A while back I posted about how you might be able to stop some of your less savy competitors copying your links.

As a follow up I’ve just done a guest post over at sem-group.net entitled ‘How to stop your competitors copying your links‘ which discusses a few more ways you could go about hiding your backlinks or better still making them impossible to copy.

The guest post was part of a guest posting competition with some pretty decent prizes for the best post. If you enjoy my post (or if you don’t) it would be great if you could leave me a comment or a tweet to help me win:-)

Below are the sponsors of the guest posting competition…

$200. Prize Sponsors

Daniel McGonagle Link Building

BestTravelWebsites.com Best Travel Sites

ArrangeYourVacation.com Vacation Rentals

$150 Prize Sponsors

Candy Store

$100 Prize Sponsors

Alisa Bowman Marriage Advice

Internet Marketing Services

SEO Software

SEO Ottawa

Virtual Assistants

Hesham Zebida Thesis Skins

Gerald Weber Houston SEO

Toni J Young Network Marketing Training

Quillcards Distinctive Ecards by Quillcards

Ghostwriter Dad Ghostwriting services

$50 Prize Sponsors

Whitney Segura Whitney Segura’s Internet Marketing Blog

James Brown James Brown

Raxa Design Houston web design

Eric Brantner Freelance Copywriter

The Servant Media Houston Business Development

Ana Hoffman, Traffic Generation Cafe Increasing Targeted Website Traffic

The Lane Real Estate Team Kennewick Homes

Other prizes: Non-Cash Sponsors

Hostgator.com 1 year Business Hosting plan. Includes toll free phone number and SSL $179.40 value Business Web Hosting

David Harry A full years membership at SEO Dojo, a value of $250.SEO Training Dojo

Link-Assistant.Com SEO PowerSuite Enterprise (max. functionality license) $599 worth with a life-time Live! Plan subscription. SEO tools

Special thanks to the contest media partner: My Blog Guest. MyBlogGuest.com is the free guest post exchange community where users meet to exchange guest posts and network.

I’ve seen and heard people misunderstanding the idea of long-tail keywords right from newbie SEO account managers to SEO expert speakers at £500 a seat conferences. I really don’t think its that hard to get but there’s a lot of mis-information about, like the definition Google have chosen…

Long tail keywords actually have nothing to do with the number of words in the query and everything to do with the volume of searches that query gets (demand). A one word keyword can be long tail if it rarely gets searched for. A query made up of 10 words can be short tail if it gets 10,000 searches a day.

While its true most long tail keywords are made up of 4 words or more I personally think concentrating on this is a misreading of the original concept.

Or am I completely wrong here?

ps – more proper posts coming here soon

OK so I know the fashion for SEO blogs these days is bashing link buying or ‘outing’ sites or agencies who are doing it but allow me to play devils advocate for a minute. Forget about ‘ethics’, ‘outing’ or the threat of penalties for a couple of minutes and bare with me while I discuss the right time and place to buy links.

Despite what certain figures in the SEO world would like you to believe buying links is still a pretty integral part of the SEO process for many sites in competitive industries. I have no idea how much the text link industry is worth but I’d speculate hundreds of millions annually- after all Google see it as a big enough threat to their own ad income to have a whole department dedicated to trying to police it.

I think there’s a few reasons link buyings got such a bad rep these days

• People remember TLA and the mess left when Google kicked that network into touch.
• There’s a huge plethora of new and even dirtier link buy/ sell networks which continue to pop up and offer absolutely no value to advertisers
• People associate link buying with sites from dirty industries like gambling
• It’s a bit of a shady world dominated by under the table, under the radar deals completed over Paypal

This is all fair enough and I see why SEO agencies in particular try to steer clear. Not least of course because every penny a client spends on links they could be spending on ‘link bait’!

Problems with competitor link analysis

Here’s how most webmasters go about buying links.

“our competitors have got links on these 20 sites. We’ll buy links in all the same places and then find a few extra sites to get links from and we’ll outrank them”

This approach is flawed in so many ways I almost don’t know where to start. By using competitor link analysis to find paid link opportunities you’re starting a game of catch up which you can never win.

While the theory of co-citation analysis is sound, what it doesn’t account for is the effect of link age. If you’re competitors already had a link in place for 6 months, a year, maybe more, any link you buy on the same pages isn’t going to be as valuable, because it’s not as old. At the same time for every new outbound link which is added to a page the value of all the other links on that page goes down so while you may be able to buy identical links to your competitors you will never be able to get the same value from them as they have. Furthermore you’re likely to end up paying more than your competitor for the same link because webmasters know you’re going to want it more.

So you’re paying more for links which aren’t as good- not a smart strategy.

The other thing this approach doesn’t take into account is what your competitors are doing right now. Maybe another competitor will come in behind you and buy the same links as you, diluting the value of your link down again- now the link which you might of decided was worth £50/month last week when there was only 1 other outbound link on the page is probably only worth £40 with 2 other links on the page so you’re now tied into overpaying for a link just to keep up with your competitors. Where does it end?!

As a result of this what you see in most industries is the sites who got in early, brought links aggressively back when everyone was doing it and have left them in place continue to dominate. While new entrants can have some success, you can never win at SEO by playing this game of cat and mouse.

So when is the right time to buy links?

Paid links should only ever be the last resort and never the default type of link building which you turn to first, early on in your campaign. This is simply a question of economics. It takes time for the links you buy to deliver improved rankings. Unlike conventional advertising when you’re getting value from the time your ads go live on a site, with text links you might have to wait months to start scraping back the money you’ve paid out for your links. That means you need to minimise the time you’re paying for a link which isn’t delivering rankings. So quite simply, the higher you’re ranking when you start paying for the link, the shorter time that links going to be a loss-leader.

Again most link buyers don’t take this into account. They’ll buy links for a site which is ranking nowhere because it’s the quickest way to get the site to rank somewhere. But nowadays you can’t buy a new site straight to the top in any semi-competitive industry.

To give another hypothetical example let’s say you set up a £10k/ month cost base with paid links for a newish site and it takes 6 months to get that site onto page 1 of google for whatever keyword you’re targeting. That’s wiped £60k straight away from the value of that first page position.

(I started drawing a stack graph to illustrate this example but gave up, I’m sure you can picture what it would of looked like!)

So to spell it out.

• Do everything you can to get links without buying them.
• Push your site as high as you can with non-paid link building before you buy links
• When you buy links target your highest ranking keywords first, that way they’ll start paying for themselves quicker and the indirect effect of the link on your “domain authority” should also improve lower performing keywords as a by-product, thereby reducing the requirement for more paid links for those other keywords.

Think of your link building project as a marathon, and paid links are the bit of energy you keep in reserve for a sprint finish once all your competitors have burnt out!

Managing risk

If you’re concerned about the risk of getting penialised by Google for your link buying then this is definitely the safest approach to take as well.

