Usability testing (software self-help)

Often we find in product research that research is difficult to address the nature of product issues, does not work on product improvement, and http://www.aliyun.com/zixun/aggregation/8135.html "> Usability testing approach to the emergence of these problems.

Why do we have to do usability testing? In our long-term research, we find that the user's actual operation is different from what he says, and the subjective questionnaire has no way of capturing the user's behavior. There is a very large deviation between the suggestions and the real situation of the users. Real usability testing is the process of allowing real users to operate, which is based on the evaluation of user performance, which is one of the important methods of user-centered design.

We can at least use usability testing to achieve the following:

n Check that the design is valid

N Diagnosing problems

N Add new Features

n A/b test two choices

n Verify that the design objectives are met

In usability metrics, we need to focus on the following issues:

N are users able to perform tasks?

n Do you expect the user to see the information that the user is discovering?

N did the user find a valid path?

n Do users know what they are doing?

n What problems have they encountered?

N Professional Terminology

n don't know where to go or what the next step is

N did not find what they were looking for

Can n recover from an error?

General usability testing standards are as follows:

n Set your usability criteria before setting the list of issues

n Behavior test criteria (for example, air ticketing)

Users can change the address within two minutes

Users can enter the home page only through one click

Users are able to purchase tickets without the wrong success

4/6 of users can buy tickets online

N Behavioral Performance criteria

Time to buy Tickets:

Good:under 2 minutes

Ok:2-3 minutes

Problem:3-4 minutes

Failure:over 4 minutes

N Basic Preference criteria

Under the easy to use standard, the user will give at least 5 points (7 points full Mark System)

More than 80% of users prefer the new design to the old design

In general, filter users:

N User selection needs to be based on the user profile or user description generated by the market data (quantitative data) concentration

N Select participants based on:

Use of the website:

Current user

Potential users

Product Use Level

Other factors are important.

Language

Education level

Gender

Geographical

Main culture

N 5 users will find approximately 80% of product availability issues (virzi,1992 and nielsen,1993)

N Faulkner (2003) found:

Test 5 users about 85% availability problems

10 participants found 82% to 85% availability issues

N general rule: using 5-8 people as measured users

Usability Test Cases:

Usability Labs:

Usability Test Discovery:

Source: http://www.chinapmd.com/?p=135