Seven Questions

Successful Prototypers

Denal Dahl, UNC doctoral student, Kalia Health
jeff terrell
Josh Terrell, Department of Computer Science
Glenn Walters, Department of Applied Physical Sciences, BeAM
QUVI student team

Student and faculty innovators explain how to build the right kind of prototype for your situation, helpful digital and physical tools to consider – and how to get started.

You have a great idea – a really great one. You’ve done your research, spoken with peers, colleagues and potential customers, and now you’re ready to create a prototype to simulate a product. But where do you start? Can you create a prototype on campus? When is the best time to create a prototype? Innovate Carolina supports many students, faculty and alumni who are on this same quest. We’ve heard similar questions and checked in with faculty and student experts who have guidance to offer.

We recently spoke with:

Denali Dahl, doctoral student and CEO and co-founder of Kalia Health, engineering safe, affordable, early-detection products to improve maternal health.

Jeff Terrell, professor of the practice of computer science and director of the App Lab.

Glenn Walters, professor of the practice of applied physical sciences and BeAM director.

Kush Jain, Alekhya Majety and Harshul Makwana from student startup QUVI, which is on a mission to make water bottle sanitization more accessible to students.

These experts give you insights on functional vs. aesthetic prototypes, how to work within your budget and the physical and digital resources they recommend to get started.    


What’s been your experience with prototypes? And what makes the perfect prototype?

Glenn Walters: I have designed and made thousands of prototypes both for clients and for my own purposes. These prototypes have ranged from tiny mechanical devices to large-scale systems and from furniture to software designs to electronic devices to wastewater treatment systems to laboratory instruments. The “perfect” prototype is the one that serves as the most appropriate prop for telling the story that needs to be told at a specific place and time to a specific audience. Prototype functionality is most important when meeting with engineers and when demonstration of true technical feasibility is necessary. Prototype aesthetics/ergonomics are most important when marketability and buyer feedback are being evaluated. Sometimes I’m the audience that I am prototyping for. Many of the early prototypes that I make are for my own thinking and visualization purposes – sometimes to test the capabilities of a tool, sometimes just to get a sense of scale. For all but the simplest products, there is no single “perfect” prototype. Identifying audience/user, developing the important stories and quickly producing the minimally viable prototypes to serve each story -- that is perfection.

Denali Dahl: Kalia Health is developing a home-based urine test for preeclampsia. We developed a proof-of-concept prototype that can detect our biomarkers of interest from a fresh urine sample using a lateral flow assay format. For us, what was critical about this prototype is demonstrating that our diagnostic idea is possible. We had the idea and knew that it should be possible based on pre-existing research and literature, but being able to scientifically prove that the diagnostic idea is possible was a critical step. This means it's an idea that's worth continuing to pursue (go/no go decision). For investors, the proof-of-concept prototype is essential to de-risking the idea and showing that our idea can be translated into reality, which is essential for attracting funding. Other prototypes that are focused on aesthetics and showing what the device would look like are also helpful, even if they are not functional. For example, if we are giving a presentation or at a showcase event, having an example of what the device would look like makes the explanation to new people easier – It gives them something to look at, they can handle it, see how it feels. The perfect prototype is one that helps you meet a specific goal or milestone. For us, the functional prototype de-risked our idea and demonstrated proof-of-concept. The non-functional aesthetic prototype helped us explain our ideas at competitions and presentations.

When is the best time to design and develop a prototype?

Kush Jain, Alekhya Majety and Harshul Makwana: When the team is approaching a time to pitch in a competition or a group of investors. Having a non-functioning prototype or MVP will offer validation and help progress your company. We created our first non-functioning prototype of QUVI for the UNC Makeathon to help judges visualize our product and watch us demonstrate it. Furthermore, we created 3D CAD models of QUVI that we put on all presentations to help people visualize QUVI. However, we did not start pursuing creating a functioning prototype until we received enough funding to do so. We applied for several grants and researched places that could build our prototype through connections and our network before we started to build our prototype.

Jeff Terrell: The best time, in a word, is early. People need something to see and interact with to really get your vision. And it's difficult to persuade a partner or investor to engage if they don't really get it. Plus, having a prototype shows people that you believe in your own idea enough to spend some time working on it, so they're more likely to take you seriously.

Once you start talking to others and envision possibilities, you find more opportunities to connect with people from different disciplines who have new ideas. It’s been an amazing opportunity to take that risk and try to reach out and find people who are also interested in thinking a little bit differently. We’re pushing the boundaries of what it means to teach in the humanities.

What’s the first step you’d recommend people take in building a prototype? Sketch them out? Do people need a virtual version?

Jain, Majety and Makwana: Prototyping is really a multi-step process, and it requires a mindset that the prototype you build will likely have numerous iterations before the final product. Therefore, the first step is to take the idea of the final product, and strip it down to the most basic elements. It is important to decide which elements you need to test/work on at first, and ignore (for the time being) everything else. The next step is sketching the prototype out, whether it's drawing on paper or building a 3D model using a computer. This makes it much easier to collaborate with engineers or whoever is going to build the prototype, and can even lead to changes and improvements further along the way. With these two preliminary steps, teams can start making progress almost immediately. It took time and experience for us to understand this early on, but it has helped in iterating on our first, non-functioning prototype.

