Monday, February 7, 2011
Social Networks as filters
A few weeks ago, I found myself in the position of having to explain why the company I work for should register with various social networking sites. The question was clear enough: aren't these sites just for killing time and keeping up with friends? The answer was a little more complicated. There are a variety of reasons why a company should join a social networking site. Some of those reasons will be specific to the company's business model, but at least one of those reasons will be universal. Social networking sites filter third party influence from individuals' votes of interest in a website or online feature. You trust the link from your friend more than you trust the link from Nike. Let's get nerdy on this one.
A company must be visible and offer value. If no one knows I exist, I won't get customers and if the customers that do see me don't want my product, I might as well not exist. Online, visibility is very much a product of search engine rankings ("rankings"). The following data is pretty much old news, but it drives home the point that visibility equals income. This is all the more true when a company can convert a visitor into a customer online (think Amazon.com).
Search engines have helped to democratize the web by placing a strong emphasis on providing users with the most relevant results to their search query, regardless of the results' origins. In a world without search engines, we would only know of the URLs that were shared between friends or those that were marketed to us. This would favor the largest and wealthiest companies. So, no search engines, no internet start-ups, etc.
With click through rates being as high as they are for the top ten search results, it is always in a company's best interest to be highly ranked for relevant queries. Hence the market for SEOs. What these click through rates have really done is spawn two, simultaneous races. The first race is between companies, to see who can achieve the highest ranking. This is done through a large variety of methods that we can briefly touch on later. The important part here is that this has as much to do with making a better site as it does with trying to decode Google's algorithm (Bing and Yahoo's market shares are negligible in comparison http://www.seomoz.org/blog/top-10-things-the-microsoftyahoo-deal-change-for-seo). If a site can figure out how get a high ranking without restructuring their website or creating new features for the user, then their return on investment ("ROI") will be higher. It is much easier to write a great title tag than it is to generate new and compelling content.
The second race is within the search engines themselves and, of course, between them. The search engines must always refine how they determine website rankings. If Google stops providing you with useful results, you will likely begin using a different search engine.
The interesting point is how these two races play off each other. Google refines the criteria for its rankings, companies then chase the new algorithm, and Google must further refine its rankings, less they be co-opted by clever SEOs. If Yahoo! and Bing see Google slip or slowly address an issue with their rankings and are able to address it first, they may be able to take some of Google's users away.
A company must be visible and offer value. If no one knows I exist, I won't get customers and if the customers that do see me don't want my product, I might as well not exist. Online, visibility is very much a product of search engine rankings ("rankings"). The following data is pretty much old news, but it drives home the point that visibility equals income. This is all the more true when a company can convert a visitor into a customer online (think Amazon.com).
Google SERP Click Through Rates – The Raw Numbers
Rank# Click Throughs % Delta #n-1 Delta #n1
19,434,540 100%
1 8,220,278 42.30% n/a n/a
2 2,316,738 11.92% -71.82% -71.82%
3 1,640,751 8.44% -29.46% -80.04%
4 1,171,642 6.03% -28.59% -85.75%
5 943,667 4.86% -19.46% -88.52%
6 774,718 3.99% -17.90% -90.58%
7 655,914 3.37% -15.34% -92.95%
8 579,196 2.98% -11.69% -92.95%
9 549,196 2.83% -5.18% -93.32%
10 577.325 2.97% -5.12% -92.98%
11 127,688 0.66% -77.88% -98.45%
12 108,555 0.66% -14.98% -98.68%
13 101,802 0.52% -6.22% -98.76%
14 94,221 0.48% -7.45% -98.85%
15 91,020 0.47% -3.40% -98.89%
16 75,006 0.39% -17.59% -99.09%
17 70,054 0.36% -6.60% -99.15%
18 65,832 0.34% -6.03% -99.20%
19 62,141 0.32% -5.61% -99.24%
20 58,382 0.30% -6.05% -99.29%
21 55,471 0.29% -4.99% -99.33%
31 23,041 0.12% -58.46% -99.72%
41 14,024 0.07% -39.13% -99.83% (http://www.redcardinal.ie/google/12-08-2006/clickthrough-analysis-of-aol-datatgz/)
Search engines have helped to democratize the web by placing a strong emphasis on providing users with the most relevant results to their search query, regardless of the results' origins. In a world without search engines, we would only know of the URLs that were shared between friends or those that were marketed to us. This would favor the largest and wealthiest companies. So, no search engines, no internet start-ups, etc.
With click through rates being as high as they are for the top ten search results, it is always in a company's best interest to be highly ranked for relevant queries. Hence the market for SEOs. What these click through rates have really done is spawn two, simultaneous races. The first race is between companies, to see who can achieve the highest ranking. This is done through a large variety of methods that we can briefly touch on later. The important part here is that this has as much to do with making a better site as it does with trying to decode Google's algorithm (Bing and Yahoo's market shares are negligible in comparison http://www.seomoz.org/blog/top-10-things-the-microsoftyahoo-deal-change-for-seo). If a site can figure out how get a high ranking without restructuring their website or creating new features for the user, then their return on investment ("ROI") will be higher. It is much easier to write a great title tag than it is to generate new and compelling content.
The second race is within the search engines themselves and, of course, between them. The search engines must always refine how they determine website rankings. If Google stops providing you with useful results, you will likely begin using a different search engine.
