Blog >

Should We Use Mad Libs Style Form?

By:  Yu Centrik's Cynthia Savard Saucier, UX Designer & Josée Laganière, Usability Expert

One of the primary goals of form design is making sure users can accomplish their tasks quickly while having an effective and positive experience. Usually, we design forms with the intent of improving efficiency and time of completion, reducing complexity, ensuring consistency to minimize the user’s cognitive load and enabling speed scanning. Still, it feels like the design of forms hasn’t changed drastically in the past ten years and we repeat a routine which works well.

However, we recently saw a new form emerging. Praised by the form guru Luke Wroblewski and the usability expert Jared Spool, Mad libs have become increasingly popular. [1]  Let’s try to explore what is going on. 

Where do they come from?

One of the first designers to create a “Mad libs” style form was Jeremy Keith, for his audio sharing site Huffduffer in 2008. He wrote on his blog:  “I share Luke W.’s rallying cry that “Sign up forms must die!” While I wasn’t able to kill off the signup form on Huffduffer entirely, I was at least able to make it human-friendly.”  [2]

According to Wikipedia, a Mad libs is a phrasal template word game where one player prompts another for a list of words to substitute for blanks in a story...[3] 

Basically, the user is asked the same information, but in a narrative format instead of the regular label and field way. Here are some examples on websites: 

Huffduffer

Crowdfavorite

Vaultpress

                                                                                                                                              It’s interesting to note that some software has been using this strategy for a long time, making programming language accessible to non-programmers. See the smart mailbox in Apple’s Mail App, for example.

Apple’s Mail App

Do they really work better?

Although this format is fundamentally against the classical UX best practices, it seems to offer a greater conversion rate than a typical form, in given cases. The literature shows only four studies comparing Mad libs and “traditional” forms in terms of conversion rate – and we can find flaws in methodology in almost all of them.

The Vast.com Example

Lu ke Webrowski’s post started the buzz, reporting on Ron Kurti’s implementation of the Vast.com contact form- the classic example found all over the internet. Their A/B Testing showed an increase in conversion rate between 25 and 40%. 

This Mad Libs form has a real “letter” feel to it. There are additional minor modifications: the phone number field and the comment box. 

http://www.lukew.com/ff/entry.asp?1007

 The Vast.com Example

The Kalzumeus Study

The counter-example is from the post “Lesson from Mad libs Signup Fad: Do your own Tests.” The tone of the article shows the author’s opinion of the Mad libs, calling them “a compellingly goofy design decision”.

According to his study, the A/B testing shows a 22% decrease in conversion rate (27.55% for the traditional vs. 21.73% for the Mad libs).

This test has been severely criticized in terms of aesthetics and copy. The author states that he designed the Mad Libs in about 10 minutes and it shows. However, the conclusion is valid: do your own tests!

http://www.kalzumeus.com/2010/02/27/lesson-from-madlibs-signup-fad-do-your-own-tests/

The Kalzumeus Study

Netmedia Group

After hearing about the Wroblewski study, the people of Netmedia Group decided to try it for themselves, carrying out their own testing. In his post, JR Farr reports their results of an A/B test on one of their landing pages. They put some effort in both design and copy to give their page a personal feeling.

After a few weeks, they reported a 25% increase in conversion rate. More precisely, the Mad Libs form had a 4.3% conversion rate as opposed to 3.4% for the traditional version. Unfortunately, they don’t give any information about sample size.

http://www.netmediagroup.com/blog/25-percent-lift-using-the-mad-libs-form-for-lead-gen/

 Netmedia Group

CFD Group

The CFD Group also heard the buzz and, although they were a little sceptical, tried the same type of A/B testing. They quickly designed a landing page for a lead generation campaign for financial services.

In terms of actual data, this study is the most complete. They report with about 1800 visits to each site after 20 days. The trend showed a 12% increase while using the Mad Libs form (11.79% vs. 10.49% conversion rate). It’s not as spectacular as the 25 to 40% increase from the previous studies, but it is still a nice improvement.

http://www.cdfnetworks.com/mad-lib-form-style-testing-results/

CFD Group 

Should I use it right away?