The reality of paid link penalties when they happen (which is historically rare but perhaps getting more common) is that the buying site doesn’t get penialised at all. All the sites which they’ve brought links from get hit with a penalty meaning they no longer pass ‘link juice’. When this happens top ranking sites which only had paid links see a big ranking drop because it’s as if all their links have been removed at once meaning they go back in the rankings to around where they would of been without any paid links. This is often mistakenly called a ‘penalty’, where actually it’s just Google resetting the rankings to how they would have been before any paid link manipulation.

If you’ve only used paid links to get you from the bottom of page 1 to the top then even if the worst did happen, the lowest you’re likely to drop in any reshuffle of the rankings is back to the bottom of the first page- not ideal, but not the end of the world either.

The other reason you don’t want to buy links from the same sites as your competitors is that you don’t want to go down with their ship. If all the sites they’ve brought links from get hit with a penalty you’re going to see the same drop in rankings as them. If you’ve sourced your own link prospects these are unlikely to get penialised, at least not at the same time, as your competitors linking sites so you benefit from their misfortune.

Feel free to disagree with any of my points here in the comments…

This is a pretty boring post but it might be relevant to anyone doing SEO for travel sites.

I’ve noticed a few links I’ve got on lonelyplanet.com have had nofollow’s added to them. These aren’t links I’ve dropped in the forum of anything like that, its the proper links which the editors have handpicked for each location. If you look at the Mallorca guide page (yeah I nofollowed that link!) those links in the article have had the rel=nofollow tag applied to them.

Who cares right? Well I do cos I’ve lost some links but I do also think that if any site in the world should be following external links its the Lonely Planet. Lets face it those links:

• Are editorially researched by professional travel journalists
• Offer key information not covered by the page
• Are reproduced in print (once you’ve printed a book you can’t go back and nofollow it)

If big trusted sites like this keep making ill informed SEO decisions its going to make Google’s job more difficult because they’ll have to work out which nofollows to obey and which to ignore. After all following the link path from editorial trust centres like lonelyplanet.com is still the cornerstone of the algorithm. It wouldn’t be so bad if lonelyplanet had anything to gain from it but I think its unanimously agreed now that trying to hog PageRank in this way doesn’t help you in the slightest.

If anyone’s been on my Twitter feed you may have noticed I came unfashionably late to the Twitter party. When I first saw short URLs popping up all over the place I didn’t really think much of it but now that Twitter is bigger than Jesus and tinyurl and bit.ly links are as common as regular links its suddenly got my attention.

I did some background reading on the subject of url shorteners and SEO and although people are talking about this, at length, there’s some glaring problems which in my mind make URL shortening services a terrible idea for anyone who cares about their SEO.

Apologies if any or all of the following points have been said a hundred times before, I got bored of my background reading on the subject after the tenth article which did little more than pointing out that you should choose a shortening service which uses a 301 redirect. If anyone has links to places this stuff has been discussed properly please share them in the comments.

Anyway here’s my run down of the pro’s and con’s of URL shorteners for SEO as I see them.

Pro’s

None that I can see.

Con’s

URL shorteners create an extra point of failure

This isn’t just an SEO issue. With anything you do on the web you should start with the realisation that things break. If you think about the huge volume of traffic some of these shortening services must be serving they’ve got to have some serious server farms and load balancing going on. I can’t see how they make money either so I’d assume these aren’t the worlds best servers they’re running on.

If that service goes down your link is broken, simple as. Sure your own site can go down but when you add a shortener service into the mix you’ve got 2 points at which something can go wrong and the request can fail. From an SEO point of view it doesn’t matter if these services use a 301 redirect, 302 redirect or print out your web page and fax it to the user, if their server goes down Google and anyone else won’t follow your link.

Not to mention if the company goes bust billions of links just get vaporized!

Hacking

I think the only thing worse than using URL shorteners on sites like twitter is when people paste those short links onto their own sites. If I was a hacker (a good one) I’d be concentrating 100% of my time and effort on penetrating some of these shortening services. Imagine the damage you could do hacking a shortening service and redirecting every short URL to a piece of malware. If that happened every site with short URL links would be linking out to spam sites which is obviously bad news for SEO and will likely piss off your visitors a bit also. I don’t see any reason to ever use short URLs in blog posts except sheer laziness copy pasting links out of twitter.

Users don’t know what they’re clicking on

Nothing to do with SEO, but personally I’d rather click a link with some idea of the site I’ll end up on.

Link scraping

Twitter feed with followed links on review centre blog

Although links on Twitter are nofollowed there’s a chance your links are going to end up on other sites like with those wordpress plugins which display latest tweets or in the example I gave above where bloggers copy and paste a tweet into a blog post.

At BrightonSEO last week Kevin from SEOptimise demonstrated a plugin called blackbird pie (WTF!) which makes this copy pasting easier, although it doesn’t look like it nofollows the links which get scraped off of Twitter so if you look at the bit.ly links on this page are followed and should pass PageRank.

If those links were direct rather than shortened they’d be pretty valuable SEO wise. Even if your shortening service uses a redirect which Google could potentially follow I think its pretty much universally excepted now that link value will diminish through redirection of any sort and I’d speculate especially if that redirection goes through a 3rd party site rather than from 1 page on your own site to another.

Furthermore with the shift of emphasis to brand since Vince I think its fair to say an anchor text link your-domain.com has more value than someone-elses-domain.com.

so:

bit.ly/1234 redirect to johnmcelborough.com/url-shorteners

Passes no anchor text and link value diminishes through the redirect. Where as

johnmcelborough.com/1234 redirect to johnmcelborough.com/url-shorters

Passes branded anchor text and less link value diminishes in the redirect from one page to another on the same domain.

Twitter links and ranking

On my post about building links to ecommerce product pages I floated this idea that links on the real time web can directly influence your rankings, just not in a conventional, pagerank/ anchor text sorta way.

Again its just speculation at this point but for all the same reasons why I want links which get scraped off twitter and followed to link directly to me rather than a URL shortener it stands to reason that links actually on Twitter itself could have more SEO value if they don’t go through redirects

The influence of brand citations

Another fairly speculative idea this one but its an established local ranking factor that unlinked citations of your business on the web could help your local rankings so with the shift towards brand in the algorithm and just generally from  a branding perspective if I’m putting a link out on  Twitter I’d rather it had my name/ brand name in it than the brand of another company!

i.e.

johnmcelborough.com/1234 is a brand citation for johnmcelborough.com

where as

bit.ly/1234 is a brand citation for bit.ly

The increased importance of load speeds

Redirection is an incredibly bandwidth hungry exercise. If you’ve ever seen the setups of some of the big adservers you’ll realise that a couple of redirects and a bit of tracking script become pretty serious business when you multiply it by 1o0 million ad impressions a day. The sames true of url shorteners. Now, again, its not likely there’s a direct SEO ramification from this I just think best practice should be to take the path of greatest speed and least bandwidth which means linking directly rather than through another server. If indeed Google is following Twitter links to some extent its reasonable to assume that they’ll prefer to follow links which don’t slow down their bots.

