Doc Searls has discovered a paradox
of Web use:
He used to
sell many books
from his site when he linked to
to do the fulfillment.
not sold a single book
since he changed his site and started to link to
an independent bookseller, Wordsworth,
to handle fulfillment.
Why would sales stop just because of the choice of a different fulfillment provider?
The answer is found in my Alertbox from November 1999:
Usability as barrier to entry
. Web users are getting so impatient that they demand instant gratification and
do not want to learn any new user interfaces
For the average user, Wordsworth is a new fulfillment channel, so they do not want to take the time to learn how to use it. Most people already know how to use Amazon.
In addition to learning a new UI, users would also have to overcome the pain of registering with a new site. Most people already have their credit card number and shipping info on file at Amazon.
The current Wordsworth design places additional barriers to having anybody buy because they violate several of the
usability guidelines for e-commerce
They call their shopping cart a "bag" instead of a "cart". A few users will stop right here.
When you "add to bag", you don't get a confirmation screen that shows the shopping cart and its contents - you just get the same page back, causing many users to think that nothing has happened (only confirmation is a subtle change to a marginal area on the screen that now says that "number of items" is 1). Most users will stop here because they think that the site doesn't work since (seemingly)
when they clicked add to bag. People don't have time to investigate a site that appears to be down.
There is no obvious checkout button (the word "checkout" does appear but as non-standard white text that does not look like a hypertext link). Many of the few remaining users will stop here. People don't have time to study how to use a site if it is not immediately clear how to do so.
As I have said many times,
non-standard interaction design
equals lost sales. (After I posted this critique, the site has fixed some of the problems.
The above list refers to the way the site worked the day I wrote my article.)
Note that all of these arguments apply to
Web users. New users would certainly be equally motivated to learn Amazon and Wordsworth, but it would be rare for new users to come across a specialized site like The Searls Group and use it for their first e-commerce purchase.
Branding is a Temporary Advantage
New users would still be more inclined to use Amazon instead of Wordsworth because of Amazon's
current advantage in branding
Users are very worried about being cheated: will I get my package, will it be the right product?
Users also worry about getting trapped in an endless marketing mill: will I get a million "valuable offers" in my inbox?
Better to use a well-known company.
In the future,
branding will be replaced by
as the way users decide where to do business on the Web. Whenever you go to a website, a flag will appear on your screen to indicate whether that company has treated previous customers well or poorly. Similar with banner ads (if any survive that long): all ads will be stamped with an independent rating advising you whether the offer is genuine and worth checking out. Bye, bye deceptive advertising.
Reputation managers will make users feel as safe doing business with a small, unknown company as with the largest corporation. In fact, since small companies usually offer better customer service, they will benefit the most from the shift from branding to reputation.
The usability barrier will also be lowered as more sites recognize the need for simplicity and user-centered design. Following the basic rules of Web usability will be another great equalizer.
Web Wallets Overcome Type-In Barrier
Having to register and type in loads of personal information is a great barrier to using new sites. As soon as we get a decent
that is a wide-spread standard, this problem will vanish.
Users would have "1.5-click shopping" enabled on every single site on the Internet: any time you see a product you like anywhere on the Web, you would click a button called something like
"buy with my WebWallet."
This would bring up a very simple dialog box to allow the user to authorize the charge. The wallet server would pay the vendor (and charge the user) and would provide a privacy-secure way to transfer the shipping address to the vendor's fulfillment house.
Traditional mailing list vendors have invented a great way to police the use of this shipping info: every now and then the wallet service would place a test order using a dummy address - if this address ever received
unsolicited spam, then the vendor would be dumped from the system for abusing private information. Not worth the risk if you value your sales.
Amazon May Be Over-Valued
The analysis in this column leads to two opposite conclusions:
In the short term, Amazon is unbeatable: other sites will have a very hard time taking business from Amazon unless they provide incredibly superior usability. Unfortunately for these other sites Amazon has always had great usability, so it will be very hard to surpass.
In the long term, Amazon is going to lose its current special status and will have to compete on an equal footing with other sites. Reputation managers will eliminate the marketing advantage Amazon enjoys from its brand. Web wallets will eliminate the usability advantage Amazon enjoys from one-click shopping.
I think Amazon will thrive even when it is forced to compete on an equal footing with other sites. Most websites violate basic usability rules and are much harder to use than Amazon. As long as Amazon retains the lead in usability, it can survive the loss of its two temporary advantages.
Even so, Amazon's stock valuation probably assumes that it has a stronger lead than it actually has. Even though
websites are clueless, some of the better ones are starting to simplify their designs with a focus on customer needs instead of glitz.
on this Alertbox (including a case study of a price comparison engine that closes 30% of its sales from Amazon even though they are usually more expensive).
For newer user research data on when you should — or should not — emulate Amazon.com, attend our full-day seminar
Big, Famous Sites: Which Design Patterns are Good Enough to Borrow?
at the annual
Usability Week conference