Although the early studies show an improved conversion rate, there are still many elements that need to be considered. Whilst reviewing Internet articles, looking for examples and information, we found a great number of pros and cons of this new practice. Here’s a compilation:

Pros

1- It’s new, it’s different, it’s fun and it’s eye catching. 

Even if any change in conversion rate is disregarded, the style often creates a buzz around your website since it’s different and amusing. Just look online! 

2- It encourages the user to read the text

In many cases, especially for repetitive tasks, you might want to avoid forcing your users to read long instructions. But in a sign up process, when transferring a small amount of information, encouraging more complete reading may be a good way of preventing error. Perhaps saying: “My name is_______ and my last name is __________” could prevent the majority, if not all mistakes of first and last name inversion. The context and logic sequence of conversation helps in deducing information asked.

3-It’s engaging

Like JR Farr from NetMedia Group, we think that once users start to fill a Mad libs form, they are more likely to finish. FR Farr wrote “With the right copy and layout, it gives the page a personal feeling, while possibly captivating your user.” 

4- It’s “human-centric” and not “data collection centric”. You are a person that lives somewhere. Not a “name, Last name, location”

Instead of asking for information the way companies want it, it presents it in a manner the user familiar with from conversation. This can create a coherent tone of conversation throughout the complete experience: online, on the phone, in presence. An example of this would be the registration process on the phone. It’s a lot nicer to be asked: “Good morning, may I ask your name?” instead of a cold: “Good morning. Name?” It shifts the mindset from “Us asking you questions” to “You telling us who you are”.

5- It’s flexible

Here in the East part of Canada,most of what we do has to be bilingual. With English to French translations, some labels become excessively long, forcing the designer to either leave blank spaces next to the label, pushing the field further away to make sure it will fit once translated or to create a different grid for the French version. However, since the Mad libs style form is not stuck in a typical left or right aligned grid, its restrictions to the length of instructions are almost nonexistent.

 

Cons

1- It slows the scanning speed

Even if you dislike forms in their usual design, they are the quickest way to scan information. In addition, they offer a real dissimilitude between fields and labels reducing the risk of confusion. In the narrative style, you can’t tell right away the number of fields required and a neophyte could get confused.

2- The different contexts may limit its use

Because of the informal tone, it could be inappropriate in certain contexts.  For example, we wouldn’t imagine a national insurance company changing their contact form right away. Some users could feel like the conversational tone used is almost childish. Although a banking company is currently using it on its corporate webpage, proving that it could be appropriate for a wider range of industries than first expected. 

banksimple.com

3- It might create translation problems

Localization is a permanent hassle. In a Mad libs form, the copy may work in English, but how about other languages? Every language has its own syntax and the order of the blanks to be filled may vary from one to another. Yes, we could either rephrase or tweak the code underneath, but traditional forms are easier to manage (but not perfect as stated before).

4- Harder to create

For this type of form to work, you have to create an engaging copy – using complete sentences in a friendly tone and possibly requiring a word order that works in several languages. And it has to be well presented. On the other hand, regular forms have become so standard that even a programmer with no design skills can make a decent sign up.

5- It’s a change from what the user currently expects

Forms are typically used in critical phases of interaction with the user (sign up, making a purchase, becoming a client etc.). By making a major change in that phase, you may lose a potential client, or user. People can be perplexed by a variation of what they are used to and comfortable with... just look at the fuss around every interface change on Facebook!

6- Requires a good level of literacy

Because there is more text, some people may feel more effort needs to be made in order to complete the Mad libs style form. This is problematic for people reading websites in a language they don’t fully understand. This can also be problematic for native speakers, in Canada alone, close to 6 million people have a literacy level below high school level in their mother tongue. [4]

Conclusion

After our research, we still have unanswered questions:

  • To what user profile is that type of form suitable?
  • In what context is it appropriate? (Work, entertainment, gaming?)
  • What are its limitations? 
  • What information can and cannot be collected using the Mad libs style?

In conclusion, although this new method seems promising, more serious studies need to be done, internationally and with a variety of users.

[1]http://www.uie.com/brainsparks/2010/02/11/spoolcast-moving-beyond-static-forms-with-luke-wroblewski/

[2]http://adactio.com/journal/1521/

[3]http://en.wikipedia.org/wiki/Mad_Libs

[4]http://abclifeliteracy.ca/en/literacy-myths

 

Home
User research insights per country...