Why do you need to shorten a URL?

Not withstanding all the SEO stuff above, most of which is admittedly going to net you marginal gains, the question still remains why would you want or need to shorten a URL?

In offline advertising short URL are used for tracking and to make page addresses easier to remember so you might shorten:

http://www.moneysupermarket.com/insurance/travel-insurance.aspx

to

www.moneysupermarket.com/travelinsurance

Online the URL doesn’t need to be recalled and bit.ly/ Tinyurls aren’t great for recall anyway so its just about the space saving when using services like Twitter which have character limits on posts. But with a blog post how much do you really save by shortening a URL? Look at the URL for this post:

http://johnmcelborough.com/url-shorteners – 41 characters (you can safely drop the www because that will redirect anyway)

Now if I shortened that URL with bit.ly I’d get:

http://bit.ly/b0B1qT – 20 characters, a saving of 21 characters

or with Tinyurl:

http://tinyurl.com/33bo7p9 – 26 characters, a saving of 15 characters

Now I haven’t exactly got the shortest domain name as it is (damn those Scottish/ Irish roots) but pasting in my full URL I’ve still got about 100 characters to add my message, any tags and allow room for people to retweet and add their own comments. So for bloggers, when its so easy to write your own permalinks or just set up your own redirects I really don’t see the point in shortening URL’s just for use on twitter.

Obviously long ass URL strings like you get from Google maps or product pages on big shopping carts- fair enough you’re going to need some sort of shortening service but I’d definitely be inclined to set this up yourself if you’re anticipating a lot of people wanting to link to you from social sites. It might be a bit of work so if it looks like its going to be expensive definitely don’t make it a priority but most open source CMS’s will have components which let you setup your own redirects quickly and easily so redirection for wordpress or ushorten for Joomla.

So what do you think? Like I said at the top I really don’t use Twitter so I’m looking at URL shorteners from an outsiders perspective.

This is one of those questions which comes up every now and again in the SEO community but tends to get skimmed over because there’s not a huge amount of conclusive information available on it.

“If a bad (spam) site links to me, will it damage my rankings by association?”

This naturally has to lead on to the more important question for SEO’s:

“If I buy bad links for my competitor will it damage their rankings and therefore help mine?”

I doubt many SEO’s haven’t ever though about ways to do a number on their competitors and mess up their rankings. As search results become more and more competitive, competitor sabotage is at some point going to become to the SEO world what click fraud is to paid search. Except with SEO you’ll have no Google account manager to run to when someones out to mess up your rankings. Let me start with a hypothesis:

Probably the fastest and easiest way to take down a site is by pointing spammy paid links at it.

There I said it.

This isn’t just speculation. I’m going to share with you a few examples of exactly how this could happen. In these examples I nobbled myself but I believe I could just as easily of nobbled a competitor with the exact same tactics. Hold tight as we cross over to the dark side people!

Look at the graph below. That sudden drop in Google traffic from January to February is where Google knocked this site for 6 after I pointed about 50 TNX links at it over 2 weeks.

That was January 09. Until last week when the site started ranking again for its main keyword (which incidentally was when I decided to look into this further) Google sent a total of about 5 visitors to the site in 18 months, and those were just me running site: commands.

Now for obvious reasons I’m not going to tell you the domain but for those who are interested

• It was a pretty new site when I pointed the links at it.
• Exact match domain for the main keyword on a .co.uk
• Only a few pages, but all unique content, no outbound links
• It had some OK quality links from other sites (which I own but on other servers) before I brought those spam links

Could be a one off right, after all its not the best site and probably doesn’t deserve to rank that well anyway.

Here’s another example, this one where I pointed 3 Inlinks at a different site for one month, then removed them.

Again its a new site. That peak is where the main keyword moved onto the first page of Google, the dip is where the paid links were removed. To this day the homepage of the site isn’t even indexed (although the other 15-20 pages are) and it gets no traffic from Google. The funny thing with this one is the site ranks #1 in yahoo and #5 in Bing for the main keyword.

According to Yahoo Site Explorer the site has 44 links from external domains, all of which were built at the same time as the Inlinks were added so the ratio of paid to non paid links was something like 1:10. A year later despite the site being 12 months old, with some non-paid, non-spammy links which haven’t changed since the site was first indexed- the homepage is well and truly planted on the infamous ‘Google blacklist’ because of 3 paid links which were removed about 11 months ago.

Now don’t get me wrong- these are pretty thin affiliate sites on new domains and I was playing around with some link networks. I deserved and expected the slap I got. However to me this stands as evidence that Google not only devalues spam links but actively penalises sites which get linked to by spammers or known link sellers. I don’t think the Inlinks would of been detected if I hadn’t removed them because the blogs they were on were fairly decent but some of the sites on the TNX network were shocking.

Google has to look at inbound links for spam detection

When you really think about it the idea that Google is able to ignore spammy paid links to good sites while being able to effectively punish link sellers, link buyers and link networks is just implausible. With inbound links being so important to improving your rankings, it stands to reason they can hurt your rankings as well- surely that’s just common sense.

An information vortex

Now I’m definitely not advocating the use of link spamming your competitors as part of an SEO campaign, I just think webmasters and SEO’s need to wise up and be aware of this stuff because in reality Google will never admit that they have no way of policing paid links and link spam without inadvertently making it possible to bring down good sites with links from bad sites. There’s no information I could find on webmaster central about this and the advice which comes from Google reps is inconclusive at best and deliberately misleading at worst.

If you scroll the video below with Maile Ohye from Google to about 6 minutes she gets asked a question on this subject which I think she probably misses the point of (we’ll forgive her, she’s pretty cute!)

SEOmoz Whiteboard Friday – Interview with Google’s Maile Ohye

Maile goes on to answer something like:

“Those are usually such low quality sites that we wouldn’t consider them to be something that you paid for”

That point immediately resonated with me. Pointing actual spam at your competitors probably won’t damage them but if you make the spam links look ‘like something that they paid for’ then you could cause all sorts of problems for your competitors. I would personally question where the distinction lies between ’spam links’ from scraper sites and ‘paid links’ from spam networks like TNX but I guess the takeaway here is you don’t need to worry about really bad links, its the links which are bad but don’t look that bad which can hurt you.

The SEO industry and link sabotage

For such a vocal industry not many SEO’s talk about this kind of stuff either and when it does come up it tends to be approached with the same level of non-shalance that Google employ.

