Prototyping: Iterative vs. Parallel
You hear the word “prototyping” thrown around a lot. Not just by designers, but by just about everyone in the tech scene. It seems to be one of those jargon words that means different things to different people. It’s also a word that gets misused a lot (just like the term “user experience” — but that’s a story for another day). I remember when I first entered the world of design and started hanging out with design folk who used all these seemingly big words like “prototype”, “agile”, “iterate”, and “MVP” that all felt a bit intimidating, but that was only because I didn’t know what the words meant.
So, what does “prototype” actually mean? (If you feel like you’ve got a solid understanding of what it means, feel free to skip down to “Parallel and Iterative Prototyping — What’s the Difference?” a few paragraphs down).
Furthermore, what’s the best way to prototype? (Hint: there’s more than one strategy for prototyping.)
In this article, I’ll give a quick intro to prototyping, and then delve into the differences between iterative prototyping and parallel prototyping (two different approaches to prototyping).
Prototyping — just another design buzzword?
Prototyping is an activity that allows designers, entrepreneurs, and engineers to rapidly create designs and evaluate how useful or successful those designs are. This allows businesses and campaigns to increase conversions, improve usability, or convince people to donate $60 million. It’s a crucial part of UX design and research (as well as innovation and creativity as a whole), so knowing what prototyping is and the different approaches to prototyping is essential.
Prototyping in HCI/Academia
I highly recommend this video on prototyping, created by Scott Klemmer when he was a design professor at Stanford:
Essentially, prototyping is rapidly creating a “rough draft” of a design idea that can be used to gather feedback. By testing out different ideas with prototypes, you can improve your designs. The critical part of prototyping is not the artifact itself, but the feedback you receive, as this is what will inform your next iteration (or version) of the design.
Think of prototypes as a means of asking questions. For example, “Will users like feature/design X?” Don’t just ask them. Make a prototype and give it to users to find out. As professor Klemmer says: “Prototypes are questions. Ask lots of them.” (I love that line, but that’s because I’m just a design-thinking nerd.)
Prototyping in Everyday Life
In one of my favorite presentations, Tom Wujec’s TED talk illustrates the power of prototyping using spaghetti and a marshmallow.
During Tom’s “spaghetti tower challenge”, teams have 18 minutes to build the tallest spaghetti tower capable of supporting a marshmallow on the top.
Who builds the tallest towers? He’s seen that kindergarteners consistently outperform lawyers and business school students by making taller spaghetti towers.
Why are kindergarteners better at this task?
It’s because kindergarteners engage in rapid prototyping. They, unlike adults, don’t spend their first 15 minutes debating how to design the best tower. Instead, they immediately start making towers, allowing them to quickly discover what works and what doesn’t by testing out different structural designs. This is the process of prototyping — designing, testing, (learning from the test,) and repeating. So, to answer the original question, prototyping is certainly more than just a buzzword (even if it gets used like one).
Parallel and Iterative Prototyping
What’s the Difference?
Traditionally, designers prototype in an iterative/serial fashion: they create one prototype, evaluate its success through testing, then use what they’ve learned to make improvements, and repeat the process. (At least 3 repetitions of this cycle is recommended by the Nielsen Norman Group.) It’s an effective way to design. But, there’s more than one approach to prototyping…
Parallel prototyping is an approach to design that involves creating multiple designs at a time (in parallel), opening up a wide variety of options before testing them out and soliciting feedback.
So which should I use? Parallel or Iterative?
There is no doubt that prototyping (whether iterative or parallel) will vastly improve your design. Both will help you achieve your goals. Some evidence suggests that combining both techniques is most effective. But I’ll make the case that parallel prototyping is most often the way to go. There is a slew of evidence highlighting the benefits of incorporating parallel prototyping into your design process. Here are just a few of my favorite reasons why parallel prototyping is better than iterative prototyping:
Better Usability in a Shorter Timeframe
In a case study done the Nielsen Norman Group, a team of designers worked in parallel by designing prototypes independently of one another and then combining the best of each design. In addition to speeding up their process, this method measured out to a 70% improvement in usability from the first to the second design (as compared with just 18% when using traditional iterative design).
Diversifying to Find the Global Optima
Sometimes it is quantity over quality.
When brainstorming design ideas, it’s best to come up with as many ideas as possible (of course, you’ll need to narrow it down, and you can’t test every idea, but…). The more numerous and varied your design ideas are, the more likely you are to come up with a better solution. One problem with iterative prototyping is that it pushes designers towards a better version of just one design, rather than considering a wider range of possibilities. Parallel prototyping, on the other hand, encourages designers to explore the spectrum of possible designs, and helps designers discover the global optima, rather than just the local optima.
Additionally, HCI researchers have seen that parallel design leads to more diverse designs.
Parallel prototyping helps designers reach better solutions by encouraging a more diverse set of design prototypes.
Avoiding Emotional Attachment… to Your Design
If a designer didn’t have any constraints, they’d spend all the time in the world perfecting their design and all of its details. After putting so much effort into their design, they’d naturally become attached to it. But attachment to a design isn’t helpful because it can prevent you from seeing better possibilities.
Parallel prototyping also helps designers receive critique more objectively. In one study, more than half of the participants who did iterative prototyping reacted negatively to expert critique, whereas none of the participants who did parallel prototyping reacted negatively to critique. Designers who prototype in parallel are less invested in any single idea, so they’re more receptive and open to critique.
How to Incorporate Parallel Prototyping into Your Work
Parallel prototyping is all about coming up with many design alternatives at once, rather than focusing on just one at a time. But it can be hard to generate lots of original ideas, especially if creative block sets in. Try these tips with your team to come up with as many design ideas as possible:
- Keep up the momentum with paper prototyping. Drawing with a marker and paper allows you to quickly sketch out lots of alternatives while keeping you focused on the high level things instead of spending too much time on the details.
- Hold a collaborative sketching session. Collecting input from team members not only gets new ideas on the table, but it helps foster discussion and investment in the design among teammates. Even teams that aren’t colocated can hold a remote sketching session.
To summarize, parallel prototyping offers a variety of benefits that traditional (iterative) prototyping doesn’t, like improved usability, increased diversity of designs, and improved chances of discovering a better design.
Thanks for reading! Any thoughts on iterative vs parallel prototyping? Is parallel prototyping actually better than iterative? Or is it just design-thinking nonsense that doesn’t hold up in the real world? Please feel free to share your thoughts :)
Adapted from my original publication on Viget.com
I highly recommend the following videos, research, and articles about prototyping (which were linked above in the article):
- “The Power of Prototyping” video lecture by Scott Klemmer provides a great introduction to rapid prototyping with some classic examples. It’s just one of many lecture videos from Klemmer’s Human-Computer Interaction video playlist. You can also take any of his online Coursera courses, like the “Human-Centered Design: an Introduction” class.
- “How Prototyping Practices Affect Design Results” article by Steven Dow forgoes academic jargon to provide a concise (and even entertaining!) summary of design research findings.
- “Improving System Usability Through Parallel Design” research article by the Nielsen Norman Group gets a bit technical and academic, but provides a nice description of a study demonstrating the benefits of parallel prototyping. The Nielsen Norman Group is just another authority that designers should have on their radar.