the dark art of link building

I understand why link building is so crucial; it gives Google an idea of how important a page is with regards to the number, quality and context of links that are linking to a the site.
However I still find it a bit of a dark art; why would people want to link to you? how do you keep it ongoing? Reciprocal vs one-way links?

In order to demystify it, here are some key points;
1. have the right ratio of qualitative vs quantitative links. You do need a lot of links, but err on the side of fewer links from sites with some sort of authority (Google pagerank of 6 and 1,000 inlinks). Ideally these should be within your industry; so if it’s a cooking site don’t link from a car site.
2. Wikipedia is great. If you can link to your site from Wikipedia/write an article about your industry- do so. Google gives Wikipedia a lot of clout (the second click phenomena)
3. Good anchor text– ideally someone should link to your site and the text next to the link (underlined) should say “cooking tips” (or whatever you want to optimize for), rather than saying e.g. click here. This gives Google some context about what the site is about if the link has a description.
4. don’t limit yourself; put relevant links back from blogs, forums, video sites, anywhere where there is a “follow” on the link (which means that Google will give it credence)
5. PR is a great way to get more links. News sites are usually authority sites, if you can get a well optimized release on to a news site it will help.
6. Link bait is an interesting opportunity. Do a zany stunt to get people interested and linking to you. once the newspapers start linking to you, sit bank and watch your pagerank rise.
7. Be persistent. Get a couple of interns in, and get them to get a spreadsheet going with prospects. This involves going through the sites that link to sites that are optimized for the keywords that you want to optimize for. put them all on a list and send them a personalized e-mail. be persistent with unanswered mails and buy your interns lunch every now and again. This is a fairly monotonous task.
8. Don’t buy links- it’ll cost you and it isn’t a sustainable strategy
9. The best way is to have good content that people want to link to. Sort of a no-brainer, but this is the quickest, most effective and guaranteed return method of getting links.

These are some tips to improve your links (concentrate on the bottom one and you’ll go far). Get busy.
Related articles by Zemanta

Your site is optimized, now what?

Michael JacksonMichael Jackson (via

So you’ve done it. The checklist is done;
1. site is in all the directories.
2. Everything on the page is looking good; canonical domains, H1’s et al ()
3. you’ve got a load of inlinks

time to sit back and watch the traffic pour in, right? Nope, now its time to start optimizing. The first thing you need to do get a good perspective of all the possible words in your area, and the potential volume on them. Put yourself in the shoes of your audience and look at what terms they would be using. Google Insights and your own PPC search campaigns can help in determining volumes.

Once you have the list you need to start prioritizing;
1) what is the volume of searches on each term
2) how much competition is on each term
3) what data do you have from PPC to assist (CTR and conversion)

The ideal scenario is that you find words with a hell of a lot of people looking for them, which no one else is optimized for and you know that people will buy loads from you as a result of it. Lets assume you have a dental health website. All of a sudden Michael Jackson sings the praises of plastic toothpicks and it becomes a hot trend. The searches for “plastic toothpick” are going crazy on your paid search campaign, and you can see them going off the chart in Google Insights too. With each successive picture of MJ sucking a toothpick, more and more people are looking for this must-have summer accessory. You would want your product page for plastic toothpicks to rank well for this, because you know (from your PPC campaigns) that people who search for plastic toothpicks usually end up buying them, and you make a fortune on each toothpick

Start tweaking your site accordingly. You’ll need to change your domain structure to account for this, e.g. create a page that has plastic toothpicks in the domain name, as close as possible to the root (e.g. or even better, buy and put up a site there. Ensure that the content that you write has a high keyword density around plastic toothpicks, this means that you have a couple of paragraphs that talk exclusively about the teeth cleaning benefits of these toothpicks and mention the words “plastic toothpick” as often as possible. Then you need to get some links into the domain and you will be printing money. Show celebrity magazines where to buy these lovely toothpicks, and ask people to link to your cool content. Link building for this page is not too difficult when it is within your own domain, because your other links will do the heavy lifting.