Take SEOMoz who are an easy target because they’re regarded by most as the centre of all SEO knowledge in the known universe. They put out a post talking about how buying links still works (not sure why they were so surprised as about 9 months ago their consulting page still listed link buying under their SEO services) but at the end of the post they add a warning:

Some friends of SEOmoz who run a fairly well-established site recently ran into a snag–they vanished from Google. They had ranked in the top two for many moons, raking in the lucrative spoils of their hard-won rankings. Then they got greedy; they thought a couple of paid links (four to be exact) could secure them the number one spot for all eternity. They wanted to be like the lone Highlander atop his mountain. They bought their links, and it worked for a minute. Then Google beheaded them (to continue the Highlander theme) by abso-friggin-lutely burying their site. Their links were discovered and now they can’t even rank for their business name or their full title tags. Suffice to say, this has made business a tad difficult.

Fair enough, this is useful information, if its true, but then the next week comes this post about how bad links can’t hurt your site which basically regurgitates Google’s advice on the subject to not worry about the bad links as long as you’ve got trust. So which is it? Did their friends knock themselves out with bad links, meaning their competitors could just as easily, or should we just chill out and let any old site leave footprints all over our link profiles hoping Google are smart enough to work out what links I’ve paid for and what links my competitors have brought on my behalf?

Real world applications of competitor link sabotage

Maybe normal webmasters don’t need to know this stuff but professional SEO’s should definitely be aware of it. Coming back to the examples above I brought down 2 of my own sites with less than $50 worth of link buying. Its an extreme example but its too easy to dismiss this as the backroom experiments of an SEO geek- put it into a real world situation for a minute and it becomes more of an issue. Lets role-play for a moment…

• I open a brand new bed and breakfast in the village of Littlebottom
• You live right across the street and run the only other B & B in Littlebottom
• I launch my website but I know nothing about SEO, after all I’m the owner of a fictional B & B in a fictional village in the home counties, why would I!
• You, happening to know a bit about SEO, see my new website go live and fire a couple of hundred cheap, dirty paid links at it.
• New domain, 200 obviously paid links, no good links… An alarm goes off in Google HQ, Larry Page gets woken by a phone call in the middle of the night and my site gets blacklisted
• I go about my business, don’t really think to ever Google my own website, don’t know anything about link building so I never pick up any other links apart from the ones courtesy of my friend across the road. My site might never reappear. In cyberspace you’ve still got the only bed and breakfast in Littlebottom.

Another extreme and unlikely example? Perhaps. But I can tell you right now if I dangled the option of competitor sabotage in front of some of the hoteliers I know they’d bite my hand off for it! Its no good sitting back and relying on the honesty of webmasters. If the short history of the web has taught us anything its that webmasters aren’t honest- they’ll buy links, sell links, steal links, cloak, comment spam, hack, spam report, leave fake reviews and then lie about doing any of it.

Ask me the question can bad links hurt a good site? I’ll tell you yes

Ask me what you can do about it? I’ll say I don’t know

Ask me whether you should be buying spam links for your competitors? I’ll tell you that’s up to your conscience!

This is a sneaky little trick more than a hard and fast strategy but I explained it to some SEO people the other day and they found it interesting so I thought I’d turn it into a post.

Here’s the problem- one of the best ways to beat your competitors rankings is to mine their backlinks and copy all the open links. Unfortunately that means your own backlink profile is vulnerable to getting mined.

Now the best way around this is to create a backlink profile which can’t be easily replicated. But realistically parts of your profile like your directory listings, forums you participate in and associations you’re a member of are essentially FFA links for anyone with the time and possibly money to pursue them.

There’s plenty of link data sources out there – yahoo, linkscape, majestic seo etc and there’s no way to block these web services displaying your link data so instead we’re going to rely on good old fashioned user error.

Here’s what we do:

Step 1: Set a trap for your competitors

First you’re going to need to uncover the IP addresses of your competitors and/ or their SEO agencies. This is by far the hardest part of this process but it’s absolutely worth doing anyway because this is valuable information to have at your disposal for other reasons which I won’t go into here.

The trick is to lure the person responsible for SEO in the company onto a private page then store the details of their machine. Some creativity may be required here but to give you an example I used:

1. Create a directory for whatever niche you’re in. Let’s say you’re a florist so you setup a little localflorists.biz directory of flower shops.
2. Stick a page on that domain which will record the visitor IP. Let’s say you create a page at localflorists.biz/mycompetitorsname.html. The trick is you want a page that only the person from the company will ever land on so block it from search engines and don’t link it from the main site.
3. Now you put in a call to your competitor saying you want to give them a free listing in your directory and want to speak to the person who deals with their SEO. Get an email address in the SEO team.
4. Send an email to the competitor saying you’ve listed their company for free in your high PR florists directory and they just need to visit the page and update the company links section and make the page live. Include the link to the private page localflorists.biz/mycompetitorsname.html
5. Now it doesn’t really matter what happens on that page, the important thing is that they were curious enough to click the link and have a look.
6. Once they’ve visited the page you can check the visitor log files for the page mycompetitorsname.html and hey presto you’ve got their IP location.
7. Repeat for other competitors

Step 2: Creating a shadow site

OK so with that data you can do all kinds of interesting and fun things to throw your competitors SEO teams off the scent like removing page titles or displaying the whole site in an inline frame! For now though what we want to do is point the competitors to an almost identical version of your website to the one you show everyone else. We’ll call this a ’shadow site’ for lack of a better name.

Chances are you’ve got a bunch of domain names which are very similar to your main domain. For example I might have johnmcelborough.com and johnmcelborough.co.uk or john-mcelborough.com or even misspellings like jonmcelborough.com. For our purposes here the more subtle the difference the better so the mispelling is probably the least obvious. For this example I’ll use jonmcelborough.com as my shadow site.

Now use a mod_proxy to map johnmcelborough.com to jonmcelborough.com so the same content is displayed at both, the only difference is that little mispelling in the domain.

Use an X-robots header to stop search engines indexing all pages except the homepage of your shadow site to minimise duplication.

Step 3: Build some garbage links

Next you want to create a convincing looking but complete garbage backlink profile for your misspelt domain. This is easy just employ the first few companies which come up for these searches:

Bulk directory submission service
Dofollow blog commenting service
Dofollow social bookmarking service

Step 4: Cloak your competitors to the shadow site

Now cloaking search engine bots can get you into trouble but cloaking individual human users shouldn’t be a problem. What you need to do is create a little rewrite rule which will lookup a visitor IP, check it against your list of competitor IP addresses and if a match is found 301 redirect to the misspelt domain. If no match is found i.e. it’s a normal user or a search engine, do nothing, show them the real domain as usual.

Here’s what we hope will happen…

• Your competitor Google’s your main keyword and you rank #1 in Google (of course).
• They lick their lips and say something like ’sweeeet, I’ll have a look at their backlinks’
• They click on the google result and while they look away for a millisecond to pick their nose your rewrite sends them to the misspelt domain.
• They scan around for a minute then either use their SEO toolbar to run off a link report or copy and paste the domain from the address bar into their backlink analysis tool (yse, ose, majestic etc)
• They literally salivate with excitement as they see hundreds of free web directories, terrible paid links, social bookmarks and blog comments which they can spend hours replicating.
• They download the link data and occupy themselves with reverse engineering the crap link profile you created for your misspelt domain.
• Meanwhile the kickass links which are actually making your real domain rank remain elusive!

