Traversing the "Googlearchy" 67
baloney farmer writes "How much do search engines influence the availability of information online? A new study gives some surprising results. Search engines help with popularity, but not as much as you'd think: 'Traffic increased far less than would be expected if search engines were enhancing popularity. It actually increased less than would be predicted if traffic were directly proportional to inbound links. In the end, it appears that each inbound link only increases traffic by a factor of 0.8. The results suggest that the reliance of web users on search engines is actually suppressing the impact of popularity.'"
Re:A factor of 0.8 decreases traffic (Score:4, Informative)
Instead they saw a straight line with a slope of 0.8, meaning the hits increase 0.8 units for every 1 unit increase in incoming links. More links still correlate with more traffic, but, for example, doubling the number of incoming links increases the traffic by a factor of 1.6, not by a factor of 2.
Re:Self-reenforcing cycle? (Score:5, Informative)
It has nothing to do with what the search engines do or provide per se. Search engines aren't always needed to a certain extent any more, particularly when it comes to popular sites, specific uris, etc. The reason (IMO)?
Word of "mouth". Actually, email messages[1] are sending names of services or specific uris for a particular site (e.g., something particularly funny on youtube) and people are pointing their browser in that direction, then exploring what else is there. If there are uris to other locations on the web, people follow those. One of the local affiliates in Indy played a considerably portion of this [wthr.com] last night and made sure everyone knew there was a link on their web site. Lots of people likely pointed their browsers and youtube had a lot of extra traffic[2]. On the youtube page is Explore other videos. Lots of information conveyed, but no search engine activity in the process.
The web has enough toys^w services which people regularly visit (e.g., blogs, youtube) they don't necessarily need search engines unless somethings isn't found via the normal means. And normal now includes the various discussion forums where people provide the advice from the voice of context. IMDb.com has a professional side (reasonably priced paid service) where people who are in the biz can post things they're looking for or are available for. A couple of nights ago, someone was asking about the best software for scriptwriting on a small budget. ca. eight people chimed in with what they knew about different packages, including a couple of free ones, a commercial one for $25, a template which can be downloaded for MS Word, and some of the pros & cons about the ones they'd used. Where will you find ad hoc information in that context on demand in a search engine?
__________________________________
[1] Unless you're in the media and use "emails" as a noun.
[2] Several years ago, I had a client who helped small to medium newspapers get online. Someone build a web site for them (taking six months, #include files nested six deep, every call to the server required 20'000 lines of code to be processed, regardless of the function involved. Once more than twenty people hit a site, the server showed you its impression of the La Brea tar pits. One site for a reasonably small city, perhaps a handful of a thousand people had a sheriff's deputy arrested for pedophilia, a ten-car pileup on the nearby interstate, and the largest employer (a substantial percentage of the citizenry) was going to be dismissed. All of this hit CNN with a reference to their newspaper's web site. That's about the time Chrnobyl and Three Mile Island happened at the same time. Fortunately smarter people are starting to anticipate resource issues a little better than they used to.
# of inbound links == page rank?? (Score:1, Informative)
If they really are using # of links as an approximation to search engine score, then they're flawed from the beginning. If they aren't, then somebody isn't very good at conveying information.
Re:Self-reenforcing cycle? (Score:3, Informative)
This survey's results make a great deal of sense to me. We write and market an extensive graphics package that in many ways surpasses Photoshop in capabilities. What we don't do is "market" it in the traditional sense (though you can find our zero-dollar "footprints" for instance my sig here, all over the web one way or another.) Adobe spends an amazing amount of money talking about Photoshop, compared to our zero dollars approach. We used to market like they did, back when we wrote Amiga graphics software. Sure, we sold a lot of product, but we spent so much on marketing that the actual gains after accounting for all the thrashing around with ads, reviewers, shows, distributors, dealers, packaging, brochures, mailings and so on weren't all that impressive. Classical marketing is expensive!
These days, we only sell direct and we have significant amounts of detail on the web — about 70 megabytes of docs, images and animations — the search engines, particularly Google, do a great job of finding us when people search out the types of graphics specific things our software does. We're astonishingly successful for the type of company we are today; I have absolutely no reason to complain, nor does anyone else who stuck it out with me all these years. We've been marketing online well prior to the advent of the web; we started doing it on Compuserve no less, and the more we did that, the better results we got. When the web came, our approach, which is put everything you can online and then some, began to work for us much better, the web being a richer environment than CIS ever was; and when search engines began to get reasonably smart, that pretty much taught us to kill our standard marketing. I don't regret it one bit (and by the way, I own the company outright, so that is the opinion of the company. :)
By comparison, every once in a while we get some press, though again, we don't actively seek it out. When that happens today, we see small spikes in sales. In the past, say, in the late 1980's, serious press (like a review) would make huge difference in sales for a month or so. My impression is that the impact of self-described "news" outlets has dropped in a big way since I began writing and selling software back in the late 1970's. I'd say they're essentially in the "don't count for much" category today, though I'm reasonably sure they'd offer a different opinion. :)
We do all of our business as a consequence of word of mouth and search engines. By all, I mean to say in the high 90th percentile. Not that we discourage anyone from reviewing, far from it... it's just that when you've got something specific, something technical, it's pretty much as the report says: people can find you, and they will. Another thing that helps is having truly unique content and capabilities; for instance if someone needs nondestructive geometric image manipulation, or morphing, they're going to find us, and quickly. If a site has "me-too" content that is duplicated in large part all over the web, I don't imagine the search engines would be all that useful, because now you're into trying to trick your site uprank with semantics, and while that may work for a while, the search engines are always mutating how they rank things and if you don't spend a heck of a lot of time on it, I could see any such effort sinking beneath the waves, as it were, within a relatively short time. We don 't spend any time manipulating our site's ccontents. We just write about what we offer, in as much detail as we can think of. If we think of something new, or need to make a correction, we certainly do that, but it isn't possible (or even reasonable) to try and fudge a site as large as ours based on today's particular version of how Google is going to rank things.
All in all, the results these researchers got fits in with my mental image of what the web "does" quite well. People who know what they are