How Do People Evaluate a Web Site's Credibility?

theduck writes "Ever suspected (or feared) that web users are mostly mindless sheep evaluating your website more by the eye candy than your carefully crafted content? Well, it appears you were right. A study resulting from a collaboration between Consumer Webwatch and The Stanford Pervasive technology Lab reports that even though consumers say that they look for content first when evaluating the credibility of a website, they actually focus primarily on design look and information design/structure (i.e. ease of navigation). Of course, the study's methodology might have something to do with the results..."

Comments

Not that a good design is a bad thing.

[novakreo]

What good is the best content in the world if it's difficult to navigate your way through it?

In real life communication people are able to get much non-verbal information from the speaker, giving hints as to whether they are passionate about their topic of conversation, or even whether they really believe it or not.

While ultimately the content itself is paramount, having a well-designed site will show visitors that you at least care enough about it to put some effort it.

Easy

[Compact Dick]

They view the source to make sure only valid XHTML and CSS are used. Those that do get top scores :-)

Re:Easy

[dtobias]

A few of my "turn-offs" that lower my estimate of the cluefulness (and hence credibility) of the site's developers:

• hardcoded-pixel-width tables instead of sites that gracefully adapt to any window width
• microscopic fonts for main body text
• pop-up ads (though since I've set Mozilla to reject them I don't actually notice which sites have them any more)
• notices telling me to "Get a Better Browser" or change my resolution, etc., to whatever their clueless developer prefers
• pointless splash pages before you get to the real content
• main site content or navigation that's nonfunctional without some feature not present or disableable in some browsers (JavaScript, Flash, PDF, sites opening pop-up windows to put all their main content in -- the latter often show as blank pages to me due to my Mozilla configuration to disable popups)
• stupid or careless title text for pages, like "Untitled Document" (it's amazing how many pages out there have this as their title), or just mindlessly replicating the site name alone as the title of all pages, or trying to spamdex keywords for search engines -- these things result in a useless and stupid-looking set of titles in the browser back-button history and when you bookmark a page in the site
• images with no ALT text
• images with stupid ALT text, showing a lack of understanding of the purpose of this attribute
• frames (almost any use of them)
• any use of A HREF="#" as a link; the "#" pseudo-URL seems to somehow be trendy as a dummy link element when developers are really using the anchor tag to hold JavaScript events; they should instead put a meaningful URL there for graceful-degrading purposes. At any rate, the JavaScript events should always end in "return false;" so that the URL isn't actually gone to; otherwise you get an unsightly jump to the top of the page and add another URL to the back-button history.
• any use of ".htm" as the extension for HTML documents instead of the more proper ".html" -- that's a dumbass Microsoftism dating from ancient and obsolete operating systems that couldn't take four-letter extensions. It's HyperText Markup *Language*, dammit.
• links back to the site's homepage, or to subdirectory default indices, using A HREF="index.html" or A HREF="subdir/index.html" (or other default name) instead of the more elegant and consistent use of the directory alone like A HREF="./" or A HREF="subdir/".
• links to subdirectories that omit the trailing slash, forcing an extra server redirect (NOTE: Both this and the preceding item cause links to show in the unvisited color even when you've already been there, due to the URL variation.)
• sites that muck around with the link colors so as to make visited and unvisited links the same color, thus suppressing a useful feature of the browser's interface
• sites that try to disable parts of the browser's controls, like the back button or the "Save As" feature, perhaps to "stop people from stealing their content".
• domain name abuse, like sites that are clearly noncommercial in nature but insist on using a .com address, or sites that are chapters or branches of parent entities but use stupid unnecessary domain names of their own instead of logical subdomains of the parent site, or sites that keep registering ridiculous flavor-of-the-month domain names for every last marketing gimmick they come up with, when it all could have been done perfectly well through the domain they already have.
• sites that have their own domain name but for some inscrutable reason use something stupid that's not in their own domain as their contact email address -- the use of (gag, vomit) aol.com addresses in this manner is somehow rampant.

Depends on the user

[Smidge204]

...and what they're looking for. Obviously a site that is impossible to navigate is't going to be very popular, but the savvy user who knows what he/she is looking for will certaintly be more focused on content than presentation.

Of course, the 'internet surfer', who is typically not looking for anything specific, is more likely to be captivated by 'shiney things'. Given the nature of the study's methods, I'm thinking that was the case.

Since there was no guarantee that the person in question had any interest whatsoever in what they were showed, how could they honestly judge the page based on *content*?
=Smidge=

Mindless sheep

[jukal]

Of course, the study's methodology might have something to do with the results...

Yes, if you invite mindless sheep, it is quite likely that it is reflected in your results. In this case, it turns the whole study into a very good case of black humor :)) From here.

We began recruiting participants in May 2002 by contacting nonprofit groups, such as the Children's Brain Tumor Foundation, and offering a $5 donation for each supporter who completed the study

*** Why not read an analysis of the Slashdot Effect instead :))

I don't buy this

[WolfWithoutAClause]

The design/look of their consumer webwatch website is much too poor for me to ever buy their conclusion. ;-)

Very obvious

[tanveer1979]

Looks count. So if your webisite aint good to look at content can take a walk. This is a simple truth in this world. Suppose a merc looked Like Lada(that russian car you see in bond movies) I doubt if many people would pay the moolah.

Of course a pretty webpage is no excuse for bad content. Just looks will ensure that people visit your front page and dont really come back. For the invitation you need the eye candy, for the substance you need content. so you need balance.

