Corridor testing: we get fast feedback on layouts
I once wrote about the frequent mistakes of designers, when we completely forget about users, considering that we know how they think and how they feel more comfortable.
It is clear that user testing helps a lot when designing interfaces and whole products, but in order to test something, you need a finished product or at least a prototype.
Creating a prototype makes testing much cheaper, but it also takes a lot of time and effort. Yes, and before creating a prototype, there are often variants of the interface concept, which would be good to test, too, without spending more than half an hour. In this case, the so-called “corridor testing” can help the designer. On Habré this topic was not paid much attention, so I decided to share a few tricks from my experience.
You draw the layout on paper or in your favorite program, exit your office, come up to your colleagues in the office and show them the layout in printed form, or on a laptop / tablet. If according to your layouts you can test the entire script - great, link them to some kind of invisionapp.com and get an interactive prototype.
Testing is called corridor because it allows you to get feedback very quickly on your layout anywhere, literally in the hallway from a passing man.
Creating a model, you obviously understood in which scenario it is used, therefore, demonstrating a prototype to a person, you can ask at least three types of questions:
This question is asked when testing user expectations from the interface. Your users must have tasks that must be completed with your product. Thanks to this question, you can quickly understand whether the user will find the desired function or information.
')
The second question is testing expectations. With it, you can test not only the information architecture - whether the information is organized and the links are named, but also the expectations from the system behavior. For example, a user can tell you that when you click on the “buy” button, you expected a transition to checkout, and you designed a pop-up window with a notification that the item is in the basket.
The third question helps to test not only the interactive interface, i.e. buttons and links. It helps to understand how clear the information on the layout is. The question can be reformulated: “What do you think they want to show the user with this phrase / schedule / picture”?
The answer may surprise you greatly and you will understand that the idea that was originally laid down is understood by people in a completely different way.
A small addition is a five-second mock test. You show the person the layout for 5 seconds, then remove it and ask what he saw or remembered in the layout.
Perhaps this is not the best approach when testing interfaces, but it works well when testing promotional materials, landing pages and banners. For example, you can test the effectiveness of banner spots on the portal or the content of the banners themselves, asking after 5 seconds: “What was the banner on the layout about?”.
You can approach people in the corridor with a laptop or tablet, or you can with a layout printed on a paper. On the one hand, on the screen, the layout is more familiar to the user, on the other hand, on the paper layout, the testee will be able to draw something or depict his vision. My choice is both. If possible, come up with a tablet, but keep the printed version ready and, of course, do not forget to take a pencil with you.
Corridor testing helps get fast feedback on your sketches and move on. On the other hand, such feedback may not be very representative, because:
But corridor testing and is not designed for testing large scenarios, and also does not cancel the conduct of full usability testing.
With all the drawbacks, I would still recommend using this technique. It will not help you to make the product perfect, but it will make you think about possible design problems, when the cost of correcting errors is still very low.
It is clear that user testing helps a lot when designing interfaces and whole products, but in order to test something, you need a finished product or at least a prototype.
Creating a prototype makes testing much cheaper, but it also takes a lot of time and effort. Yes, and before creating a prototype, there are often variants of the interface concept, which would be good to test, too, without spending more than half an hour. In this case, the so-called “corridor testing” can help the designer. On Habré this topic was not paid much attention, so I decided to share a few tricks from my experience.
What is the meaning of corridor testing
You draw the layout on paper or in your favorite program, exit your office, come up to your colleagues in the office and show them the layout in printed form, or on a laptop / tablet. If according to your layouts you can test the entire script - great, link them to some kind of invisionapp.com and get an interactive prototype.
Testing is called corridor because it allows you to get feedback very quickly on your layout anywhere, literally in the hallway from a passing man.
How to conduct corridor testing
Creating a model, you obviously understood in which scenario it is used, therefore, demonstrating a prototype to a person, you can ask at least three types of questions:
- Where would you click to do / find something
- What do you think will happen when you click / click on this link or button
- How do you understand this phrase / title?
Where would you click to do / find something
This question is asked when testing user expectations from the interface. Your users must have tasks that must be completed with your product. Thanks to this question, you can quickly understand whether the user will find the desired function or information.
')
What do you think will happen when you click / click on this link
The second question is testing expectations. With it, you can test not only the information architecture - whether the information is organized and the links are named, but also the expectations from the system behavior. For example, a user can tell you that when you click on the “buy” button, you expected a transition to checkout, and you designed a pop-up window with a notification that the item is in the basket.
How do you understand this phrase / title
The third question helps to test not only the interactive interface, i.e. buttons and links. It helps to understand how clear the information on the layout is. The question can be reformulated: “What do you think they want to show the user with this phrase / schedule / picture”?
The answer may surprise you greatly and you will understand that the idea that was originally laid down is understood by people in a completely different way.
5 second test
A small addition is a five-second mock test. You show the person the layout for 5 seconds, then remove it and ask what he saw or remembered in the layout.
Perhaps this is not the best approach when testing interfaces, but it works well when testing promotional materials, landing pages and banners. For example, you can test the effectiveness of banner spots on the portal or the content of the banners themselves, asking after 5 seconds: “What was the banner on the layout about?”.
By the way, all of the above questions can be asked during the classical usability testing.
Laptop or paper
You can approach people in the corridor with a laptop or tablet, or you can with a layout printed on a paper. On the one hand, on the screen, the layout is more familiar to the user, on the other hand, on the paper layout, the testee will be able to draw something or depict his vision. My choice is both. If possible, come up with a tablet, but keep the printed version ready and, of course, do not forget to take a pencil with you.
Pros and cons of corridor testing
Corridor testing helps get fast feedback on your sketches and move on. On the other hand, such feedback may not be very representative, because:
- you asked not real users, but those who caught your eye
- people were not in the conditions where they usually use a computer
- people did not have enough time to get into trouble.
But corridor testing and is not designed for testing large scenarios, and also does not cancel the conduct of full usability testing.
With all the drawbacks, I would still recommend using this technique. It will not help you to make the product perfect, but it will make you think about possible design problems, when the cost of correcting errors is still very low.
Source: https://habr.com/ru/post/219699/
All Articles