You’ll note there’s a lot of assumptions here. We’re assuming your competitor won’t notice the misspelling in the domain or will think it’s a recent change. We’re assuming they’ll be running backlink analysis manually rather than using a tool like Raven’s site explorer which scrapes the urls straight from the google results page or that they won’t copy the URL from Google rather than visiting the site first.

Also there’s going to be virtually no way to know if it’s worked or not. None the less, if you’re in a competitive industry and are defending valuable rankings this kind of tactic becomes a serious option.

If you’re an SEO agency, have a look at your own process for backlink analysis – do you think your link builders would fall for this cunning ploy?!

Ps – I deliberately haven’t included any code samples in this post. If you don’t know what you’re doing with this stuff you shouldn’t be attempting it- I wouldn’t myself without the supervision of a more experienced developer!

This is a follow up to my post on building lots of links to ecommerce product pages. Someone commented on that post something like:

Since no one is linking to these FFA pages they will have a “low” page rank value with a small page rank voting share to distribute.

Because I don’t get many comments here I have the luxury of being able to answer this in great depth!

To understand why the link building techniques for product pages I discussed in that post do help your product pages rank you first need an understanding of the challenges of ecommerce SEO.

For many owners of online shops the following is true:

• Many products are buried deep in the sites architecture, meaning they get no ‘PageRank’ passed from high value pages like the homepage and top level categories- pages which don’t look important to the site might end up in the supplemental index and are unlikely to be returned in SERPs
• Product pages are unlikely to attract links from external sources naturally
• Product pages might use manufacturers descriptions or have no textual description at all meaning they are light on content at best or duplicated on other sites at worst.
• Lots of products will have near identical descriptions, names, prices and attributes meaning they look like internal duplicates of other products making it likely they’ll be treated as supplemental results

In this post I’ll share some ideas for optimising your product pages to eradicate some or all of the problems above. I want this blog to actually add value and not just recite best practice so some of the stuff I talk about here might not be what your agency recommends, but it will work if you know how to do it properly.

Step 1: Making your product pages (look) unique

There’s a few techniques you can use here. As a general rule (this is true with most stuff in SEO, and life in general I tend to find) the more time and money you invest in your product pages and producing content the better these pages will perform.

Writing unique thick content product pages

The best thing you can do to optimise your product pages is to write your own, unique descriptions of each and every product. Now if you’ve got hundreds or even thousands of products this is the route I’d go down. You don’t need to launch your site with this content, you can release it after launch and add it in order of product priority (your most important/ profitable products get unique copy first).

You could outsource this job but it’s better managed by your product teams or someone with actual product knowledge- if you’ve got a bricks and mortar store your shop floor sales people are best placed for this as they talk about your products all day so try and work out a way to put them on the job.

You can treat this exercise in a number of ways:

1. do it just for the basic SEO benefit of having unique content on the page. If this is the case you might want to display the content further down the page where your visitors are less likely to dwell on it. You can then keep the manufacturer description at the top of the page, selling the product.
2. use the copy to introduce new keyword variations for the product. Aim to get as many words of unique content on the page as possible without negatively impacting conversion rates. Crude as it sounds the more words you have on product pages the more likely they are to drive long tail search traffic.
3. use the copy as a sales device and write really useful product descriptions which not only talk about the product but actually sell it’s benefits. Like a sales person would on the shop floor.

Needless to say I’d advocate the last approach as its by far the best from an SEO, usability, conversion rate and general ‘making a good website’ point of view. For example check out the awesome thick content product page on Firebox for the musucbag.

Like I said adopting unique content production for each product is the best tactic, but it doesn’t scale up well if you have tens or hundreds of thousands of products and variations. While it’s still worth adding sales copy to your key products some of the following methods will help you to dynamically create ‘uniqueness’ on other pages which are either in the queue for the copywriting treatment or which you don’t see as important enough to invest time in (i.e low profit margin/ loss-leading products)

User reviews

It you can get users to review products on your site they are doing you a massive favour because they’re adding unique content to your product pages. Amazon do an amazing job of this, have a look at the detail added to this product page on Amazon by the user generated reviews. Having review options on every page of your site and no actual reviews looks pretty crap though so don’t just add the functionality and hope people will bother to take the time to add their reviews. I would always recommend incentivising a review with reward points, a money off coupon or a prize draw. Email your customers a few days after the product has arrived offering the incentive in exchange for the review.

Related products

Dynamically adding links and text snippets from related products or ‘customers who brought this also brought’ type widgets doesn’t really create unique content because the snippets are taken from other product descriptions but when the content is mashed together the % of duplication between similar product pages should be reduced (assuming the same related products aren’t displayed on each product page).

Boilerplate or macro content

I don’t really recommend this but it does work. You create a generic description and dynamically insert the unique attributes for each product into it to create some semi unique copy. You can use synonyms to randomise the copy further as well. So your macro template might look like this:

The [product name] is available from [store name] at the [bargain|low|cheap|great] price of [product price]. You can buy the [product name] in [colour options]. Select the quantity of [product name] you require and add to basket to [checkout|pay|purchase] securely online.

I.e. for a tv you might have you might have:

The Sony Bravia plasma is available from cheaptvs.co.uk at the bargain price of £499.95. You can buy the Sony Bravia plasma in black, White or grey. Select the quantity of Sony Bravia plasma’s you require and add to basket to purchase securely online.

Obviously the more product attributes you have the more you can personalise each piece of text. You probably wouldn’t want to use this on high value products like TV’s but for something low value where you’ve got stacks of products (like TV cables) it could work. Seatwave use templated copy like this at the bottom of their event pages (just after the table of tickets).

i.e.

Love Never Dies market information on 05 July 2010
Seatwave has 50 Love Never Dies performances with 2664 tickets in 1 cities across 1 countries.
At present you can buy or sell Love Never Dies tickets for performances in London.
There are currently dates set at the following venues: Adelphi Theatre.
The current lowest priced tickets are £34.42 for the performance on 07 July 2010 at Adelphi Theatre in London, UK.
The performance with the most tickets available is Adelphi Theatre on 31 July 2010.
The performance with the least tickets currently available is Adelphi Theatre on 05 July 2010.

The orange stuff is dynamic content, unique on each page- the rest is boilerplate

Make use of those meta keywords

Think meta keywords are defunct? Well you’re right. However we might be able to find a use for them… I had a client who had spent quite a bit of time and money writing meta keywords for each product in their catalogue (this was a few years ago but still probably pointless lol). They’d done a good job of it though, researching keyword variations for each product. Sitting in the meta data of the page they were doing nothing so we moved them into the page itself and output them under a ‘tags’ section at the bottom of the page. This is a cheap tactic but I have to say it worked great.

