If you have already scheduled appointments with test subjects, you are ahead of the game. If those people match the personas you’ve identified, even better! If neither of those are true, it’s time to scramble. While the designers and/or developers are making the prototype, other participants spend the day recruiting and scheduling the tests.

Depending on your persona(s), there could be a variety of means to find users. If you’re in a large city and your persona is a consumer, you may be able to recruit some via a Craigslist ad, or even by approaching people on the street. However, like Laura Klein, author of Lean UX for Startups, we also discourage approaching random people in a coffee shop.^[31] First, it’s just plain creepy. Second, you are less likely to find your target user. Remember all that thought you put into the personas and journey on the first phase? Find someone who’s about to embark on that journey, someone who needs that job-to-be-done!

Your existing user/customer base. If you have a user or customer base to draw from, this is a great place to start. Internal sales and marketing teams can often point you in the right direction to connect with the subset of existing customers. At Constant Contact, the Innovation team has built something called “Test Drive,” a network of more than 7,000 Constant Contact small businesses customers that have opted in to test and provide feedback on new products and services. Trace and Richard work closely with their respective clients to obtain and identify the appropriate users for testing. Further, Richard and his team have developed relationships with different organizations such as the Alzheimer’s Association and The Perkins School for the Blind to source test participants when they need to test prototypes for accessibility.

Craigslist. The large classified ad platform has many communities across the globe, and we’ve used their “gigs” section to recruit users for a variety of tests. The warning here is to be sure you’re screening appropriately with a survey, which we’ll get to in a moment. There’s always trepidation given the negative press Craigslist sometimes receives, so be careful and screen your test subjects. We don’t want to read about your prototype attempts in the daily news, OK?

Amazon’s Mechanical Turk (MTurk). This crowdsourcing network gives you access to participants from around the world who are willing to take short online “HITS” for low cost, often pennies. MTurk may not have high-quality responses, nor will you find users that fit into a narrowly defined category. Further, many users on MTurk quickly take tests and can falsely answer your questions, so be sure to include some screener questions to determine which results to keep or discard.

Panel agencies. Services such as Opinions for Good (Op4G), which focuses on nonprofits, and Research Now offer lists of users you are able to filter and select according to your needs. There is often a direct cost associated with these types of agencies as that’s their business model. But they can be a worthwhile investment: you’ll get a higher quality response from someone who represents your user base.

Web-based services. User testing sites such as usertesting.com, trymyui.com, and applause.com offer captured narrative services with a population of users to select from. Because these are remote and unmoderated, you may not be able to interact directly with the users. While these can be hugely valuable at later stages of product development, they may not be best for a design sprint. You want to be able to see the user’s face and body language if you can, because it can often tell you more than she is actually saying.

Your existing social networks. We’ve seen many requests on Facebook, LinkedIn, and Twitter calling for users to test a prototype. It’s always tempting to reach out to people who you know and have an established relationship with; however, be careful, as they may not make great test participants, as they are less likely to give you the honest, critical feedback you need.

A screener survey can help you filter out responses of users who do not fit the profile of the users you have identified (personas, FTW!). Conduct one or two short surveys that filter out the types of users you seek. Here is an example screener survey we used at Constant Contact:

These five simple questions help us to find a subset of users who are small business owners, who have a website and a Facebook page, and also participate in some frequency of email marketing activity. You’ll want to check the business website and Facebook pages to verify it is not someone looking to answer a few questions for a few dollars. Speaking of dollars...

You may be thinking “what if I’m not a designer?!” That’s OK, prototyping digital products can be done using paper, Keynote, PowerPoint, HTML, or a prototyping tool (e.g., Invision, Proto.io, Balsamiq, etc.). You don’t need to be skilled in HTML/CSS or the Adobe Creative Suite to make something to test. The act of making something gives the product life beyond the concepts and allows the design to be experienced for the first time by people outside your immediate team. In the world of digital prototypes, the process of developing a prototype might be more important than the end product itself. This is very similar to writing a business plan. It’s not the end product that matters; it’s the mental exercise of thinking through the plan that will give you insights you need in the future.

The prototype, whether it’s on paper or online, is a test vehicle that will include some content, some of the primary navigation, and possibly important copy or images as well as the key functional elements.