However some webpages are all eyecandy and everything else is a pain. A proper balance is needed. A good example of balance is this site.

Then CNN is also an okay example though it gets a bit cluttered.

Sadly, most webdesigners use some invalid M$ only code in their website, and non windows ppl have a problem surfing. These guys realize that 95% of surfers wont mind the nonsense.
This is web democracy majority rules. Most sites which are coming up new are focusing all on eye candy.

You have other extremes too. Some people want to make it easy for lynx users, result, sites look good only in lynx! This is going too much over the board. IMHO a good site should have valid HTML, and simple valid javascript. Also a site map is a must. I really hate those sites with lots of clutter. Looking for info is like looking for a needle in a haystack. Perhaps if they put some good search engine on the front page. As far as that is concerned even /. is horrible. The search engine can take some serious work. Otherwise I just love it, its simple, has never crashed my browser and dosent take too much time to load.

I'm not interested in credibility

[Subcarrier]

...and neither are you. That's why we're both reading Slashdot.

Optional Flash

[evenprime]

Spoonist said: Sites that do not use Flash get my Seal of Approval

I don't have a problem with sites using flash liberally, so long as they provide a non-flash way to get to the same information on the web site. This is especially important for an index page...you ought to allow people into the site without them having to wait for your animation to download.

Google and Design.

[The Cydonian]

Here's an alternative thesis.

Is it possible to say that poorly designed websites reflect a certain apathy on the web designer's part? Surely (the argument goes), if designers didn't care enough about the design, they wouldn't have cared enough about the info they provide.

Now, that's a statement that wouldn't apply to me. I use Opera 6.05 to navigate, so I get more than my usual share of poor design (bad html, javascript, MS-proprietary tags etc). However, personally I don't care; my focus has always been on finding the info I want, for which I use the excellent the find-in-page and the google search buttons that Opera provides.

Bottomline: It's probably not poor navigation per se, but a bad impression on the viewer.

Not quite so obvious

[evenprime]

tanveer1979 said:

Looks count. So if your webisite aint good to look at content can take a walk. This is a simple truth in this world.

I don't believe that holds true for all users or all types of information. The more detailed the information you seek, the less concerned you are with the look of the site. Example: I've been thinking about building a guitar, and have literally spent hours reading articles at Frank Ford's site. That's one of the plainest web pages you could imagine, but the information there is pure gold.

Re:NO FLASH

[LordLucless]

Its all about what's appropriate.
I know some people who hate all web applets on principle. I know a lot of people who hate all flash on principle. But what the whole thing really boils down to is this:

Does it add anything to the site, without taking anything away?

Flash with no way to avoid it detracts from the site because people with slow connections are inconvienved. Fancy applets everywhere detract from the site because not all browsers can handle them.
But applets that add optional extra usability, or flash navigational elements with a HTML alternative are fine by me.

A programmer's perspective

[NBarnes]

Code that runs perfectly but is uncommented, undocumented, and mostly consists of one 3000 line file titled main.cpp with five methods == bad code

The distinction between information and data is critical. If Site A has the data I want, but Site B has it and has already intelligently decomposed it into information for me, then Site B wins. It's not even a distinction that only matters to non-power users; any thoughtful person will prefer to spend less time digesting data into information and more time applying that information in interesting ways. This is a dynamic that is seen in good coding practice, in (G)UI design, in web design, in short, in any sort of content presentation.

This is not to defend presentations that _obscure_ the information being presented, but rather to highlight the importance of _correct_ content presentation. Clarity of presentation is a creature of balance; neither too little nor too much. If Flash can make your information clearer, use it. If Flash obscures your information, ditch it.

Well duh!

[croftj]

Of course they judged the books first by their cover. We do it with books, with the people we meet, with stores at the mall and around town!

We need to do that. It helps us decide whether we want to pursure things further or move on to the next thing. Otherwise we would have to fully investigate everything.

Does it work 100%? No, but it works well enough that nobody is going to stop doing it.

Captain Obvious

[limekiller4]

michael writes:
"Ever suspected (or feared) that web users are mostly mindless sheep evaluating your website more by the eye candy than your carefully crafted content? Well, it appears you were right."

Thanks, michael.

In other news. water is still wet, the sky is still blue and yes, Barbara Streisand still sucks.

this smacks of focus-group thinking...

[mlas]

While this study produced some food for thought, I will ingest it with a liberal amount of salt. Whenever you ask people about their comments on a site they just visited, you will invariably get a fair amount of what they think you want to hear. This is why focus groups are suspect; it's an interesting truism that people are bloody poor at articulating what they really want.

Even with 10 sites per category in several categories to choose from, it's doubtful that any of those sites have content people really want at the given moment. Without providing content that will satisfy a perceived need, of course people are going to respond more positively to the shiny, candy-like buttons...

The only way to really understand what makes people's web site experiences a satisfying one is to observe people as they surf, and watch their results. Good web design is more like gardening than anything else: plant the site, watch traffic grow, remove some weeds, water with fresh content, fertilize with fresh design. A good strategy is to post alternate designs often, and then watch your traffic logs to see if clickthroughs increase. Amazon, whatever your criticism of them, is great at this... their site changes features and layouts just about weekly. Not enough to make people lost, but menus move, contract, expand; buttons move from GIF to HTML and back; layouts widen and narrow. Then the proof is in the traffic, and the visitor never gets a chance to tell you what they think you want to hear.

I'm glad to hear the conclusion that good design has value (I'm primarily a designer), but I think this survey is an oversimplification of reality which leads to an oversimplified conclusion.