Homebase do something remarkably similar at the bottom of their category pages:

Image optimisation

The cool thing about images is they’re free from duplicate content issues so if you’ve got 10 photos of a product it doesn’t matter they’re the same photos every other shop has. Make the file names unique on each page (even if you’re using the same images on different pages) and dynamically build unique alt text using the product name so for example:

Image 1 - file: images/[product name]-1.jpg, alt: “[product name] main view”

Image 2- file: images/[product name]-2.jpg, alt: “close up of [product name]

Syndicated content

Now from a conversion marketing point of view linking off site probably isn’t a great idea but this is worth mentioning for SEO. One way to distinguish similar pages from each other is to feed in content from external sources. This of course can lead to a different and potentially more serious type of duplication but if it’s implemented correctly it can work.

The idea here is to start by gathering a bunch of open data sources which you can (legally) scrape into an RSS feed. These could be blogs, tweets, news results, yahoo answers, forums, where-ever the products which you’re selling will be mentioned.

Next you’ll need to feed in relevant items to the product name onto the page. You’ll only want to publish an excerpt of the article to avoid being seen as a content scraper. What this might actually look like on your page would be something like a sidebar module with latest news about your product so on your Sony Bravia page you’d have an excerpt from the latest news stories, blogs or forum posts which have mentioned Sony Bravia tv’s.

If you think about those twitter plugins you see on blogs which pull the latest tweets with a certain hashtag onto the blog its the same principle.

Step 2: increasing internal link weight to product pages

Ok so if you followed one or more of the content optimisation recommendations your product pages should be looking more like the kind of unique, content rich documents which Google likes to return in search results. However the pages are still buried deep in your site so they don’t look particularly important. If a product page is a long way from the homepage and doesn’t have any external links pointing to it it’s unlikely to rank well unless you have a really strong site.

For small sites sculpting your PageRank with internal linking is easy, you just link to your important deep pages from your homepage. With ecommerce it’s different. You can’t list 50,000 product page links on your homepage and if you did it wouldn’t help your rankings much anyway. This is a problem best tackled at the sites IA planning stage but there’s a few modifications and quick fixes you can make on live sites too.

Flattening the information architecture

If you have a big site with lots of products chances are your IA is a few levels deep like:

Home > Electronics > TV’s > Plasma TV’s > Sony Bravia 32″

At best this product page is going to be buried 5 levels deep, that’s assuming there’s no pagination of the category pages.

The quickest win for your most important products is featuring them on the homepage or section homepage like in the example above the Sony Bravia would be a featured product on the ‘electronics’ homepage.

But that only takes care of a few individual products. To make site wide improvements to your product pages you need to raise the deep categories up the IA so plasma TV’s can be reached from a number of places like:

Home > Plasma TV’s > Sony Bravia 32″

Or

Home > Electronics > Plasma TV’s > Sony Bravia 32″

Doing this your Sony bravia’s now only a couple of levels from the homepage, so are all the other products in the plasma TV category.

It might mean listing more bottom category levels on your homepage than you’d like but this is usually the quickest fix for big ecommerce sites. Conventional wisdom may disagree here but realistically a trusted site can get away with 200-300 links on your homepage before you start really diluting things so it’s just up to your designer to make all those links fit onto the page in a user friendly way. Have a look at how Asos manage this with their drop down navigation listing lots of deep categories.

Brand pages

Another quick way to flatten the IA is to create brand category pages (if you haven’t already got them) so:

Home > Shop by brand > Sony > Sony Bravia 32″

Or even

Home > Sony > Sony Bravia 32″

This is an additional internal link to each of your product pages which is good news in theory. However be careful creating new category pages because the more you have the more you dilute the others so you’ll need to be building external links directly into the brand category pages in the example above in order to keep things balanced.

Related links

These were already discussed in the content optimisation section but if you missed it there, these are the product links at the bottom of product pages which take you to related products. They’re useful for spreading link equity horizontally across the site architecture (where as the category page examples spread link equity from the top down). For this to be effective though you really also need to be building external links to product pages, else you’re just batting around worthless links between your products.

In-content links

If your sites got a blog, ‘how to’ or other ‘thick content’ area you can add deep links to product pages from within these content pages, either embedded into the text or using a related content widget in the sidebar. Nice example of this on the topshop blog:

Subdomain expansion

This borders between internal and external link building. It also borders on ‘don’t try it unless you’re sure you know what you’re doing’. I’m assuming you do…

One way to flatten your site is to use subdomains for categories or sub-categories so you might have a homepage either in addition to your plasmatvs.mysite.commysite.com/plasmatvs category or instead of it.

My personal preference would be to host category and product pages as normal on the main domain so:

• Plasma tvs category at mysite.com/plasmatvs
• Product page at mysite.com/plasmatvs/sony-bravia-32

But also

• Host a category microsite at plasmatvs.mysite.com which links back to the product page on the root domain.

You obviously make more work for yourself by doing this because you need to promote the subdomain with external links but if you have the resources I think it’s worth exploring. You could also create brand subdomains like sony.mysite.com.

Step 3: building external links

This is the final and also hardest part of the process. Internal linking relies on your homepage having some authority in the first place but you also need external links to category and sub-categories as well as direct to product pages to ensure there’s enough link equity in your site to spread around all your product pages.

Link building to category pages

I’ll usually look to link build to deep categories first as these have the closest proximity to my product pages so backing up to the example in the last section I’d favour link building to the plasma TV’s sub-category than to the electronics category page. Not only will these links pass more weight direct to the products, they should also be easier to source because the more niche you go the more link building opportunities there are. For example you could link to your plasma’s category page from specialist forums, geek blogs, tv show fan pages, sports sites, home improvement magazines etc etc

I’ll give a few practical tips for link building to category pages, because I’m nice like that!

• Botw accepts deep page links in it’s directory- a quality link you can buy, permanently, and within guidelines- snap that mother up for as many categories as you can afford
• Ditto many other shopping directories who’ll let you submit category level pages
• Place some advertorial with the authority magazine for each of your categories so for your laptops category take an ad on laptopmag.com
• Guest post on niche blogs for each of your subcategories or if these are hard to find search for blogs with an audience/ customer overlap so you could write a post on a gaming blog about the best plasma’s for multi-player mario kart on the Wii then link to your plasma’s category in your byline.
• Buy up niche sites and keyword rich domains and 301 them to your category page
• Try for links from manufacturers, suppliers or support contract providers within each category
• Mine the links of specialists in each category so look at the backlinks of www.directtvs.co.uk and try to replicate these for your category page at www.mysite.com/plasmatvs
• Run a giveaway/ voucher code promo through niche blogs. So for example contact bloggers who write about BBQ’ing and give their readers a 10% off code for anything in the BBQ’s category.
  

