In my last column I wrote about "evil users" and how they are out to destroy you. I explained that protecting your web site is important because not all people are out there thinking good things about you or your site. In my newsletter, when I announced the new column, I served up sentences like this:
I received a few responses about the article. People were concerned about me, and about Webword.com. They also questioned my commitment to usability and human factors. When I started thinking about it, they were right. I created an environment that did not seem to relate to usability. And, I wrote in a negative tone when I usually write with a much more positive overtone. In my newsletter, I painted a very dark picture of users. This is not my standard operating procedure.
I failed to consider how my actions and words would impact my readers. I also neglected to realize that my site is getting more and more traffic and that I've made fewer and fewer changes to the structure. Users can come to my site and find things easily if they have been here before (and hopefully, even if they haven't). Change is bad, especially if your site gets a lot of traffic and if your users are accustomed to it.
Think of it as an investment. The more time you spend at a site, the easier it is to use. Your investment in that site pays higher and higher dividends. Here's another metaphor. Your site and your users form a relationship. It might be a one night stand, a date, a partnering, an engagement, or a marriage. Perhaps a search engine or a link or an email set up the relationship. The point is that as time goes on, the commitment grows (on both sides). More and more is expected, and generally, less and less change is desired. Stability plays a huge role in the relationship.
When people become familiar with a web site, it is very dangerous to make changes. Similarly, if you change your writing style, your graphics, or your navigation, expect your users to be frustrated and confused. I recently heard a great story about Dell's web site. The designers decided to make some changes to the site for one reason or another. When they made the changes, traffic dropped and sales dropped. Dell thought they understood their error: The new design was bad! So, they made the change back to the old design. But, traffic dropped again and sales dropped too. Dell learned a valuable lesson. Do not make changes unless you must do so.
But there is more to learn from my last column and the Dell anecdote. You must be worried about changes, that is certain. But, perhaps more importantly, you need to help people understand that changes are being made. You need a crutch, a ramp, or some other a mechanism to ease the pain of the transition. And, you might want to consider how you manage user expectations.
Here's some advice. Not every piece of advice will apply to your situation, so pull from the list depending on your situation:
1) Perhaps the best tool for managing change is testing with an appropriate sample. Know your user base, draw a fair sample from that user base and conduct testing. Be sure to conduct your testing on a non-production server (i.e., use a test box).
2) If you don't have one, create an easy to use feedback form. Be sure that you place the feedback form in an appropriate position. Keep it out of the way of user tasks, but make it prominent enough to gain some attention.
3) Announce the change(s) in your newsletter. Also, feel free to ask for advice and solicit feedback from these trusted users. Drive them to the (well designed and easy to use) feedback form. Reward these users; their data and feedback is priceless.
4) Be prepared for data analysis. Dell knew to stop making changes because they had data to examine. They watched their log files and saw that changes were having a negative impact. The bottom line is to establish a baseline or you won't know if things are getting better or worse.
5) Be sure to be fair in your changes. In some cases, to see the impact of changes, you might need to wait to see how users will react. You can't always assume that your change is causally related to traffic and user usage patterns (although sometimes you can). Be sure to understand the appropriate time frame of your testing. The impact of your change might require weeks of activity to surface.
6) Decide on whether you want to stage your changes or if you should make them all at once. In general, I advocate making a sudden change to get over all of the pain as soon as possible. However, this can change based on the complexity of the site.
7) Decide when you will make the change. In my experience, if you are worried about poor results make the change on the weekend. But, if you expect success (due to trials with a sample of real users) a weekday will work better. It all depends on your situation. Just remember to think about the day of the week and the time before you make your change.
It is never possible to totally control a user's experience. So, the best user management weapon is information. Give a user enough information to manage their experience. Help them make their own judgments. Help them help you through feedback and testing.
It is entirely possible that the more you try to manipulate a user, the harder they will rally against you. The harder you try to twist their perception, the more they will make negative comments. The power of control is based with users, not you and your web site. Even the biggest giants must bow to the customer.
There are very few sites that can actually influence the whole of the web. Only the biggest and most traveled sites have the strength to mold Internet wide usage stereotypes. Amazon.com is often cited as the best e-commerce (business to consumer) web site. They have indeed shaped the way people view web shopping. As evidence, think about how other web sites have copied the Amazon look-and-feel.
Does this mean that management of user expectations requires developers to copy Amazon's design? No, but you ought to check out how Amazon does business. It is very likely that your users are familiar with the Amazon model of customer interaction. While you are at it, see how eBay, Dell, and Yahoo do things too. Think about how users come to expect email updates, affiliate programs, site navigation, and site searching. The big, powerful sites do have influence. Use this influence to your advantage.
My final bit of advice is to fall back on honesty when all else fails. Whenever you fail, and whenever you fall short of user expectations, tell users the truth. It is rare that telling the truth will harm you. Sincerity and integrity carry a lot of weight.
Search through over 300,000 Unclaimed Domain Names -- CLICK HERE!
Home | Moving WebWord | Cool Books |
Hot Web Sites | Reports
Newsletter Archive | Weblog Archive | Services | Interviews | About WebWord
to the Webword.com Newsletter
Contact John S. Rhodes, the WebWord.com Editor and Webmaster
URL: http://www.WebWord.com/moving/expectations.html© 1999 by John S. Rhodes. All rights reserved.