10 Tips for Testing Prototypes
How does testing prototypes differ from Remote Usability Testing and how do you get the most out of it?
The most obvious differences are that you are:
- Testing a hosted lower fidelity experience rather than a live website.
- Testing early designs of the website not the real thing which means you can evaluate its effectiveness at various stages before investing large amounts of money and pushing it live.
As such, remote prototype testing can prove useful in a variety of situations:
- Cost-effectively and rapidly test your potential redesign to hundreds (or thousands) of participants
- Find out first impressions with before and after questionnaires
- Test navigation patterns – ‘where would users click to access specific information?’
- Concept & communication effectiveness testing
10 tips for optimizing your prototype research:
1. Match Method And Interactivity. It’s Never Too Early.
Unlike a live website, the level of interactivity of prototypes can vary. If all you have are static images, you can do screenshot click testing to test the effectiveness of your designs without needing to host them online.
When using screenshot click testing for instance, depending on where participants click, you can then ask different follow up questions. In this way, you can link screenshot click testing together to replicate an interactive (click through) experience. For example, if participants click on a particular menu option, you can then present them with a follow up screenshot click test of that menu option’s landing page. This is ideal for testing the navigation of your early designs.
Subsequently, if later iterations of your prototype are more interactive, you can conduct a task-based usability test. Similar to when testing live websites, you can provide participants with ‘tasks’ or ‘goals’ to complete and capture metrics such as success, time on task, number of clicks and unique page views, ease of completion and satisfaction to name but a few.
2. Give Your Participants Choices.
While your prototype does not need to be high fidelity, your testing environment MUST be somewhat interactive and include both correct and incorrect journey paths. If you find you run short of time to add incorrect journey paths it is important to at least ensure there are other interactive links, even if they just lead to blank ‘dummy’ pages with text instructions along the lines of the following:
“This area of the prototype has not been built yet and is still undergoing development. Please use the back button to return to the previous page.”
The danger or risk in not doing this, is that your participants will know exactly how to continue with the task if there is only one interactive link (the correct one) on the page.
3. It’s A Prototype
It is important to explain to your participants that they will be using an interactive prototype and not a live website. This will ensure participants know how much functionality to expect when completing tasks. For example:
“Please note that the prototype does not represent the look and feel and functionality of a finished website. Not everything is functional. Please be patient, if you click somewhere and go to a blank page – simply use the back button to return to the prototype.”
4. Apples With Apples
If you are using remote prototype testing as a means of benchmarking against competing designs or tracking the evolution of a design, it is necessary to be sensitive of the ‘fidelity’ of your prototypes. Higher fidelity prototypes, which are close to the final product, might be perceived more favourably than lower fidelity prototypes simply because they look more polished, rather than being easier to use or more engaging, per se.
Something which might slip your mind, but is important to remember to check, is whether or not your prototype works as expected in all browsers.
You may sometimes want to restrict the study to certain browsers. If participants attempt to access the study on a browser which is not compatible, they will be presented with a message indicating which browsers they can use in order to access the study.
Always check that your prototype works from within the Remote Usability Testing software environment, not just in its own hosted environment. You may find it behaves differently, in particular when testing mobile prototypes.
8. Behavior Tracking
If you are interested in capturing behavioral data such as clickstreams and heatmaps, make sure you have a means of tracking it.
The best way to do this is to set up a study and run through it as a participant, then check your data. If something you need is not being tracked by your study software, you can use advanced tracking to do so. This involves adding JS code to your website or prototype to do things such as measure success based on the click of a certain button.
9. Don’t Forget Mobile
Test mobile prototypes as well as desktop designs with a task-based for Mobile method which should work for both smartphones and tablets. All of these tips apply for mobile prototype testing as well as desktop.
10. And Agile Prototype Testing
If research is to be done at all in agile development projects, it needs to be quick and prototype-based.
As such, typically, small numbers of participants are used in prototype testing in agile projects. However, this is due to the fact that traditional ‘face to face’ lab-based approaches are used, as they are perceived to deal better with all the associated time (and financial) constraints, rather than any reverence to the idea that a lot of issues can be identified with 5 participants.
While small scale research might well reveal issues about a design, it can be more cost effective to use larger scale remote testing to evaluate designs with hundreds of users in different locations. The latter approach has the benefit of providing sufficient numbers of participants to get statistically significant results, and also provides a broader picture of how users in different countries/cultures respond to the prototype.
The key to this type of rapid remote testing is the ability to recruit quickly. Having a research panel made up of pre-screened participants is useful. Alternatively, building relationships with recruiters in order that they are familiar with your segmentation or relevant personas, such that they are able to act as quickly as possible to calls for participants, will also help.
A good remote prototype test is only as good as your planning. Hopefully the above tips will provide you with food for thought during the planning of your Remote Usability Test.
However, the biggest tip we can offer you when it comes to testing prototypes is: Do it. You should be testing prototypes, as the longer you wait to test and find how effective your design is, the more costly it becomes to make changes to it!
The ability to test prototypes remotely is a huge opportunity and can save your company a lot of money further down the line when you might have invested a lot into your designs.
Customer Success Manager
Customer Success Manager