Unlike product pages the game with category pages is about getting some decent links, not volume. Most stores have few category page links so a few go a long way. As old school as it seems I think good old fashioned PageRank is as a guide as any when assessing link prospects for category pages.

Building links to product pages

Not going to cover this again here. If you’ve reached the end of this post (sorry, I know it’s long) and haven’t read my big post on building links to product pages that will be another 10 minutes of your life spent on this blog. Hopefully it will be worth it though.

That’s a wrap. Just a reminder at the end that some of the tactics I’ve discussed here might not fall into the best practice handbook. Make sure you know what you’re doing before embarking on any of the more advanced stuff- if you’ve got questions, just ask…

Some thoughts on the Google takeover of ITA and what it means for travel SEO

For those who didn’t know last week (July 1st 2010) Google formally announced their buyout of ITA software, a company who provide the technology which powers sites like Kayak.com and Farecompare.

I don’t usually post search news here but this story was of interest for me as I’m involved with a number of travel SEO projects.

Now Google buying out software companies barely makes the news these days it happens so often and at $700 million dollars they got a bargain with this one. But what do Google want with flight search technology?

Although plenty of people search for flights on Google the application of search data goes beyond the most obvious way they’ll integrate the technology – displaying fares and date search boxes in the SERPS when users search for stuff like ‘cheap flights to Spain’.

This software could be used across any number of search verticals. Hotel searches on Google maps could use the technology to display fares alongside results so if I’m in London and search for ‘hotels in new york’ on Google maps they display a link or a search box to the cheapest fare from London to New York.

It could be integrated with Adwords so airlines and travel agents could add live pricing to their Ads directly from the system.

What’s most likely though is that Google will just build their own comparison site and work direct with airlines on an affiliate basis to sell their tickets directly. The Google press release on ITA deal says:

The acquisition will benefit passengers, airlines and online travel agencies by making it easier for users to comparison shop for flights and airfares and by driving more potential customers to airlines’ and online travel agencies’ websites. Google won’t be setting airfare prices and has no plans to sell airline tickets to consumers.

To me this implies Google is looking to sideswipe the comparison engines and drive traffic straight to airlines. Where does this leave the likes of Kayak, Travel Supermarket and cheapflights.co.uk?

Firstly I don’t think anythings going to be happening overnight but if I were them I’d be pretty nervous about the deal. These sites rely heavily on traffic from Google (from SEO and Adwords) making Google a dangerous competitor. Top rankings for flight related keywords will be devalued massively if Google lump a big search box at the top of the results page. Route based searches like ‘KL to Singapore flights’ will likely be the hardest hit because Google will not only be able to steal the top spot, they’ll also have the technology to make the other comparison engines irrelevant by answering the query right there on the results page.

If I were in the flight comparison business I think I’d be looking at ways to shift my SEO strategy away from the reliance on these goldmine keywords and focusing on related searches and the long tail with stuff like destination based content or events listings which Google is less likely to muscle in on. I’d also seriously consider a decent affiliate scheme for flight comparison, at the moment its difficult for affiliates to earn money on flights because the margins are so low. I used the cheapflight.co.uk affiliate program a while back and ended up switching it off because I was earning less than I was on Adsense.

What do you think of the deal?

If it’s your first time here you might want to read this post on link building for ecommerce sites first.

I recently started reading the Hobo web blog (which is really good for practical SEO advice by the way) and this post caught my attention. Shaun’s advice on using article marketing to build links with keywords which you’re already ranking for is similar to the way I use articles to boost up the rankings of product pages (or any deep page on a big site really). So I wanted to share this tip and some others for how ecommerce site owners can effectively build lots of links to lots of pages, ideally without lots of work.

To understand how and why this works you first need to accept 2 of the less spoken about but absolutely fundamental underlying principles of SEO.

1. Not every link you build has to be grade A, 100% clean, natural link building gold like many SEO’s would have you believe. Bulk links still play a quite significant part, in fact I’d go so far as to say the majority of links on the web are pretty crappy. Also while I don’t buy into the ‘bad links can’t hurt a good site’ idea because they definitely can, the stuff I’m discussing here won’t have any negative side effects.
2. The majority of product pages on ecommerce sites have no external links, they’re ranked on the basis of, predominantly,  domain authority instead. It’s usually only after a product page is ranked at number 1 it acquires natural links as anyone talking about that product online in forums or on blogs will just paste the top result in Google into their posts as a reference. For this reason, provided you’re doing the other stuff right, have built a domain with some trust from decent links to top level pages, you’ve got your technical stuff right so there’s no duplication issues etc and your internal linking makes product pages easy to spider then a few links even of low quality to product pages usually has excellent effects on rankings.

OK if you’re down with those ideas, here’s how we go about building these deep links. In no particular order…

Article syndication

I still use various types of ‘article marketing’ quite a lot as part of the overall link building mix. The type of article marketing I’m talking about here is pretty old school. Start with a list of your products and associated keywords, group them into pairs in a logical fashion. Most article sites will let you include 2 attribution links in each article.

Next get busy brainstorming a big list of article titles which are loosely associated with each product/ keyword pair. Get the product name in the title some how if you can but don’t be too pedantic. For example if your product name is ‘Sony RX-142 recordable DVD player’ an article title like ‘how to choose a good recordable DVD player’ is perfect. Don’t use brand names in your articles or titles as some article syndication services won’t accept them.

You’re going to need lots of article titles here, possibly hundreds so it’s a big job. The easiest way to do it is to search some of the more popular article directories like ezine or go articles for your base keywords (like recordable DVD player) and see what titles come up (like this), then just use these or adapt them slightly. There’s no harm in this kind of light plagiarism in my eyes anyway, your articles going to be 100% unique even if your titles not.

Next the real work starts as you need to produce all the articles you’ve just brainstormed. They need to be 500 words each, in reasonable English and unique. I don’t use article spinners or anything like that as these articles rarely get published by article directories.

Probably needless to say I wouldn’t recommend writing all these articles yourself. You should be able to farm the job out fairly cheaply to English speaking writers. I use text broker mostly and pay around $5 for a very basic article. You can find cheaper but I’ve never seen a legible article for less than $5.

Once the articles are written, add author attributions to the end which include links to the 2 product pages you assigned to that article. I also like to include a unique reference number in the attribution link so I can easily track where the article has been picked up and where my links are.

With all your articles written you’ll need a distribution plan. You don’t want to just ping out hundreds of articles at once. I tend to use content crooner for this (used to be article marketer- I have no idea why they changed the name!!). I like this service because they have reasonably strict editorial guidelines which are actually enforced meaning they don’t release spam or spinner content. Because of this articles you syndicate through them get picked up by more sites, meaning more links. Also the most important thing the system let’s you do is schedule articles for release meaning you can upload all your articles for approval in one hit then just sit them in a queue to be released every couple of days to give you a steady link growth. The only other syndication I’ve used recently is iSnare. I still use this a bit to mix things up but the distribution doesn’t seem as good and the approval time is long, like 2 weeks or more where as content crooner is more like 2 days.

