Welcome to WebmasterWorld Guest from 126.96.36.199
Forum Moderators: buckworks
I have a testbed site that I use for experimenting, and over a 3 week period I got this:
week 1, no underlining: sales approx £300
week 2, underlining: sales approx £35
week 3, no underlining: sales approx £210
While this hardly constitutes a proper study, it's food for thought. Does the average consumer know or care about link underlining? Do e-com surfers react differently from general surfers?
joined:Sept 1, 2000
Maybe there are other factors involved such as by making it (arguably) less obvious where links to other pages are the customer is driven more to the obvious link (that says buy now?).
Interesting stuff, is there any way to see stats for how often each hyperlink is clicked in each week?
Looks that way, on this site at least. Your point about pushing them towards the 'buy now' link fits in with my rapidly developing theory of reverse usability ;)
I think there is a good case to be made that you should try to grab the visitor immediately they hit the site and push them towards the sale, giving them little opportunity to get distracted by exploring the rest of the site.
As to the clicking data, I'll try and get an analysis done. Something like the ratio of internal to external referrals could be illuminating - thanks for the idea :)
However, if it increases sales that would make it good. Maybe it has it's place. But personally, I don't like it.
Edited by: Vacor
I don think i've ever done a site with underlined links ;)
I can understand what people are saying but as long as the links are used in well constructured pages the user will not need to think 'is that a link?' I have been to sites where the page is clutered and it is hard to tell which links are really links cos they highlight other words.
But after a very short moment, I had no problem with the site. My mother, however, that would be another story.
I think underlining tends to be a bit passe especially on corporate sites not that they should never be used just that they are not always required.
Personnaly I want to see the web progressing not looking the same as it did in 1984 ;)
The first few times I visited WMW I tried looking for the search feature and it was not at all obvious, now it would'nt have been any easier or harder to find had 'search' been underlined. But what if it was bright pink, was 4 times bigger than the surrounding text? You would have seen it straight away.
So all i am trying to say is that a succesful link is one that stands out above normal text - one that draws your eye to it. Underlines are very good visual clues - but not everyone thinks underline=link.
If i have a big page of text then i will use underlines, all I want ppl to realise is that as long as you give some visual clue to hyperlinks people will know to click on it. People click on buttons if they look like buttons, people click on links if they look like links regardless of what they look like.
Its all about giving people clues, enough clues to discern what is clickable and what is not.
Designing sites to impress other designers and art directors is a throwback to the print medium. Web sites are not static presentations, they are interfaces that visitors need to understand and USE instantly.
There are very few people that can't see the difference between black and blue (most color blind is blue/green).
The familiar blue color is a dead giveaway that the word is a hyperlink. If that isnt enough, the hover underlines in blue and highlights the link in yellow. No studies, but no complaints either. Personally, I think it makes the page neater(as in clean or sharp looking).
If you really wanted to do a study, do session based cloaking. When a user visits the site, randomly mark them with a cookie...underline or no underline. Then track the number of pages that user views as well as the purchase amount. Both values are important. It definitely beats having a time (week to week) variable thrown in there as well.
I agree with that, some sites "die" on legal holidays. If you throw in 4th of July week to the mix, you could get very faulty results. Everyone was off of work for the 4th legally and most had a "sick" day on the 5th. Overall, a bad week to include in any comparisons.