It seems that some of the big names in the industry decided to list their favorite ways to build links, and they are all well worth a few reads to digest them. Without further ado:

• 101 Ways to Build Link Popularity in 2006 by Andy Hagans and Aaron Wall. Notice how many times they mention tracking your results and your competitor’s results… the perfect use of What the Buzz? (sorry for the shameless plug )
• 5 Tips for Media Link Love, by the one and only DazzlinDonna. Donna focuses on getting links from the media such as newspapers, TV and radio stations, etc. Don’t know about the dog parade though!
• 12 Different Types of Links and How To Get Them by Todd Malicoat. Todd’s list is great because it is more of a how-to, giving you detailed examples and comments based on his experience.

Update:

Threadwatch has a longer list.

[tags]seo, link building[/tags]

Yahoo launched its new corporate blog yesterday, called Yodel Anecdotal. The kickoff post is very funky and has a video (a Yahoo! Video of course) that makes you want to work for them.

I wonder how Google’s blog started… This makes me think that Yahoo might overtake Google in branding power. And any iPod cynic can point out, technical superiority is not enough these days: you need branding power.

[tags]blog, yahoo, google[/tags]

The question of how search engines assign geographical locations to domain names is an important aspect of SEO. The importance of geolocating a website is in local search: the big search engines usually have country-specific (e.g. yahoo.ca) and local searches. The algorithms powering these searches may favor websites relevant to the local search; for example, Canadian sites may be weighted more in Canada-specific searches.

There are many ways how SEs assign a geographical locations to a website. Bill Slawski has some excellent coverage of Google’s local search patents, in particular, this one. There he talks about how on-page content (such as a telephone number or an address) tip the Google that the website is related to a certain geographical area.

Recently, the Google Sitemaps blog spelled out (very clearly, for a change!) how to target non-US local searches. To quote that page:

If you want your site to show up for country-restricted searches, make sure it uses a country-specific domain (such as www.example.com.br). If you use a domain that isn’t country specific (such as .com), make sure that the IP address of the site is located in that country.

This tells us that the TLD of the domain name (.uk, .ca, .au, .br, .ch, .de, etc) is a strong hint in Google’s eyes. If you have a generic TLD, like .com or .net, then where the site is hosted is another hint for Google.

So that’s Google covered; how do MSN and Yahoo! deal with this issue? Over at SEOmoz, Rand posted in detail about how MSN determines non-US targeting of a website. The idea is quite clever: if a lot of Canadian websites link to your website, then your website is likely to be relevant to Canadians. How a search engine determines if a link website is Candaian or not is again based on hints like the TLD, its hosting location, etc.

So today, I launched a tool that does exactly what I described above: trying to assign a geographical location to a website. The tool, called the Domain Geolocator, does the three things described above:

• Look at the domain’s TLD, which is the biggest hint.
• Also look at the domain’s hosting location, which is useful for generic TLDs.
• Profile the countries of the incoming backlinks using the Yahoo! API. To do that, it fetches the top 1000 links of a website (or all of them if less than 1000) and counts the TLDs of the linking domains. Further, for each linking domain, you can view where it is hosted, again, mostly useful for linking domains with generic TLDs.

The data is cached, meaning the first time you run the tool, it is slow (imagine fetching 1000 links and processing them…) but later searches are fast. Further, the tool allows you to easily bookmark the results.

As always, feedback is most welcome. There have been a few suggestions to improve the tool already (!), but I’d love to get more comments to incorporate them into a future update.

The link again: Domain Geolocator.

[tags]SEO, Google, Yahoo, geolocation[/tags]

For the past couple of weeks, ten people, including me, have been busy doing SEO interviews hosted by Rand at 14th Colony. The final result is a set of lessons for everyone including a few of the biggest names in the industry.

It’s been fun, and now I can finally tell you about it. Have fun reading!

[tags]seo, interviews[/tags]

A very cool new study from Hamburg University researchers about how people interact with websites. From their study results:

A recent clickstream study revealed new information about how we use and peruse the Web. University of Hamburg researchers found that the Web is moving from a static hypertext information system to dynamic interactive services with rapid interactivity between man and machine. The authors recommend that web developers create concise, fast loading web pages to keep pace with the speed of web navigation.

The Hamburg researchers also found an F-shaped pattern of clicking activity similar to results found in eye tracking studies. There appears to be a strong link between our eye and hand movements over the surface of web pages.

Be sure to check it out!

[tags]clickstream, analytics, internet, research[/tags]

Graywolf reports on Google indexing Lycos SERPs in a search he’s found. A quick Google search reveals the extent of the problem: 40,000 such results. Nice.

Makes you want some therapy products

[tags]Graywolf, Google, Lycos, SEO[/tags]

Who would have thought: Some DMOZ editors are spamming Google and other search engines. A quick search of some typical spam keywords in dmoz.org picks up some editor profiles. Needless to say, that’s just abuse of the power these editors have, not to mention abusing the trust Google places on the site.