Ezine articles while not technically a syndication service in the same way works well too as these articles get picked up by plenty of other sites. If you’re doing lots of articles you might want to use a few distribution services.

It’s important to track the links you pick up from this technique (and every other technique I talk about here). Things change in SEO and I wouldn’t expect this technique to work forever so keep tabs on a) where your articles are getting posted and b) the affect new article links are having on the rankings of your product pages.

Social bookmarking

This ones a bit easier as there’s no content production required. It’s less effective, especially when used in isolation but still has a positive overall impact by increasing the volume of links to deep pages.

I’m assuming you know what social bookmarking’s all about and basically how it works. If not read this.

If you’re using this technique I’d tread lightly and just add a few bookmarks for each product page, not hundreds. Mix up bookmarking services which follow links with those that use nofollow to make your link profile look a bit more natural.

There’s plenty of companies offering social bookmarking services. SocialMaximizer is a pretty good one with decent management facilities if you’re making big orders. Paying per bookmark can get expensive though when you’re looking at thousands of bookmarks, especially as most of these will never get indexed.

The alternative is using software. Bookmarking demon looks spammy but works pretty well. I’m dubious of anything fully automated when it comes to link building but this is one technique where I don’t think you gain anything by doing it the manual (hard) way rather than the automated (easy) way.

These links will probably only help you if your sites well established already or if you’re targeting very uncompetitive product terms but they do help. Again this technique is probably time limited and won’t work very well in a few years so don’t rely too heavily on it as part of your link building strategy.

Blogs and social hubs

This is the most involved technique I use for this type of link building but it’s also the most lucrative.

The idea is to maintain a mini-network of your own blog sites and pages on social sites like hubpages and squidoo which you can use to freely add links back to your product pages on your main site.

I came up with this idea a few years ago when people started selling ‘link wheel creation’ services. Although link wheels are garbage and I wouldn’t use these services the concept it fairly sound. If you control a lot of domains, even if these don’t have a great amount of trust with Google in their own right the fact that you can freely add deep links to any page you choose gives you a huge advantage over other techniques which rely on other webmasters to play ball and publish your article or get your bookmark page indexed.

Now to be clear I’m not advocating the use of a full scale link pyramid here. You should create unique content for every site you setup, keep it relevant and make it useful.

Let me illustrate with an example.

You’re an electronics retailer and one of your product categories is TV’s so you setup a blog, on a domain you own, let’s say www.myworldoftvs.com, use something like wordpress to manage it. Get your writers to produce a bunch of posts about TV’s. You can have fun with this and do some link bait type stuff which might do well on digg or stumbleupon. Or just write some useful how-to articles.

You’ll need each property to stand up on it’s own feet on the search engines so you can get it indexed with some directory submission, do some article marketing, buy some sponsored reviews or even setup a little link swap page. Just get the site on the radar.

Now you have a site which you can add as many links on as you like to your product pages. Either add the links in the sidebar or in the content of your posts or even just create some big list pages.

Once again these aren’t the best quality links in their own right but they are

a) from a relevant site

b) optimised for anchor text

c) directed at a deep product page.

I said at the start of this section you could also use social sites like hubpages or squidoo to do something similar. Setting up blogs on wordpress mu sites like wordpress.com or on blogspots is another option. The benefit of this is you might get some authority passed down from the root domain making your pages more powerful. Its also cheaper as there’s no domain and hosting costs. The big disadvantage, and why I wouldn’t recommend doing it this way on any great scale is your content could get removed or changed at any time by the site owners. As far as possible I’d choose to keep this technique on my own sites.

Twitter and facebook

Yes I know these links aren’t followed but bare with me a moment…

One of the ideas I’ve had in my head for a long time and I think seems to work, although I can’t prove it in any real way, is that Google is using a new (or probably not so new) scoring system around popularity signals. This is seperate to trust or relevance and doesn’t necessarily rely on good old fashioned followed text links so anything from tweets to facebook likes to unlinked mentions to social bookmarks to Analytics data to search query data and beyond is all contributing to some rankings, although conventional SEO wisdom may say otherwise.

Some of this stuff has been formally or informally backed up by Google like the introduction of real time search which shows just how much of an active interest Google has in Facebook, Twitter and the real time web in general.

Anyway if you buy into my theory read on, if not you can skip this bit.

I don’t have a real process for building links on these sites although anything you do on facebook and Twitter needs to start by building up a decent following. On Twitter that just means getting followers and people who will be willing to retweet your links. On Facebook that means getting a popular page with lots of fans likes so when you push a link out it gets a wide distribution through news feeds.

Once you have that sort of following it’s pretty easy to use your social presence to push out your deep links as long as you do it in a useful interactive way so don’t just drop a link like…

‘hey guys, buy this recordable DVD player – now’

but more like…

‘hey guys, Sony have just released this new recordable DVD player. What do you think of the touch screen function. Let us know’

Ok I’ve never owned a recordable DVD player but you get the idea right!

Affiliate links

I’ve talked about this a few times before and this article on making affiliate links SEO friendly is still relevant.

Now I don’t recommend this as much as I used to because I think affiliate marketing is too important a channel for ecommerce businesses to let SEO have a baring on it. However there’s plenty of good reasons to run your own affiliate program if you’ve got a strong brand which affiliates will want to work with. Most importantly you don’t have to pay commission to a network meaning you can offer a higher payout to your affiliates (or just take a bigger profit). So if you’re looking at managing your affiliates through an in-house program anyway, then you should definitely look at potential SEO benefits which you can take from it.

Systems like idevaffiliate and post affiliate pro will let you setup your affiliate links in an SEO friendly way, meaning when an affiliate links to you, providing they don’t nofollow the link or pass it through an adserver (both of which are likely unfortunately) you will get some benefit from the link just as if it was a normal organic link.

Because affiliates are also inclined to link to product pages this becomes a great way to get deep links- just look at amazon.

You can also offer widget style ad units to your affiliates which contain a block of products with deep links. If you build widgets for popular cms’s like wordpress and joomla to work with your affiliate program the affiliates are more likely to leave the links as you designed them, rather than going to the trouble of messing around with the widget code to add nofollows to the links. If there’s one thing you can rely on with affiliates it’s that they’ll take the easiest and therefore most profitable option available.

Although Google have kinda changed there tune on affiliate links now and said they should be treated as paid links I’ve seen no evidence of this being enforced on ecommerce affiliate programs which are setup without an obvious footprint. Whenever I talk about SEO friendly affiliate links I reference the botw.org affiliate program because it does a great job of implementing affiliate links in a way that they can pass PageRank without leaving a footprint. It also gives a number of options for how affiliates can link which is important to making your program a success. For example both the links below are affiliate links (as is the one above) they both also pass PageRank from my site to BOTW:

• http://www.botw.org
• http://botw.org?uid=34022