useit.com Alertbox Aug. 1997 Loyalty Negative impact of affiliates programs

Too Much Loyalty and Quality May Harm You

A reader who preferred to be known under the pseudonym Spam B. Gone sent me the following essay with scenarios that might cause smaller websites to play down quality. Mr. Gone even points out cases where user loyalty is detrimental to the business model of certain sites.

The essay makes several good points, but I hope that it does not turn out to be predictive of most smaller sites in the future. I hope that micropayments and other customer-supportive business models will make good content and good customer service more attractive than con games.

The essay clearly shows the importance of making the Internet more supportive of want of transferring value directly from users to those sites they like, enjoy, and find useful. Current models (advertising and affiliates programs) are too indirect and make non-users the true paying customers. At the end of the day, those who pay are the ones who shape the direction of an online service: the design will gravitate to satisfying the payers' needs and not the users' needs. Thus, to get useful sites, we ultimately have to make the users into the paying customers.
-- Jakob

"Affiliate programs will kill the Internet, Film at 11"
or
"Stickiness is Bad, but Bad Sites are Good"

by Spam B. Gone, November 1998

Web design gurus say that sites should strive for "stickiness". Sites should offer as many services as possible, with extensive internal navigation links to keep visitors on-site, and fresh daily content to keep them coming back for more.

That makes sense if sites make money per impression: you want to serve as many banner ads as possible. But most of us [small-time webmasters in affiliate programs] get paid per sale. We don't make money until visitors leave. 100% stickiness = 0% revenue.

The more attractive your site, the less inclined a visitor is to leave. They stay and run up your bandwidth bill. Worse, each surfer has a limit to how much "exploring" they will do before they get their daily Internet "fix" and log off and do something else. The more time they spend at your site, and the more satisfied they are with it, the less they will explore an affiliate sponsor's online catalog, which they might find less satisfying by comparison.

Stickiness is good if you're getting paid per impression. It's also good if you're an online store, or if you have a company website that supports a brand. But for the rest of us, stickiness is bad.

"If you want a picture of the future, imagine a boot stamping on a human face, forever."
-- George Orwell, 1984

Well, OK, it won't be _that_ bad. But here is the future: web sites with minimal content, seldom updated but stuffed with search-engine keywords and text links to affiliate programs (text links, because no one clicks on banners anymore). They will be just adequate enough to prevent an instinctive click on the Back button, but the surfer will be left "wanting more".

In this condition, they will be more likely to click on a (carefully-targeted) text link and more likely to spend time exploring the affiliate sponsor's site, which will seem like a delightful oasis after the "desert" they have just come from.

Meanwhile, high-maintenance labor-of-love sites with plenty of multimedia (high-bandwidth) content will wither and die. It is entirely self-defeating if you outshine an affiliate sponsor's website... the visitor will just click the Back button and come back to you.

Unfortunately, the above is based on actual clickthru and revenue figures rather than just idle hypothesizing. I will try to create plenty of the former type of site and none of the latter.

In 1984, "War is Peace, Freedom is Slavery, Ignorance is Strength". In 1999, with apologies to George Orwell, "Bad Sites are Good".

Adult Sites have Been There and Done That

The "adult" web is apparently already well advanced in this evolution. "Free site" owners who toned down or outright eliminated the graphic content from their pages due to concern over obscenity laws found to their surprise that their revenue went way up, and they never looked back. Such sites do not rely on or even want repeat visitors -- "bookmarkers" is used as a term of scorn, synonymous with "bandwidth-sucking locusts". Traffic comes from search engines, "type-ins" of judiciously chosen domain names, and other sources. Revenue is earned by sending that traffic straight back out the door to "pay site" sponsors with rich content that requires subscription by credit card.

Needless to say, pay-per-impression banner advertising doesn't exist on adult sites -- it's mostly commissions per sale, or productivity-based clickthru that amounts to the same thing (sales are tracked and a sliding scale pays more per clickthru for more productive referrals). There are some "raw" clickthru programs, but even those often have a provision for kicking out affiliates who send unproductive traffic, and in any case they turn a profit on "raw" clickthrus by reselling their exit traffic at a markup to other adult sites.

In the mainstream non-adult web, it seems increasingly clear that per-impression advertising will never support the vast majority of sites out there. As affiliate programs spring up like mushrooms and increasingly emerge as the most viable revenue source, it seems that the "ugly site" model will be equally lucrative on the mainstream web.

Get Them to Explore if You Want More ($$, that is)

Affiliate program sponsors would do well to emulate some of Amazon.com's features that are designed to draw surfers in to explore deeper. For instance, the "people who bought this book also bought..." feature.

My experience with Amazon has been the following: visitors hardly ever click on banners or buttons or even text links that go to Amazon's home page. They are much, much more likely to click on a text link to a specific book that closely fits in with your site's theme. But then, paradoxically, they rarely buy that book! Instead, most commission revenue comes from purchases of other books or CDs, often entirely unrelated.

It's important to provide text links to books that closely match your site's theme... surfers won't click on unrelated book links (unless the word "sex" is in the title :-). But ironically, it seems this is merely the "bait" that lures the surfer to the store. They don't buy the books you think they ought to buy. Instead, the "supermarket" effect takes over and they make impulse purchases.

The above anecdotal evidence only seems to emphasize the importance of creating an "oasis" effect. Since surfers rarely buy the merchandise that they clicked-thru on, we don't get a commission unless they get a favorable enough first impression of the affiliate sponsor site, so that they will be enticed to explore further and buy something. And since the design of the affiliate sponsor website is beyond our control as affiliates, the only thing we can do to influence this is to create ugly sites that will contrast and set off the affiliate sponsor's site in the best possible light :-)