Gay Google Search
Perhaps as a celebration of the change in the law as regards same-sex marriages in California, Google are displaying the rainbow flag next to the sponsored links for “gay” and “queer” searches.
Perhaps as a celebration of the change in the law as regards same-sex marriages in California, Google are displaying the rainbow flag next to the sponsored links for “gay” and “queer” searches.
Google Dominate Search and Upstream Traffic
Hitwise have just posted Google delivers over a third of all UK traffic in which they discuss the amount of UK traffic Google get.
Here they are talking not just of the Google.com/co.uk domains but also Gmail.com Youtube, Blogger etc. Google get over a third of all Internet traffic in the UK. They are in effect controlling not just the means of distribution with 90% of the UK search market but also increasingly the destinations visited too.
This should be a cause for concern for everyone outside of Google. They have wrapped up the search market, and now are making further inroads into publishing. They saw all the traffic being lost from Google as people where only on a search engine for a few moments before engaging with other properties, and they’ve taken step after step to keep the traffic. The phrase “to have one’s cake and eat it too” comes to mind, and we’ve willingly baked them the cake, helped them cut it up and now we feed on the crumbs they leave over.
The Android platform for mobile is their attempt to take control of the mobile market, which will eventually become as large as the desktop market is now. Notice I don’t say the “mobile search market”. Android will allow Google to control the mobile platform in the same way that Microsoft did with the desktop.
If Google continue to take strides into the publishing model just what will be left for the rest of us? Traffic to their top 100 sites is up from 30% to 36% (March ‘08) of total UK traffic in the last year, then we can predict that within a year they will have over 40% and could have 50% of the upstream within two or three years. Outisde of Google’s traffic, our visitors will continue to increase for a while as the total internet population grows, but once that steadies we’ll increasingly find sites plateauing except for advances made against one another when one set of SEO techniques increases a site’s reach over anothers’ or a particular site becomes hot on social media for a while.
Then if Google ever moves directly into the niches we’re in, it really is game over.
Building upon success - the long tail & using existing keyword traffic
In my experience, many people undertake keyword research when developing new pages or sites but neglect to continue to study keywords on existing pages and utilise current success to further increase their authority within the topic they are addressing. In short, you’re leaving traffic - and therefore money - on the table. If a search engine is providing traffic on a subject you are already ranking for, it’s easier to get traffic for subsets and expansions on that topic, than to develop a new page on a new subject.
This is essentially an expansion of the long tail idea which says that although a few pages on your site provide the vast majority of the traffic on a page by page basis, the tail of the site’s keywords will provide more traffic overall. It essentially reverses the expected pattern in that although the top terms will provide a nice big chunk of traffic, there is a lot more traffic to be found in the tail of the search terms people use to hit your site.
On my personal site which has a fairly extensive linux & webmaster tips section that I’ve added to over the years, I find via hittail that the top ten keywords are 17.4% of my search traffic, leaving 82.6% to the long tail terms. Furthermore, 103bees (that’s 10 cubed by the way) tells me that nearly 43% of my traffic was from terms used only once, though the top 5% terms generated 38% of my overall traffic.
However, when you dig into the long tail traffic much of it naturally remains closely related to the top key phrases. For instance, the phrase “MySQL Distinct” is in the top ten phrases, but I also have dozens of mysql terms in the long tail, and more variants of mysql distinct itself. Looking at the Google SERPs for those specific terms I already get traffic for I see the following results:
| Key Phrase | Position: Google.com |
|---|---|
| mysql distinct | #3 |
| distinct mysql | #6 |
| distinct in mysql | #8 |
| mysql distinct group by | #8 |
Taking those phrases into the Google Adwords’ keyword tool (Wordtracker provided just 1 result, mysql select distinct) I also found a few other phrases it would be worth exploring on mysite, cementing my position on the terms and generating more traffic via long tail searches for something I essentially I’ve already done the work for:
- mysql select distinct
- php distinct
- sql server distinct
- update distinct
- insert distinct
- inner join distinct
- delete distinct
- access distinct
- group by distinct
- where distinct
- distinct join
- sum distinct
- distinct query
- order by distinct
- count distinct
- distinct statement
- distinct queries
- distinct null
- distinct command
- left join distinct
- distinct function
- distinct row
- distinct syntax
- having distinct
I’ve already removed the less relevant terms, such as Sybase, Mssql, Postgres etc from the original list leaving a set of keyphrases now with a pretty good relationship with my existing page mysql distinct
The page itself is pretty short, containing little more than two examples of SQL and a brief explanation, so should I chose to, it would be pretty easy to add more terms to fill the page a little more and start to rank better for these other terms. Furthermore, I could also use those terms to generate new pages for the site and interlink them from existing ranking pages and to other pages I want to target.
