 TUTORIAL----Being Tops with Your Users and the Search Engines |
    |
Being Tops with Your Users and the Search Engines
By This Method
Playing Detective
There are three times in the life of a website when user feedback is necessary:
1. Before the design or redesign. At this stage, it's important to figure out who your target audience or ideal client is so that you can meet the objectives of the site. Include marketing data, and ask potential users what they want, how they want it and why.
2. During the design. First test the website on your development server with sample test users. You can include the development team but since they know how it's intended to work, they're not necessarily the most objective testers. Be sure to run a beta test on your "staging site" (a mirror of the final site) with other, more objective testers. This will allow you to work out the bugs, create enhancements and perform navigational adjustments *before* your live launch.
3. After the launch. When the website is in full production, included in the search engines and humming along nicely, it's time to start digging into your server logs and stats to track your users' interaction with your site. (Pages visited, how long on each page, paths throughout the site, etc.) You can learn a lot about your visitors this way.
All websites will benefit from this before/during/after approach and attention to detail, but few actually take the time to do it. Those that do are generally many steps ahead of their competitors.
Detective Tools
If you simply want to know how your website is doing, feedback is easy to gather. However, if you built a website where you need statistical data such as ROI (return on investment), customer satisfaction results, and competitive analysis, things will be a bit trickier.
Here is a variety of different methods for collecting user feedback (some are more easy to implement than others):
1. Contact email address - include this in your footer and/or contact page.
2. Contact feedback form - do not *require* personal information here. Leave lots of room for comments, and be sure to ask questions. Offer incentives to encourage user feedback.
3. Polls - there are all kinds to choose from (look for third-party polling software).
4. Popup forms - however, many users don't like these or have them disabled. Consider your target audience and any risks before deploying this option.
5. Exit polls - for instance, a poll that shows up after a person checks out of your shopping cart.
6. Ratings - try adding a "rate this page" function.
7. Ask for help from Listservs/Forums - there are many forums that allow you to ask for a website critique and will help you with any trouble spots.
8. Email surveys - survey a sample of your users through email (only those who've given you permission to do so). You can design longer surveys for user testing this way.
9. Web-based surveys - there are many pre-built surveys you can add to your site. Some are designed for generic feedback while others collect real-time data for analysis.
10. Focus groups - there are marketing firms you can outsource this to if it's within your budget.
11. Labs - hire usability experts to watch people using your website and record their movements via videotape.
12. Friends - if you don't have the budget for a usability expert, you can copy the lab scenario by watching friends use your site.
13. Clients - don't wait until the final rollout to show the site to your client. Allow the client's staff to beta-test it using their own operating systems and browsers.
14. Usability testing - more on this in Part 5 of this series.
How To Ask Questions
It's important to note that the way a question is asked and presented can bias the results of any user feedback. There are user-testing books to help you understand how to set up a neutral testing environment, and white papers or usability studies that teach the design of questionnaires, where to place web-based surveys and how many people to sample for meaningful results.
Types and Examples of Questions
When creating questions for your users, you'll want to categorize them into the following types:
" New ideas
" Criticisms
" Help/FAQ
" Navigation
" Performance
" Credibility/popularity
" Functionality
" Ease of use
" User interface
" Search engines
There are many ways to ask your questions, and of course, many questions to ask. Your survey format can include "yes or no" questions, five-point scale satisfaction, verbatim answers (written comments), checkbox selections or simply a question written on a piece of paper and handed to someone (usually in the form of a task you would like them to perform). Tasks are the most telling of all because watching a user shows what works and what doesn't.
Here are some example questions that will help you uncover the most common mistakes:
" Were you able to locate the login button?
" Were you able to locate the shopping cart icon?
" At what point, if any, did you get lost on this website?
" How satisfied were you with the navigation of this website?
" Please rate your experience with locating product specials and/or sales.
" Was the objective of this website clear?
" How fast did this page load? What connection speed are you using? (DSL, cable, modem)
" Were the tasks you performed self-explanatory?
" How satisfied are you with our link descriptions?
" How satisfied are you with the product images?
Why Bother?
A successful website is one where the visitors' needs are being met. User surveys and feedback are diagnostic tools to help identify problems before, during and after the site is built. It's like designing the food take-out speaker for a drive-thru restaurant and finding the speaker height is wrong for sports cars. Far better to find out before a billion speakers are built and sent to the restaurants! If enough users are surveyed you can understand the magnitude of any problems or trouble spots before it's too late, and then determine whether it's financially justifiable to make improvements.
This post has been edited 1 time(s), it was last edited by kaplish on 10-03-2003 at 12:12.
|
