Screen shot 2012-07-06 at 2.02.39 PM Demo UX Cases

A user test on decoy pricing: steer decisions and increase conversion

Do subjects like human behavior, mental models and cognitive science give you a kick? It does so to us at Usabilla, so we try to share some of the exciting stories we read in posts about Piaget’s theory or the Vampire Effect. Today we’d like to write about the decoy effect, an effect we saw described for the first time in one of our favourite books Predictably Irrational: The hidden forces that shape our decisions, written by Dan Ariely. The effect can shortly be described as follows:

The decoy effect (or asymmetric dominance effect) is the phenomenon whereby consumers will tend to have a specific change in preference between two options when also presented with a third option that is asymmetrically dominated.
Decoy Effect on Wikipedia.

Did you get confused by the definition? No worries, it is much simpler than it sounds, as it plays on one of the factors humans use in decision making: relativity. We, as humans, can easily get deceived by the illusion of relativity. For instance, in the following illustration the circles in the middle are the same size, but appear bigger when placed within the smaller circles on the left, and smaller when placed within larger circles on the right (image taken by interaction-dynamics).

Dan Ariely decided to play with the notion of ‘decoy pricing’ after stumbling upon a subscription ad by the Economist. The ad consisted of three different offers:

Subscription type Price
Economist.com subscription $59
Print $125
Print & Web subscription $125

I guess you’re wondering yourself: who the hell is going to select the only print subscription? Well, so did Dan Ariely. He set up a test with 100 students from MIT’s Sloan School of Management, asking them to select one of the three subscriptions:

Subscription type Price Participants
Economist.com subscription $59 16
Print $125 0
Print & Web subscription $125 84

No surprise here: the students saw greater value in the print & web subscription. Now, Ariely decided to remove the print only subscription and let another 100 students choose. This time the results were more evenly distributed:

Subscription type Price Participants
Economist.com subscription $59 68
Print & Web subscription $125 32

So, what do we learn from Ariely’s small test? In the presence of a decoy price, it’s more likely to control the outcome of the decision making to the desired one.

Testing the Decoy effect with Usabilla

Quite an interesting theory, that got us excited to test it ourselves. We thought the prices of our plans as an excellent subject for testing the decoy effect. Our hypothesis was that by making the small or the large accounts less favorables (decoys), we would push participants to go for the standard accounts.

To achieve we set up three independent tests: no decoy, decoy in large account, decoy in small account. We invited 50 participants to each test. On to the results!

Results

No Decoy

Decoy to small account

Decoy to large account

Account Small Standard Large
No Decoy effect (baseline) $49 $89 $139
Decoy to small account $79 $89 $139
Decoy to large account $49 $89 $199


Account Small Standard Large
No Decoy effect (baseline) 32% 47% 6%
Decoy to small account 24% 58% 5%
Decoy to large account 49% 38% 6%

Take aways and discussion

Now, using the results of the ‘no-decoy’ effect as a baseline to benchmark the two other decoys, we see some very interesting results that we’d like to discuss with you under the comments thread.

As we see in the table above, the decoy to the small account increased the clicks on the standard accounts. This is something that we expected. By anchoring the price close to the standard account, we made participants think something along the lines of: “what the heck, for $10 more I can get a standard account”, pushing standard accounts indeed.

However, the decoy to the large account didn’t push standard accounts, but small accounts! In this case the price is de-anchored from the two other options, making it more like an A/B choice, Small/Standard. Another interesting point is that the percentage of people that clicked on the large account remains the same in all three cases, as random participants might not see the value of a premium account. Last, a limitation of the test that should be mentioned is that it doesn’t take into account people dropping out because of the price change.

Increase conversion rates, and decrease drop-outs by being a choice architect yourself. Influence decision making by making it easier for your users to stay on your page, and sign-up. Good luck!

4 comments

  1. Ruben Timmerman

    Great test, inspires us to try something along those lines as well. Our product is free (lead generation: requesting brochures) but I now think we should try to offer some kind of paid brochure or “Subscribe now” to see if those paid offerings would increase the free brochure requests….

  2. Joost Diepenmaat

    Interesting test! Are the conversion rates for the three variants equal?

  3. Healy Jones

    I’ve seen very similar results with my digital pricing page (http://www.officedrop.com/digital-filing-pricing) – although I’m less concerned with the % choosing each plan and more focused on the absolute number of people picking a plan. I’ve found that the # picking a plan has gone up with the addition of a more expensive plan option pretty nicely, even though very very few people choose the highly priced plan.

  4. Jay

    You guys missed out on a big part of pricing strategies, which is anchor pricing. The reason why “decoy to large account” shifted more people to the cheapest plan was because you didn’t INCREASE the price of the standard plan simultaneously.

    I am willing to wager that the $199 of the Large plan would have served as a reference point and convince more people that a $89+ standard plan was better and more profitable for you in the long run.

    i.e $199, $99, $79 would have worked to make the standard option the most desirable and increase your margins.

Leave a Reply

Your email address will not be published. Required fields are marked *