aversion. If not, give it a read. In the meantime, let’s go over its Wikipedia definition:

Framing Effect is where people decide on options based on if the options are presented with positive or negative semantics; e.g. as a loss or as a gain.

Put simply, people’s responses are influenced by the way a piece of information is presented to them. In your case, as a UX researcher and designer, it means that led by your own biases, you may frame questions that confirm your hypothesis instead of finding the objective truth. And that’s a major fail because if you don’t recognize it early on, you’ll end up with a website designed for yourself.

With that in mind, let’s figure out how you can you keep this bias in check.

Don’t Ask Leading/Manipulative Questions

…for the simple reason that they can prime users and influence their answers. They will say what you want to hear instead of giving you honest responses.

To understand this better, read the following questions out loud. Which of them, according to you, will provide more reliable feedback:

• I saw that you were confused when navigating the website. What was the issue?

• What was easy or difficult about navigating the website?

• You were having difficulty navigating the website. Why so?

My guess is that you’re going to pick #2 because #1 and #3 are leading questions. They encourage a certain type of answer that can result in inaccurate feedback and bad decisions. Wondering how? Let’s break down all three questions.

• I saw that you were confused when browsing the website. What was the issue?

The problem: First, the interviewer assumes the user was confused. What if they were not? But now that they’ve said it, the user may believe they were having trouble browsing the website. Second, they use the word issue, which seems to imply that there must be an issue. What if there was none?

• What was easy or difficult about browsing the website?

This is a good question because there’s no pressure to answer a particular way. Left to their own devices, the user will stay true to their experience. In fact, the interviewer may even get surprising insights that enable them to see problems and solutions from a completely different point of view.

• You were having difficulty browsing the website. Why so?

The problem: Like in the first question, here too, the interviewer assumes that the user was having troubles navigating. This leads them to believe that they were indeed browsing with great difficulty. The outcome? A biased and unhelpful response that may not necessarily truly reflect their actual experience and thought process.

But how do you get around to not writing leading questions?

• You purge them out of your system and out on the paper. Forget about whether they serve your confirmation bias or not. #NoFilter
• Leave them on the table for a day or two.
• Next, edit and re-frame questions until they are neutral and don’t stick out like a sore thumb.
• Run the list past your peers because two pairs of eyes are better than one.

Et voilà! You have successfully made it to the other side of the fence. That being said, you may still make mistakes because the Framing Effect isn’t easy to overcome. So let’s move on to know what more you can do to stop it from messing with your efforts.

Watch Your Words

When a user asks a question, Kara Pernice, Senior Vice President at Nielsen Norman Group recommends you try the Echo, Boomerang, or Columbo techniques. Here’s a summary of what she says in the article:

Echo

Repeat the last phrase or word the user said in a slight interrogatory tone. This will naturally put the user in the mindset of answering the question by elaborating on what he meant by those exact words. For example:

• User: This table is weird, well, hmmm, not sure what, uh…
• Facilitator: Not sure what? Or, Table is weird?

Boomerang

Bounce user’s questions and comments right back at them to make them solve the issue by imagining they aren’t in a research environment. For example:

• User: Do I buy without creating an account?
• Facilitator: What would you do if you were doing this on your own?

Colombo

Ask only a part of the question so that the user helps you by answering their question. It’s effective because fewer words mean that you’re less likely to influence their answer. For example:

• User: If I close here will I lose my work?
• Facilitator: Uhm, you are wondering if [pause] you might [pause.]
• User: I am just not really sure if I should pick “close” or “cancel” or “ok.” I guess I don’t know the difference between these buttons.

Kara concludes the article by saying that these techniques aren’t a ‘license to interrupt the user any time he makes a sound’. Determine:

• Whether the user was thinking out loud or if it really was a question that must be answered.
• Whether the comment made was audible to you to get talking with them.
• Whether there’s any real benefit from probing them, or you have enough information.

Bonus advice: Record usability sessions to analyze your non-verbal behavior because that can be leading, too. Check if you were you smiling and nodding to get a response, or were you poker-faced?

When all is said and done, let’s not completely discredit framing. It can be useful, especially when you want to elicit a user’s true behavior. Let me illustrate this with an example.

Imagine you want to know a user’s path on your website. Frame your question in a manner that introduces context and focus without being suggestive at all. How do you do that? Instead of saying, ‘You are here to buy a bookcase. Now find one, add it to your shopping cart and checkout’, you say:

‘You have 50 books lying stray in your house. Please look for a suitable product to stack them, add it to your cart and complete the checkout process.’

Doing this, you’ll have your hands on rich insights because neither did you instruct nor prime users to search for a bookcase. Had you done that, they would have straightaway used the search functionality on your website instead of acting on their own.

In short, framing allows you to establish boundaries. You become more focused to form research questions that support your research goals. Everything else gets excluded.

Moving on, let’s talk about how design decisions are vulnerable before the framing effect.

How the Framing Effect Affects Design Choices

The way test findings are framed can heavily influence how it’s interpreted. NN Group tested this and found out that decision frames do affect design choices.

About 1037 UX practitioners participated in a quiz and were randomly shown a version of a hypothetical study’s findings: Half of them saw the negative frame and the other half saw the positive frame. In the end, both groups were asked the same question: ‘Should the search function be redesigned?’ Practitioners that saw the findings expressed as a failure rate were 31% more likely to believe the design needed an overhaul.

Then, there were a few of them who said they weren’t sure or that the design needed no change at all. And, that’s quite interesting because whether a website needs to be redesigned depends on a range of factors, such as the type of website, the overall importance of the search function and implementation costs to the participants. Since none of this was mentioned in the quiz, it’s clear that these participants were more critical of their choice and didn’t want to make snap judgments.

Much like these practitioners, you, too, can make reason-backed decisions. Here’s how:

• Critically analyze decision frames rather than jumping to conclusions. Refuse to pick one frame over another in case there isn’t enough data. Find out what’s amiss to support your argument.
• Re-frame the findings and look at it from a success/failure or gain/loss perspective. You should then have a fair idea if you were being framed.

Connecting The Dots

We can all agree that the Framing Effect is like mercury that can slip through the tiniest crack. You can’t pull a trick from under your hat to banish this deep-seated bias, but you can definitely become more mindful of the trap.

Ask if the question is framed to yield a response you want, or will it motivate the user to be unbiased. What if you word the question differently? Will it change the answer? Point being, once you are more critical you will have more confidence in your choices that will reflect in your research practice as well as your website and products.

Remember, you can use this bias to work for you, especially to support research goals and help boost e-commerce conversions through loss aversion tactics.

Share on

LinkedIn Facebook Twitter