Utilising overall long tail terms to generate traffic can work wonders on very large sites, which is essentially Amazons success, but for a normal sized website to rank anywhere high enough to get traffic the phrases will need to be related to keywords you already are ranking for.
In a case of a site wanting to rank for both apples and oranges you’d start by building on any successes on apples with other related long tail apple terms - Granny Smiths, Bramley’s and Cox’s Orange Pippins - rather than targeting apples, oranges, bananas and pears. It’s the long tail of your existing authority. Services such as 103bees make it easy, they show you the hits you have, and you can easily mine the data to find other terms you have had a tiny bit of traffic for that you can now build upon.
Look at the terms you already rank for, and then look deeper into your long tail. If you can create a small database or spreadsheet of how those terms relate you are well on the way to developing more authority within your specific topic simply by placing a few relevant sentences on existing pages, better yet, build a few new pages with the new target phrases.
Once you’ve become successful in one area, it’s becomes easier to move that success into another topic - become successful in apples and then you’ll be able to move into the oranges business too.
Indexing of nofollowed pages via directory structure.
A few days ago I placed a couple of links on my tips page to another domain to test nofollow. I’ve occasionally done this via a number of sources and in each case I’ve always found that Google eventually index the nofollowed page.
I edited the page at Feb 26th at 16:34, I had two links to a test page, the anchor text was “Ape quake island republic” on the nofollow link and “Digg’s wiry crops” on the standard link.
Yahoo’s Slurp came in first, and subsequently spidered the followed link, but not the nofollowed. Google took the followed link a little later, and then returned to read in the test directory that the pages where in. This directory wasn’t linked to anywhere other than being in the pages URL path.
Google then read the nofollowed page, and a page it linked to. However, searches on the anchor text and unique text within those pages show that Google hasn’t included the nofollowed page, or the page it found from it in thier index.
A search on the anchor text for both links, results in Google displaying the linking pagefor both searches, and the target page for the followed link only.
Yahoo on the other hand, only show the linking page for both anchor text searches, and not the target page for either.
At the time of publishing this, I’ve removed both the links and the target pages, and will monitor who long the pages & phrases take to drop from both indexes.
Digg Friends RSS Feeds
Finding the Digg Friends Submissions page next to useless as it constantly times out, I decided to add my friends’ submissions feeds into a reader. To make it easy to add the RSS I grabbed a list of names and have quickly put together (thank you Vim) a list of the feeds, so I can easily click and add you as wading through each individual pages within Digg seemed onerous.
If I’m a fan of yours or a mutual and I’ve missed you here, please comment below & I’ll add your submissions RSS to my reader.
If you’re a fan of mine (thank you!) and I haven’t friended you back, again please comment and I’ll take a look.