If you’re doing the design sprint as an internal exercise (i.e., without the help of a design partner), then you may only need to include your initial design sprint team. However, if you are fortunate to have the help of a competent designer, you can shift some of the work to them. Alex Britez at MacMillan usually finds himself and one other colleague prototyping while the rest of the team has either returned to their “day jobs” or are helping set up the interviews. If you’re running the design sprint as a design firm, then the client’s core team and your core design team will be involved. Ideally you want the prototyping team to be small. Inviting more people doesn’t invite better input. What was it about too many cooks?

The goal is to get something ready for the validate phase that will take place the following day. So once you have your prototype mocked up, get final feedback from your team and then agree on an endpoint, so you don’t wind up going in circles. This can be difficult if there are additional approvals or sign-offs required. It’s essential that you spend enough time before you start this process educating your clients, participants, and/or stakeholders that prototyping work is not to secure approval on the aesthetics or functional elements; rather, it is to define and refine the flow and navigation of the future product.

It can be very daunting, as we have seen many people conflate the prototype and test with deciding on product direction. When Marc Guy at Faze-1 arrived at this stage, the look on his face was incredibly conflicted. He initially thought that making this prototype was going to set the direction for the product. And because they were a startup at the time, this would dictate the direction his company was headed. C. Todd realized that this was going on in Marc’s head and reminded him this was only a test. No need to break the glass and pull the fire alarm—it’s just a test, and should be used as an opportunity to learn.

The day starts with reviewing the decisions you made the day before and confirming which parts of the previous day’s designs will be used in the prototype(s). If you are planning on making more than one prototype, then we recommend dividing into groups and working on the designs separately. Plan to have working periods followed by rests in approximately 90-minute cycles. We also recommend you start each new design session by bringing the group together and doing a quick review or critique of the previous session (only spend 5 minutes at most for these short critiques).

You only have one day to build your prototype, so the best approach is to spread the design work across your team. Don’t worry too much about the consistency of the design elements. Use the morning to get the critical elements in place and use the afternoon to bring them all together in a consistent design pattern. Having one person to coordinate the designs across the group or groups will help tie things up at the end of the day. This work burns calories so make sure there’s plenty of coffee and snacks. Additionally, other team members may be confirming or setting your interview schedule (more about this later).

You should have a clear idea of what you’re building from the Converge phase. There may be some discussion on the nuances remaining and depending on how far your Converge phase went, you might have a set of sketches defined. Your prototype should include all of the elements that were “touched” by the users in the Test phase. If you’re creating more than one prototype, you’ll need to decide if elements are going to be used across the board or used discretely in the different prototypes. Common elements like browser bars and fonts should be decided on before you create anything. Often we find that a company’s brand and/or style guide has sufficient depth to pull this information from. If there’s no brand guide? Go basic, it’s not about the fonts and colors, so move forward with paper prototype with POP App if you struggle with getting something built. Remember: this is about learning—we cannot stress this enough.

At this stage, less is more—only focus on the essentials. Do not be tempted to fill in all the gaps and fill up the space on the home or landing pages you are sketching. That level of detail will come later. Drawing out the site pages helps everyone involved to visualize what you are doing. Getting everyone on the same page, literally, is critical to understanding where the design will succeed and where it needs more help. As discussed in the previous chapter, the first prototype designs happen on a whiteboard or a piece of paper and are then transferred in more detail to Keynote, Illustrator, Sketch, or other tool. If you’ve separated the work across different team members, you may have some inconsistency in the design. Leave time at the end of the day to clean up the designs and make them look like a single product.

Having created your core product pages or flows, you can begin to build out the complete prototype. As for public-facing sections of your product, we start with the key entry points (e.g., home page, registration page, download page) and a few primary pages. If you’re designing a web application, you’ll also want to design the critical functional areas (e.g., dashboard, profile setup/edit, and uploading files). Each project will be slightly different, but you’ll notice patterns in all of them.

We’re big fans of sketching, but in order to create a “true-to-life” design of a digital product, we recommend building your prototype in something that can be made interactive. POP is a great tool for bringing your sketches to life. Keynote and PowerPoint can be just as valuable and there are lots of kits commercially available on the market for relatively low cost. The most popular one we have seen is Keynotopia,^[32] which offers file formats in both Apple Keynote and Microsoft PowerPoint. At the time of this writing, there are a number of other web-based prototyping tools available such as Marvel App, Proto.io, and Pixate, none of which require coding knowledge.

