As the author of the WWW FAQ, I regularly answer questions about the workings of the Web. If a question is frequently asked, I simply add an article to the FAQ. But sometimes a question is more detailed, more in-depth— not really a FAQ, but still of interest to others. You'll find those questions, with my answers, here in Innards along with commentary on other web-technology-related topics.

2003-12-04

1. Assume a new web page, A.

2. Let B be the set of other pages from which the author of A links personally to A, plus any additional means by which A personally distributes the link to A, such as word of mouth, email, etc.

3. Let C be the set of actual people who see those new (to them) and therefore potentially interesting links in a week.

4. Let D be the portion of C that actually click on one of the links to A.

5. Let E be the portion of D who are actually find A interesting.

6. Let F be the portion of E who maintain blogs or web pages of their own.

7. Let G be the portion of F who also decide to link to A.

8. Let H be the set of actual people who see these links in a week...

Lather, rinse, repeat!

Now let's plug in some numbers:

I'll assume that B has remained roughly constant -- it depends only on the motivation of the author of A. C depends on the popularity of the resources containing the links in B; you can't really "register your web site with all the search engines" anymore, much less post it to NCSA What's New. There are still ways, google will find you eventually, and word of mouth certainly still works. But C is definitely smaller than it used to be.

D, of course, has shrunk phenomenally as we all get buried under a billion links in spam, in blogs, on major commercial sites, on television, et cetera. Most people do not "surf" the web for the thrill anymore. Assuming these links are in blogs, which are read somewhat for the fun of seeing what friends link to, we'll be charitable and assume D is 20% the size of C.

We'll assume E is fairly constant. I could whine that offering the same old thing doesn't turn people on anymore, but that's obvious, and I'll assume people will bother if something is really novel. But nobody shares all of your interests. Let's call E 25%.

F is a bit paradoxical. In the early days of the web, F was huge -- everyone had a home page, and everybody's home page was virtually and sometimes literally a copy of their browser program's bookmark list. F plummeted as search engines became effective and people got out of the habit of maintaining their own links to everything they found interesting. F popped up again in the last few years with the popularity of blogging, livejournal, etc. F might be as high as 20%, or even close to 100% if your friends are all bloggers and only friends read your blog. I'll split the difference and call it 60%.

G: the percentage of those with the habit of doing so who choose to link to this particular new page, given that they do find it interesting. There are lots of reasons people don't bother: you already linked to it, why mention it again and look like a Johnny-come-lately on the LiveJournal friends page. They linked to six other things today, that's enough for one day. Their blog is mostly about tropical fish, although they also enjoy your cat pictures. Call it 10%, tops.

H: it's easy to imagine that as more and more people, in total, see your web site, the number of people seeing it every day will also go up. But that frequently isn't true. D, E, F and G can all be thought of as percentages of C, the number of people that the author of A directly led to it the first time; D*E*F*G is likely to be a very small percentage. Let's say that every blog has a readership of about 12, as has been recently suggested. You post a link to something new in your blog; 12 people see it. Multiplying all of the above percentages together, we get 3.6%. 12*3.6% is .432. That means that, a little less than half the time, one other person will bother to link to your new page. If we multiply .432 by 12, we get an average of 5.184 (oh, call it five) people to your page in the "second wave." You can repeat the depressing arithmetic, if you like; 5.184*3.6% = .186 people linking to the page, therefore 2.24 people in the third wave, and so on. It peters out. It waves its little legs in the air, kicks feebly, and dies. Not to get all depressing and stuff.

So what can you do about it? Assuming no cash-money marketing budget, most of the variables are beyond your control. There are two exceptions: E and G.

E is obvious: be more brilliant! I'm sure you're already working on this. But G is a bit more interesting.

I can see several ways to increase the percentage of enthusiastic users who actually do link to your stuff, short of paying them:

• Ask them nicely. Done right, this doesn't have to be a disaster. Some people may assume you are already famous since you are obviously so brilliant. Try letting them know you could use a hand reaching your public.
• Make it really, really easy to tell a friend about the site by providing a "tell a friend about this page by email" facility. This adds people who don't make pages of their own to the set of those who can potentially link to you. Help get that flong started! Make very, very sure it is crystal clear to the recipient from the headers and a personal part of the message that somebody they really know really did forward it. Guaranteed to wind up in the spam file otherwise, with a nasty aftertaste.
• Provide nifty icons for linking to the page and make it clear users are allowed to use them for that purpose. Doesn't always work, but a surprising number of people think they need permission to link to things.

In other words, viral marketing. Because remember, if D*E*F*G is less than 1/12, you're SOL.

---