Want mine? liamvictor’s Digg Submissions RSS
Numerical
A
B
- babychen
- badwithcomputer
- Balactus
- bamafun
- BartTheBear
- beammeupscottie
- BenisBenis
- bhartzer
- Blackmane
- BrettFromTibet
- BuckNutty
C
- canudiggit2
- charbarred
- Chazzer3
- ChristinaRowe
- chumpsucker
- cosmikdebris
- cowsgonemadd3
- ctacebu4
- CyberSally
D
E - G
H - J
K - L
M
N
P
- parislemon
- PatrickA
- Pebbles329
- phaedoHD
- pictureDIGGER
- pizzler
- pleeker
- pogle
- polymath22
- poontangpirate
- purrp
R
S
- sandanista
- ScorpioPT
- seanpackage
- SebastianX
- simpro
- sixthplanet
- Skepsis
- Skitzzo
- smitas
- sravkum
- sweetdeals
T
- Taiman
- TalSiach
- tamar
- TheMadHat
- TheWalkingDude
- Tinu
- titandegroot
- todesfee
- ToeCracker
- tomboy501
- topcat31
U - Z
Why Clara Net Sucks and Finally Lost a Long Time Customer through Ridiculous Bureaucracy
I’ve had a server with Clara Net for many years. So long in fact that it’s now obsolete - that’s another blog post altogether, (wanting & asking to upgrade the server and getting no response or outrageous quotes as they completely misunderstand or ignore my needs), but lets get to the point in hand. I’ve spent something like £30,000 on server fees, and I used to have super double quick ADSL in an old office too which would have added in another good few grand.
I had a problem today with the server I have with them. It simply stopped responding properly, pages on websites would eventually (after thirty seconds or more) report back with a MySQL error. At fist I figured a client’s site had got Dugg hard, and I needed to make the page static, or huge influx of spam was tying up all the resources. I tried to connect to the server in a shell session and whilst I got a login prompt, and then eventually a password prompt I was twice kicked out of the shell before I could investigate.
So I called tech support, the “3rd line” section who are supposedly the more technical guys, although it seems that they just put everything into a queue for NOC (Network Operations centre) to deal with. In the past I’ve found dealing directly with NOC (which you can do if contacting them out of standard hours) to be generally good. Thirdline seem to regard a reboot of the server as the first natural course in any investigation. But again I digress. Once I get through I explain my problem, I give them my name, and the server’s IP address and then the representative said, that because they didn’t have a password recorded on file she couldn’t get the server restarted (as I said their first option for almost anything), and would I fax through a signiture on letter headed paper.
Naturally, this is 2007, and I haven’t had a fax machine for the best part of ten years.
I get told I can scan through the document and email it to them. Okay, I can deal with that. Now will they get the process started and I’ll get this through to them. No. They have to have confirmation first. At that point I got angry. Why do I find that unreasonable?
Firstly, I have multiple clients sites on the server completely unresponsive. The clients are phoning and asking me why their sites aren’t working, that they are losing sales and money on AdWords etc, where’s their email and will it still arrive when things get sorted, and so on… and most of all they’re asking WHEN it will be sorted. What can I say? I’ll sort it when I get these hoops jumped through because Clara Net have set up security measures so I provide a password to gain access to technical support but unfortunately neglected to either tell me or set one up on my account?
Secondly, the tech support person knows there is no password set up, nor ever has been set up, and most importantly that the server isn’t working, so surely, simply from the POV that the damn thing isn’t running they don’t need any security measures to restart it. It’s a web server, it’s unresponsive. Restart it. Where’s the security issue? I ask to be put through to my account manager, who has been changed yet again with no one having the decency to actually call or email me and let me know, but naturally I wait several minutes only to be connected to voice mail.
I then call this new account manager directly several times, but can never get past voice mail.
Meanwhile I’m still trying to connect to the server in a shell session, if I can just get in, then I can see what the issue is, and perhaps resolve it. Whilst I’m still trying I go to print out a letter headed piece of paper so I sign the damn thing and then scan it to pass their security measure. Though, quite why a random piece of signed paper that I could knock up in photoshop would pass any sort of security check I don’t know. Where they really going to head down into the archive and find a contract from 2001 and compare signatures? I somehow doubt it.
Alas, I can’t print off the letter headed paper. I bought a new PC last week and I haven’t as yet installed photoshop on it. I also haven’t yet installed the printer/scanner, naturally I can’t find the install disks.