And the spam is hidden using the classic techniques: you won’t see it if you just view the page, but it’s obvious if you view the source.

The full details are over at SEO Scoop (what a scoop this is!). Try the searches the article points to in Google, Yahoo, and others to see just how spammy the pages are.

Update: It’s made it to digg.

[tags]spam, SEO, Google, Yahoo, DMOZ[/tags]

Yahoo! offers a set of Javascript-based buttons to help webmasters "turn site visitors into evangelists" (their words!) The buttons let your visitors quickly bookmark the page to del.icio.us, email the page, blog about the page, or print it. Quite cool stuff, but of course, you should also check out the Socializer .

The point of this post though: Yahoo! has now published a guide for these social media tools. The tools are part of the Yahoo! Publisher Network, which also includes the contextual advertising program of Yahoo!. Interesting how ads and social media are tied like that…

[tags]Yahoo, YPN, Social, bookmarking[/tags]

A new paper published by Korean physicists tells us that the best way to win bids, on auction sites like eBay is… sniping. And how do they know that? To quote their abstract:

We find that the number of distinct bidders who bid k times up to the tth bidding progresses, called the k-frequent bidder, seems to scale as n_k(t)~tk^–2.4. The successfully transmitted bidding rate by the k-frequent bidder is likely to scale as q_k(t)~k^–1.4, independent of t for large t. This theoretical prediction is close to empirical data. These results imply that bidding at the last moment is a rational and effective strategy to win in an eBay auction.

To rephrase that: if you model the behavior of bidders, you find that the most successful ones are the bidders that do not bid until the very last second, i.e., snipe an auction and win it. That does not mean they always win, but it is still the most successful strategy.

[tags]ebay, sniping, bidding[/tags]

I have a free fonts download website. While looking through the Google Sitemaps, I saw this list of common ‘words’ in my site’s contents:

Eh? These are not words! What’s going on? Thinking it might be Google trying to display binary data, I searched for the download script of the site, and sure enough, binary data is being shown in the SERPs:

Clicking on those links brings up the ZIP file download for each font, exactly as expected.

Anyone else seeing this?

The Journal of Law and Economics has a special issue about online file sharing and its effect on the music industry. Their 2-month-late press release states:

Downloading reduces per capita expenditures by $25, but increases the amount of music each individual consumes by the equivalent of $70
An important collection of papers from the current issue of the Journal of Law and Economics addresses the battle between copyright owners and copying technologies. Together, these essays by some of the leading researchers working in the field offer new perspectives on the economic and social impact of sharing sound recordings over the Internet, including a groundbreaking analysis of the impact of well-publicized legal threats against individual file sharers.

Of particular note is an exploration by Rafael Rob and Joel Waldfogel (University of Pennsylvania) of the sales displacement induced by downloading, incorporating consumer valuation in dollar terms of purchased albums vs. downloaded music. In a sample of college students at four U.S. colleges, the researchers found that downloading reduces per capita expenditures by $25, but increases the amount of music each individual consumes by the equivalent of $70. Thus, among their sample group, $45 worth of music – three albums – would have otherwise been forgone.

"While perhaps paradoxical to the law-abiding citizen, illegal downloading may actually alleviate the monopoly deadweight-loss problem" write Rob and Waldfogel. "Indeed, downloading allows consumers to engage in a crude “do-it-yourself” form of price discrimination."

It sounds very interesting, but you need to be a subscriber to read the papers. I’m not unfortunately, so I cannot tell you more

For the past few weeks, I’ve been collecting the number of clicks each social bookmarking service receives in the Socializer. The results rank the popularity of each social bookmarking site, and are very interesting.

Read the rest of this entry »

Seth Godin pointed out last April that lawyers are part of your marketing department. Today, over at ClickZ, Jeanniey Mullen explains an email blunder of hers and talks about the power your email marketing people have over your brand. So let me kick open this partly-open door: EVERYONE in your company who interacts with the outside world is part of your marketing and your brand.

An example: your recruitment/HR people are part of your marketing. What does your company do when they receive a new application? Do they send out a blunt/terse/almost rude auto-reply that says ‘if you don’t hear from us, consider yourself rejected’ or does it send out an email that addresses the applicant by name, thanking him/her for the application and explaining what will happen now? Do you even send out a reply?

Another example: what do your employees say about your company in private? Do they slander it or say it’s a nice place to work?

So let’s sum it up: everyone in your company is part of your marketing department and your brand. If they love the company, the whole workforce will tell the world about it. If they hate it, the whole workforce will tell the world about it.

[tags]marketing, brand[/tags]

What happens when Web 2.0 meets scientific publishing? The answer is called PLoS ONE. Everything from wikis, to blogging, to tagging, commenting and all the other buzzwords are represented. Not bad for one press release!