Now you’ve increased traffic to your site of more relevant people who spend lots. Just hope that Michael doesn’t bring out his own brand of floss.

search engine DE-optimisation

MOUTAIN VIEW, CA - MAY 4:  Employees of Google...Image by Getty Images via Daylife

It seems like a stupid question when you think of it. If you want to optimize your site for search engines, you would always want to end up on top, wouldn’t you? That’s where all the traffic is, and more traffic equals more revenues. Some statistics I read say that position 1 = 56%, position 2=13%, position 3= 10% and position 4= 4% of search volume, which means that there is a huge premium on being number 1.

But then I think of my own search behaviour, if I search for something you need to acquaint yourself with the data a little better before you can make a purchase decision. So, the first click tells you about the space, the second click gives you an idea about competition and by the third click you’re ready to convert. I don’t think it’s as simplistic as that (and I’m sure this varies significantly within industries), but it does take into account that you shouldn’t only steer on traffic, you should look at your conversions too.

To add a further layer of complexity it depends on the nature of the search how exhaustive the search will be, and which position is the best to aim for. For navigational searches its fairly immaterial, if someone is looking for you they will find you. For informational searches top 3 is good, depending on how high involvement the issue is (buying a car would be high involvement and would require a lot of searching around a topic). With transactional I’d go with my advice from above.

I’d love to know if anyone has experience with de-optimising their website to get a lower ranking and more conversions

6 tips for using Google Insight

Search-Engine-MarketingImage by Danard Vincente via Flickr Google Insights for search has just been released, offering a whole wealth of data that previously was not available. For your SEO and PPC campaigns there are a lot of words that can be targeted based on this. However if you are looking at this from an industry perspective and want to gauge yourself and how you rank it is also extremely valuable. Having played around with it a little recently these are some skews I would advise;
1) Look at your category
It may be difficult to find the exact niche that you inhabit, but drill down to as close to it as you can. What is the trend in terms of searches, is it seasonal, is their growth in the category, what are the top keywords. Then look how much share you get proportionally of the big terms, this is illuminating because it gives you an idea of the size of the opportunity, can you optimize your page better, should you bid more on this area.

2) look at your brand
how have the searches for your brand been performing (as marketing person you hope that your brand becomes even better known). Can you tie in bigger peaks in spend with promotions that you undertook. Can you see what is working to get your name out, and what isn’t. Are there cycles on how people look (time of the year/month/week), is there some way that you can optimize based on this (this ties into how you manage your campaigns and the information you have about customer behaviour- day/week parting, bidding more round payday etc)

3) Look at different geographies
Start off on worldwide for all your searches and then split them into different countries. Look at these and see if you have some large market share in an unknown geo (that’s the joy of having an online business, no borders). Is your site big in Norway, why is your site big in Norway, should you start adding localized payment systems and structures? This can isolate areas where you should focus, similarly it may outline areas that you’re underperforming.(as this is based on IP address I wouldn’t take the extremely local/city perspective too seriously)

4) major searches
look at what the core search terms are around your brand and in your category. Are you bidding on them, maybe if you drill further into this they will give you some interesting phenomena. E.g. marketing in other languages you may not know what the up-to-date slang terms are and these may be reflected in the upcoming terms.

5) segment over time
this I think is one of the more interesting functions. You can find out how things developed with your brand/category/search term. How the geographic makeup changes over time, and what happens with related terms. If you have made specific promotions around a time, how did people react to it, did they search for you for terms related to it. Where do you experience the most organic growth, also look at your share of the search, how does this change over time. Looking at the matrix of marketing spend, competition and time and try to figure out how it fits together.

6) competitors popular search terms
are a mine of information. What is their positioning, what are the major terms that people are searching for around their brand, what should you be bidding on, what are the major traffic drivers from a seo perspective. See what share they have of the top terms. In most industries/categories there are a few key terms which drive the majority of the traffic (good old 80/20 rule). We call these the k10 (for our top 10 keywords). Who owns the largest share of these and how does it change.