If you are planning on using sophisticated design tools like Illustrator, Sketch, or Photoshop (or even HTML/CSS code) in the Prototype phase, you risk becoming bogged down. Unless you are an experienced designer who can create the prototype in a few hours (or you have someone on your team who can), we recommend sticking with something simpler.

If you are working with a copywriter, add simple text fields, which can be updated later. We are not in favor of using lorem ipsum to fill in blank spaces. No hipster ipsum nor cupcake ipsum either, though C. Todd might be swayed if cookie ipsum was a thing. Lack of clear content is an indication you are unsure what the value proposition is, and it creates administrative debris to clean up later. In testing, you don’t want to hear things like, “What does this mean?” It distracts from the goal of your prototype: to learn about your assumptions. So use real words and if necessary have your content strategist or copywriter update it either in parallel, or after you are done building your prototype.

In the design firm setting, to achieve the speed of execution our clients expect, we move directly from prototypes to validation and then once we have feedback we move to low-fidelity mockups. The client reviews these mockups and any changes required are made immediately. Because we publish the designs to the Web via Basecamp, we find progress is only limited by the speed at which we can get feedback. In some cases, we’ll move from sketches to HTML/CSS immediately so we can make updates even faster. This can only be achieved if you have sign off on a design flow or concept.

You’ll want to select the simplest, fastest tool that can best validate your assumptions. We recommend any of the following to create your prototype:

Project: Great Engagements

Summary: Before couples can address the details of planning a wedding (decoration, guests, caterers, dress, etc.), couples usually have to decide on a location and venue. The couple is filled with excitement, as the wedding day will usher forth a new life for the couple together. However, the couple might experience a bit of uncertainty as to how even to proceed planning their wedding. We want to capture couples at their “excitement” period and put aside all uncertainty and potential anxiety that the wedding planning process may cause. A customer’s initial engagement with our website should parallel his/her engagement with the wedding process in general.

Prototype: An interactive website (see image)

Learning: With this prototype, the team wanted to learn how well their website captured the excitement of a recently engaged couple.

Project: SMB Marketing Smarts

Summary: Due to a trending increase in mobile device usage and web page analytics showing a greater percentage of hits to its support forums and help articles from mobile devices, Constant Contact was interested in creating an easy-to-navigate mobile app that packaged its wealth of educational resources and included the ability to push a button and connect via phone to the customer support center.

Prototype: An interactive mobile app created with Proto.io (see images). In total, eight different screens were created with interaction hot spots.

Learning: With this prototype, the team wanted to learn if and how Constant Contact users would interact with a mobile app to get help for marketing-related questions.

We mentioned previously that the design sprint is like a mini science experiment. What’s done here is to have another look at the assumptions table. Are there any assumptions missing? At Constant Contact, C. Todd used a validation board at this stage to determine what you may test at the end of the design sprint. You will not complete the entire board, because you have no test results yet, but you should have a clear idea of what needs to be tested. Prioritize your assumptions and determine which your prototype will test.

We use the term “pre-roll” to mean any questions you ask before getting the user interacting with the prototype. These questions establish the context and surface information regarding their behaviors, attitudes, and beliefs. It’s similar to a contextual inquiry, but on a smaller scale. You’ll build rapport with the person you’re interviewing because... they are a human! We’re not saying this to be condescending—we have been guilty ourselves of being a little too robotic on test day. Humans are complex, so establishing a good relationship at the beginning will help you learn what you need.

Example “pre-roll” questions that went along with the SMB Marketing Smarts example shown earlier:

When testing your prototype, you want to see how the user will navigate to accomplish particular tasks. While this may not be 100% like a textbook usability test, there needs to be a starting point, or a reason for the user to interact with your prototype. What specifically would a user do? Consider the job stories you worked on earlier in the week.

Example tasks for SMB Marketing Smarts prototype:

Task #1: You’re just about to send your next newsletter. How would you find ideas for the newsletter content?

Task #2: You just sent your monthly newsletter and now need to share it on social media. How would you accomplish that?

Once the user has completed the test of your prototype, you’ll want to ask a handful of questions to wrap up the session. These questions should help you to understand how easy or difficult it was for users to accomplish the tasks you asked them to complete. We always recommend two concluding questions: one to gauge the necessity of the prototype in the users’ lives and the second to determine what the most pressing things are in their worlds at the moment.

With the prototype made, and interviews scheduled and structured, you’re now ready to put rubber to the road!