The interesting point is how these two races play off each other. Google refines the criteria for its rankings, companies then chase the new algorithm, and Google must further refine its rankings, less they be co-opted by clever SEOs. If Yahoo! and Bing see Google slip or slowly address an issue with their rankings and are able to address it first, they may be able to take some of Google's users away.
Let's use a quick hypothetical here to illustrate how damning it would be if a search engine algorithm was decoded. If Nike ever figured out Google's algorithm, it would be very difficult to buy anything other than Nike products off of Google. Nike would have every incentive to create additional, satellite sites to consume the top ten rankings. A relevant query would produce a variety of websites, but only one option. The result would be a serious blow to Google's market share.
So isn't this post about social networking sites? Yes, it is. The search ranking criteria has evolved over time. It started off basic: search engines wanted to see that your website used the searched terms. A lot. Keyword stuffing reigned supreme. Then we moved on to meta-data: meta-tags; meta-descriptions; meta-keywords. This was all hidden data that web developers generated to describe their site. This played a large role until it was abused. It is now a much smaller factor. What we have slowly moved towards are the importance of external links. External links, their sources, the anchor text in the link code, and things of this nature play a very large role in rankings now. Why? Because they represent votes, by people at large, of a site's trustworthiness. When I link to X site, I am saying that its content is worth reading and that I trust it. When I describe X site with text in the link, it is my interpretation of what that site is. This is more trustworthy than a web developer's meta description. Additionally, the more links that go to a site, the more trustworthy Google knows it is.
Of course, when external links were first discovered to play a significant role, they were abused with linkfarm websites (think massive, useless directories). However, Google quickly remedied this by placing a greater emphasis on link source. A link from this blog is nowhere near as valuable as a link from the New York Times. The authority of the linking source plays a large role in determining the value of the link. It filters out third party influence and lends its authority to the linked site.
Social networking sites play a similar role. We can trace the link to its source because users must register with social networking sites. This means we know whether it's Nike or Nate who is advocating the newest Nike product. Search engines can trust these sources more because the source of the vote is transparent. Further, links and chatter from social networking sites don't impact race 1 (search engine v. company). Even if a company gets wise to Google's algorithm, they cannot easily manufacture mass approval of a product.
So isn't this post about social networking sites? Yes, it is. The search ranking criteria has evolved over time. It started off basic: search engines wanted to see that your website used the searched terms. A lot. Keyword stuffing reigned supreme. Then we moved on to meta-data: meta-tags; meta-descriptions; meta-keywords. This was all hidden data that web developers generated to describe their site. This played a large role until it was abused. It is now a much smaller factor. What we have slowly moved towards are the importance of external links. External links, their sources, the anchor text in the link code, and things of this nature play a very large role in rankings now. Why? Because they represent votes, by people at large, of a site's trustworthiness. When I link to X site, I am saying that its content is worth reading and that I trust it. When I describe X site with text in the link, it is my interpretation of what that site is. This is more trustworthy than a web developer's meta description. Additionally, the more links that go to a site, the more trustworthy Google knows it is.
Of course, when external links were first discovered to play a significant role, they were abused with linkfarm websites (think massive, useless directories). However, Google quickly remedied this by placing a greater emphasis on link source. A link from this blog is nowhere near as valuable as a link from the New York Times. The authority of the linking source plays a large role in determining the value of the link. It filters out third party influence and lends its authority to the linked site.
Social networking sites play a similar role. We can trace the link to its source because users must register with social networking sites. This means we know whether it's Nike or Nate who is advocating the newest Nike product. Search engines can trust these sources more because the source of the vote is transparent. Further, links and chatter from social networking sites don't impact race 1 (search engine v. company). Even if a company gets wise to Google's algorithm, they cannot easily manufacture mass approval of a product.
Importantly, the value of the link, tweet, post, etc., increases with the amount of additional social networking support it receives. If I tweet something and it is retweeted 100 times, that tweet gains more influence than the tweet retweeted 5 times. Why? We know it is relevant, we are more sure that it is not an artificially manufactured tweet, we know that the terms and information contained in that tweet are highly relevant to the website and thus are a good indicator of what the site should be ranked for, and this is a safer and easier way to determine internet rankings. Social networks are filters in that they ensure the activity surrounding a website is genuine and reflective of popular opinion. Accordingly, every company must be active on social networks.
And, just to back this up, here is a snippet from a recent NY Times article: "The move is part of a larger trend toward custom content. As news streams and Twitter feeds multiply on the Web, there is a flurry of new programs trying to help users cut through the noise. For instance, the iPad app Flipboard finds content from a user’s social networking feeds and displays it in a magazine-like layout of photos and text blocks that can be flipped through. AOLrecently announced that it would release Editions, its own personalized iPad magazine application. The tagline is 'The magazine that reads you.'" http://www.nytimes.com/2011/02/07/technology/07yahoo.html?_r=1&hpw
One last night, whether this could be accomplished through online publishing is a different issue, but I would argue that information sharing will only become more personalized and I question whether obtaining enough big links (NY Times) is possible to beat out people who obtain more small links (blogs, Facebook posts).
# posted by seventeennightsinthecity @ 12:22 PM 
Subscribe to Posts [Atom]