Question: with all the craze of APIs, will spam increase? My answer: yes, for two reasons.

Social bookmarking site, such as delicious, can be accessed using an API, an application programming interface. The API is simple a set of function that sit on the delicious servers that you access using URLs. The URLs are specially formatted to pass information to the functions, and the functions send back XML-formatted results. Simple and useful, and can be automated.

However, one of the functions allows you to post links to delicious. Think about this for a second. You can, without human intervention, add links to delicious. Spam anyone? But you have to sign up for an account, right? Yes, sure. Imagine you spend a few hours signing up for many delicious accounts using proxy servers. Next, you write a little script that automatically posts to delicious your spammy links, using a different proxy server for each account. End result? You can probably reach the most popular list, probably have a link constantly on the front page and many other things. Add to this that delicious lists are syndicated, and suddenly your spammy pages pick up enough backlinks to rank for something.

And it’s not just delicious: I’ve seen such API functions from a few other social bookmarking sites. This loophole is just too easy not to exploit.

Now part 2 of the answer: tagging. The latest fad on the net is called tag and ping. The idea is simple enough: you tag your content as belonging to a certain category, such as mortgage or viagra, or internet, cars, etc. Technorati then comes along, indexes your content and magically believes your tags. See the problem? Tag enough of your content with a high-income, highly-searched-for keyword, and you can drive enough traffic to your sites. Pinging sites like Technorati too often can trigger spam filters, but a blog network covering the same topic can ping it without too many problems. Imagine your blog content being the top 10 results in Technorati. Sweet $$$!

So what’s the solution? For the APIs, removing the posting functions will do the trick. As for tagging and pinging, I have a couple of ideas, but no real solution. It’s a matter of what Technorati and brothers believe as genuine tags.

I’ve already talked about some ways to get blog traffic. Now, marketing guru Seth Godin lists 50+ ways on how to get traffic for your blog.

The list has some funny points, but it’s missing one important (IMHO) point: answer your blog’s comments! Blogs are dialogues, and if you don’t interact with your visitors, the blog becomes like a man preaching in the street with a loudspeaker. How many people pay attention to him?

A less than technically-oriented friend of mine was searching Google today. When she clicked on one of the results, it turned out to be an RSS XML feed. Immediately, she said "Oh it’s a screwed up page" and I watched her helplessly scroll up and down the page trying to find something to click on or do.

So that got me thinking: what’s the point of having raw XML as a result in search engines? It’s not meant to be for human consumption anyway, so why are the search engines doing it? Also, it would not be that hard for an algorithm to figure out the website’s page that the feed corresponds to. For example, it would be better for Google or Yahoo! to have the blog as the result instead of the RSS feed. Isn’t the goal of SEs to deliver user-relevant information? XML doesn’t qualify, IMHO.

Yahoo! business news reports that Yahoo and eBay formed a web advertising alliance. To quote the news article:

Under the deal, Yahoo will be the exclusive third-party provider of all graphic ads throughout eBay’s auction site. Yahoo has also chosen EBay’s online payment system PayPal to allow its own customers to pay for Yahoo Web services.

In addition, Yahoo Web search features will be integrated into a co-branded version of the eBay toolbar, and the companies said they would explore developing "click-to-call" ad technologies on their respective Web sites.

This comes hot on the heals of the news that JP Morgan released a report (pdf) that Yahoo and EBay are most likely to merge.

This might be the latest salvo in the net wars: back in April, we learned that EBay was in talks with Yahoo and Microsoft. Back then as MSNBC reported:

Under one scenario, the talks could lead to an alliance in which eBay would boost its advertising spending with its chosen partner and provide access to data it has collected about its consumers…

That sounds familiar

[tags]Yahoo, EBay, Microsoft, JP Morgan, online advertising[/tags]

Yahoo! today announced that their new home page design is live. I took a look, and my first impression can be summed up with ‘WOW’. Very clean design and easy on the eyes.

One thing that caught my attention: they moved to a left-side based navigation, instead of their usual top-navigation. All screenshots of the new test design for Google also show that Google is playing with left-navigation. Will these two online giants set a trend for Web 2.x?

I’ve been using Yahoo! more and more lately – actually, I’ve set it to be my default search engine in Firefox. It’s a great search engine, and this design might help it compete better with Google.

To try it out: www.yahoo.com/preview. Do you like it?

[tags]Yahoo, Google, search engine, Web 2.0[/tags]

Lots of recent posts at my favorite forums about the long tail of search got me thinking: How can we quantify, fairly accurately, the length of the tail for a web page? I came up with a formula and I wrote a PHP script that uses it to calculate the tail length of a web page. This post gives the details and the first results. Please note that this data is very preliminary, so it’s more about starting a discussion than the data

Read the rest of this entry »