These are a couple of things that you can do with Google Insight that should give some more beef to understanding your market and where you sit within itRelated articles by Zemanta

Reblog this post [with Zemanta]

search update

SUNNYVALE, CA - JULY 18, 2006:  (FILE) A sign ...Image by Getty Images via Daylife
I saw a couple of interesting search things this weekend

– whilst reading a blog favourite of mine I saw that Microsoft have started actively started advertising that you can get cash back by using their search engine. This is a pretty aggressive way of gaining market share, not only are they paying people a portion of their money to purchase through their engine (a whiff of desperation perchance?), but they’re then spending even more on acquiring customers through it. Its still early days, but it will be difficult for them to change this inbuilt consumer behavior that Google is the default (Google’s US market share in the US is growing and in western Europe its almost insurmountable, whereas MSN/Longhorn has as yet only launched in France and the UK).

Microsoft also acquired Powerset this week which may be another arrow in the quiver in developing a better search engine. Powerset is focused on developing a more intuitive way of you searching the web, which mirrors natural language. This means you don’t have to search in a search engine friendly manner but as how you would speak to a friend or communicate in that way. I can understand that it’s a great way to query, but as a result of it you have to understand (as they used in their example) that cancer is both a disease and a star sign, and understanding the context is a crucial element. They’re apparently integrating this into the Microsoft search technology by the end of the year, but the real challenge is to get people to use the service rather than how innovative the model is.

You can innovate in search from a querying perspective as above, from an indexing perspective or from a display perspective. With regards to display people are trying to innovate from numerous different ways (and I quite like what Ask is doing with mixing video, commercial and text links in a clean looking display) I’m less enamoured with searchme, it displays the search sites in a page where you can scroll through the pages kind of like Apple’s coverflow for iTunes. It looks very slick and sexy, but it doesn’t display the content as I like to process it; looking at a full webpage doesn’t give me the ability to scan the text and find out if it contains what I’m looking for. however shuffling through pages is easier to use than tabbing through pages, so you can see a lot of pages in a short space of time.

I think the display are of search si the easier area within which to innovate, creating a good gui that is easy to navigate is easier to do than finding a new way of indexing the web or developing a new algorithm to search in amore relevant manner. It will be interesting to see how all of these (and other) technologies fit together to create the next generation of search.

In other news, if you’re thinking of quitting Yahoo! (and it seems like many people are), here is easy to use resource for creating your resignation letter… (its multiple choice, so you can choose the tone)Related articles by Zemanta

UPDATE: it seems that implementing Live Cashback has had a positive effect on Microsoft’s market share in search

Reblog this post

Trademark & scrollable ads

in other Google news (i’m not just following them, honest). it is now possible to bid on trademarked terms in the UK and Ireland similar to the US system. I’m not sure of the customer advantage here either…

also Adsense now has ads which are scrollable, it makes senseto increase the amount of inventory that is available. looking at my own consumer behaviour I’m not sure how “engaged” i would be to look for more ad content, but looking at the volumes of ads served there must be an incremental value for them

new google functionality

interesting new functionality which provides an extra search within a site functionality. quite interesting, but i’m not sure why it’s on Google’s site, does it allow for extra ads potentially to be served?
Update: i have spoken to Google about this, and apparently you can make yourself exempt from this process;
The site: search operator is not new to Google — it has been around for several years. The Site Search Box feature is just being made more accessible to users where Google has determined that it helps them in their search.
Ad triggering behavior has not changed either, but as the number of site: searches have increased, advertisers are noticing it more. The ads team is aware of these concerns and will determine if any changes are required.
It is possible for an advertiser to opt out of this feature, but sites can be opted back in as Google makes improvements to the feature.
I would recommend that you look at the analytics information to objectively look at the result of this feature before making a decision.
Because Google is constantly evaluating search results and making improvements, it is possible that this feature might change and reappear in the future.
There have been more opt-in requests than opt-out requests, so not all webmasters are against this feature.
It is not possible to opt-in to the feature at this time give that this feature is determined algorithmically. As we refine this feature, the sites for which it appears may change and new sites can trigger this.