Walters: Assuming that the appropriate amount of preliminary work has been done (e.g. characterizing user, identifying needs, defining features, etc.), the first step in my opinion is to always start with hand drawn sketches. This is true regardless of the type of product or service. Drawing and diagraming by hand is simultaneously visceral and cathartic. It helps guide the direction of the interpretation and opens the mind to more creative processes. While I do employ a wide range of methods and technologies in prototyping, I almost never start with an electronic visualization of a prototype. Creation of virtual prototype renderings is rarely worthwhile at the early stages. Creation of functional virtual prototypes is often necessary with more complex mechanical designs, but again, it wouldn’t be the first step.

Does it take a big budget to create a prototype?

Walters: Not necessarily. The primary resource an innovator needs to invest in a prototype is their time. The more time that they can passionately invest in personally creating the prototype, the less money they will need. The best innovators will tirelessly learn new tools, techniques, etc. so they can push along the development themselves. The less of their own time an innovator is willing to spend on developing their prototype, the more it will cost – and it is an exponential relationship. If the innovator reduces their time by a factor of 10 it will increase the ultimate cost by a factor of at least 100. Between personal time invested and money invested, the greatest ROI will be on personal time.

Jain, Majety and Makwana: For QUVI, we did not spend any money to create our prototype. We were fortunate to have access to great resources on campus like the BeAM makerspace that allowed us to develop and build our first prototype completely for free. When creating a low-resolution prototype, it doesn't make much sense to spend a significant amount of money developing it because it's just mostly to help yourself and others visualize a very basic vision of your idea. Additionally, this first prototype will most likely still be riddled with problems and errors and sinking large amounts of money into it probably isn't the best use of that money. We used our first prototype more as a proof of concept and to help show the judges in the Makeathon competition how we planned on implementing our product. Even for our working prototype we did not spend thousands of dollars creating it. Rather, we worked with our mentor to find someone who would be willing to build it for a heavily reduced price. This allowed us to use our money towards lab testing and future development.

Q: What process works well for building prototypes? And what role does iteration or refinement play?

Terrell: My specialty is software. In software, I recommend so-called ‘clickable prototypes,’ in which you design the screens that a user of your software might see, then connect them together so that a user can navigate through a pre-scripted experience as though it were real. I recommend Figma.com and Adobe XD for this. Then, to refine your vision, get it in front of people and draw out their impressions. Having a digital prototype makes it easy to iterate and refine your idea in response to feedback. With persistence, you can not only create a prototype that resonates, but identify a core of people who are excited about what you're doing ‒ an excellent source of valuable opinions and evidence that what you're building has value, to show to potential partners or investors.

Jain, Majety and Makwana: Our process for creating a physical prototype entailed developing a 3D CAD model, 3D printing it in the BeAM makerspace, and iterating on it as we applied for grants and received funding before we pursued creating our MVP. This process was advantageous in the sense that we were able to transform our idea into something that was presentable quickly, but we also learned that prototype refinement is a constant process. After we created our first prototype, we continued to brainstorm ways in which we could make improvements to the chassis and its durability, or how we could better contain the UV-C light inside QUVI. When we were starting to build our functioning prototype, having these ideas and concepts in place allowed us to improve QUVI’s efficiency and safety. e limitless possibility.

Q: Thinking back on prototypes you’ve worked on, what do you wish you would have known? Any pitfalls to avoid??

Dahl: It's okay to have multiple prototypes. It's more important to be able to demonstrate specific aspects of your project and be able to convey the message than have the “perfect” prototype that encompasses all aspects of your idea in one go. Even if you don't have funding for a prototype or something cool and flashy, think about what you do have access to and how you can bootstrap your way to a prototype. For us, the home pregnancy test is similar to our preeclampsia diagnostic device, so we started with examining that technology and using it for pieces for our own prototype. Pitfalls to avoid: spend time critically thinking about what's important to show with your prototype but don't let that inhibit you from making progress. It's better to start somewhere, then trying to wait for the “perfect” prototype.

Terrell: Two things. For software prototypes, the most important advice I can give is to use existing design patterns that users already know. (For more on this, see the classic book Don't Make Me Think by Steve Krug, especially the latest edition.) Second, it can be so hard to get out there and put your prototype in front of people, and even harder when they're not impressed or have lots of criticisms. But to be successful as an entrepreneur, you must learn to persist, iterating at each step and pivoting as necessary.

What's your best advice for faculty or students looking to create a prototype? Which resources at Carolina would you recommend?

Dahl: BeAM – the UNC makerspace is awesome. We also found that reaching out to local companies making similar technologies to be really helpful. We were able to shadow researchers in a lab who assemble lateral flow assays and learn the process, before attempting it ourselves. Because we were students, they were willing to help us learn the process. So the best advice would be to find individuals or companies who are doing similar things and ask for advice. Even if they can't provide hands on experience, they are experts in the field and can direct you to more focused resources.

Walters: Ideas are worth little until you personally engage in the production of a prototype. The BeAM makerspaces are a fantastic resource for learning technical skills. The BeAM trainings increasingly include basic prototyping activities, and the spaces feature a wide range of materials and technologies useful for creating prototypes. The Carmichael makerspace (temporarily closed – possible reopening in the spring) is a particularly good resource for physical brainstorming. Kenan Science Library also has excellent collaborative spaces and resources for low-fi prototyping. Our course, APPL110: Introduction to Design and Making, is a full semester dive into human-centered design and prototyping. With extensive use of the BeAM resources, this course is designed to help students immerse in the design/prototyping process. Our primary objectives are to teach a sound, need-based design process, to encourage perseverance through challenging design/build iterations and to help students embrace that failures are an essential and valuable element of the process.