If I can do one thing different from now on for my product, that will be user test. I regard it as the most direct way to
There are many different types of tests, mainly categorized by the type of question we want to find out as well as the material we have that support these studies.
Christian Rohrer, on her 2014 article "When to use which user-experience research", divided all researched based on 2 dimensions:
Behavioral tests answer the question on "How", how does a user find the a product; how they do a checkout; how they fill in a form; while attitudinal tests focus on "Why", what are the reason you will or will not use a product etc.
Qualitative methods are much better suited for answering questions about how to fix a problem, whereas quantitative methods do a much better job answering how many and how much types of questions. Having such numbers helps prioritize resources, for example to focus on issues with the biggest impact.
Adding on top of that is the "Context of Product" which indicates how the product is used on the study, or in my case, the stage where the product is in: in conception stage, we will be doing interviews to understand how people think about our product idea and what they consider to be most important; while in late stage, we use usability tests on scripted cases to see if a feature truly deliver their expectation.
To make it more complicated, there is also the differences between face to face and remote rest. But for me it is merely the form.
There are many confusion on what kind of user text is used and why it is the best for the problem. Steve Kelvin talked about the differences between focus group and usability test, that we often use the first one to test if the flows and experience we design is good for the user. He emphasized the differences between the first being about asking questions and the second being completing tasks.
Another common mis-understanding can be found between focus group and desirability study, while both aim to understand if a product or service is attractive the target audience, while the first one has no product base and let the subject's mind run wild, the second present a product or service to the audience and is thus more controlled and reflect the actual expected result.
Google Sprint discussed the "Quick and Dirty" way of doing user test with a 5-day sprint, which talks very much on what i plan to start for my new projects, to test the idea. For the current project I am working on, I need to design a shopping assistant that provides services while users shop online. We’ve identified price comparison, last leg shipping enhancement and other features that can help solve the problem.
I am starting to write the survey question and recruit users at the moment. Craigslist is a popular platform and I am looking to other public forums to reach wider audience.
My first text is about 2 things:
The first is the way users will interact with the shopping assistant, do they find it is natural in a conversational way which is one question and one answer at a time in an order I have set for them; or they will like more a menu list and they can choose which question to answer and which to skip for the default.
I’ve prepared two sets of prototypes for on the same user flow a user will have to buy a product and plan to test them out on the audiences. They are both interactive so users can simply click through and explore by themselves.
The second is how users will inquire the different kind of information, would they prefer a traditional e-commerce experience with all requests, transaction history listed out and they will scroll up and down to find it; or will they like to always go to their assistant. The assistant will give some search recommendation based on the users requests and shipment status or show the full list.
I am going to write the interview script for next week. And aiming to have my first group of testers in the beginning of Jan. I’m very excited for this and to share with you what i found from this test.
- Make sure we come out a product the market wants
- Resolve many design discussion from the start
- Identify the right MVP that create value
- Give guidance to marketing
- ... ...
There are many different types of tests, mainly categorized by the type of question we want to find out as well as the material we have that support these studies.
Christian Rohrer, on her 2014 article "When to use which user-experience research", divided all researched based on 2 dimensions:
- Behavioral V.s. Attitudinal
- Quantitative V.s. Qualitative
Behavioral tests answer the question on "How", how does a user find the a product; how they do a checkout; how they fill in a form; while attitudinal tests focus on "Why", what are the reason you will or will not use a product etc.
Qualitative methods are much better suited for answering questions about how to fix a problem, whereas quantitative methods do a much better job answering how many and how much types of questions. Having such numbers helps prioritize resources, for example to focus on issues with the biggest impact.
Adding on top of that is the "Context of Product" which indicates how the product is used on the study, or in my case, the stage where the product is in: in conception stage, we will be doing interviews to understand how people think about our product idea and what they consider to be most important; while in late stage, we use usability tests on scripted cases to see if a feature truly deliver their expectation.
To make it more complicated, there is also the differences between face to face and remote rest. But for me it is merely the form.
There are many confusion on what kind of user text is used and why it is the best for the problem. Steve Kelvin talked about the differences between focus group and usability test, that we often use the first one to test if the flows and experience we design is good for the user. He emphasized the differences between the first being about asking questions and the second being completing tasks.
Another common mis-understanding can be found between focus group and desirability study, while both aim to understand if a product or service is attractive the target audience, while the first one has no product base and let the subject's mind run wild, the second present a product or service to the audience and is thus more controlled and reflect the actual expected result.
Google Sprint discussed the "Quick and Dirty" way of doing user test with a 5-day sprint, which talks very much on what i plan to start for my new projects, to test the idea. For the current project I am working on, I need to design a shopping assistant that provides services while users shop online. We’ve identified price comparison, last leg shipping enhancement and other features that can help solve the problem.
I am starting to write the survey question and recruit users at the moment. Craigslist is a popular platform and I am looking to other public forums to reach wider audience.
My first text is about 2 things:
The first is the way users will interact with the shopping assistant, do they find it is natural in a conversational way which is one question and one answer at a time in an order I have set for them; or they will like more a menu list and they can choose which question to answer and which to skip for the default.
I’ve prepared two sets of prototypes for on the same user flow a user will have to buy a product and plan to test them out on the audiences. They are both interactive so users can simply click through and explore by themselves.
The second is how users will inquire the different kind of information, would they prefer a traditional e-commerce experience with all requests, transaction history listed out and they will scroll up and down to find it; or will they like to always go to their assistant. The assistant will give some search recommendation based on the users requests and shipment status or show the full list.
I am going to write the interview script for next week. And aiming to have my first group of testers in the beginning of Jan. I’m very excited for this and to share with you what i found from this test.
Comments
Post a Comment