I get back through to tech support, and explain the above. I tell them there must be some other way they can pass the security check. Yes, they’ll call me. Unfortunately the number they have isn’t the one I am on. Nor my mobile, they wont confirm what it is. I’ve expressly asked them to change my contact details with them several times before, both through the tech support, through my account manager and through the accounts department. We go through addresses, I supply them with every address I can think of that they might have on file, my home, the old office, my old flat, the older office. Again no joy. I give them the account code, but that doesn’t count. I give them the name of my old account manager, that again doesn’t matter.
Thankfully, as we are talking I establish a connection via PuTTY with the server and as we go through the proces of them trying to authenticate me - simply because their own systems require it, and ignoring the fact that a password has never been set for the account and that security isn’t an issue when just trying to get an obvious problem looked at with a server - I’m able to look at the issue myself. The home partition on the hard drive had got dangerously full so I had to simply delete some temp files, move some others around, and things quickly began to improve with the servers responsiveness.
Note, that restarting the server would have probably caused more problems as new log files would need to be written and with little to no space left could have killed things off completely.
At last the server is working. No thanks to technical support. I’m currently moving many of the domains from them, and will complete migration of the larger accounts ASAP.
This is on top of the fact that the server is supposed to have a SSL cert supplied by Clara Net and each and every year they screw it up, and for several days I normally have downtime - and clients lose money as strangely people don’t like to checkout on an e-commerce site with no SSL - this year I contacted them a good week in advance, and they still screwed up, and then no one was available to sort things out after 5 pm on the Friday and no one would or could deal with it until Monday when the guy who should have sorted it would return. I bought my own in under ten minutes. Clara Net have still not responded to my repeated requests for a refund for the value of the certificate I had to purchase despite it being part of my deal with them. The most I’ve had from them is my (then) account manager confirming that they’ve had problems with client certs before. There’s been other issues too.
What should they have done today? Seen there was a problem, seen that their systems weren’t set up properly and be able to work around them. They set up passwords but hadn’t on my account, and never bothered to sort that out, their system was at fault. They should have been responsive to my needs and dealt with the important stuff first - a non working server - and sorted out the details later.
Are Social Media Connectors less important than thought & shooting for the stars to hit the moon
In this article, Guy Kawasaki agrees that key influencers are less important than they give themselves credit for. The article itself is more than interesting, pointing to a summary of research by a Miami University faculty member and his colleagues that states
Old-fashioned “word of mouth” might be more useful in advertising than previously thought, especially in digital media
Digital word of mouth can obviously be social media, they argue that the key influencers are less important in spreading a message than previously thought.
“We find that trying to track down key influencers, people who have extremely large social networks, is typically unnecessary and, more importantly, can actually limit a campaign or advertisement’s viral potential”
The thing is in the most part as regards social media I disagree. One only has to watch the reach and percentage of front page hits that the top and upcoming Digg users get compared to an almost normal user to see that their influence is far greater.
In the Tipping Point Malcolm Gladwell spoke of three types of people being necessary for an idea to spread, Connectors, Mavens, and Salesmen. The top social media people will usually be supreme Connectors, and have strong influences from the other personalty types. Often they will begin their social media journey with more Maven like intentions - wanting to spread ideas & knowledge, but to really succeed in some social-media circles, especially Digg, you also have to be a good salesman - picking just the right headline for the piece to resonate with the audience.
Where as other people are I proud to call my friends on Digg such as Tamar, Zaibatsu and msaleem have a far greater reach. If you wanted your website to gain traction would you want me to digg it or one of them?
|
|
|
|
||||||||||||||||||||||||||||||||
On the other hand, I’m currently one of the top stumblers. As many reports have shown the type of long-term traffic that Stumbleupon can send compared to digg. That has it’s own rewards but unfortunately for me, doesn’t scratch the competitive itch the way that seeing the digg numbers does.
The reason for the post however was not so much the merits or otherwise of the research, of being a top digg or social media person, but to highlight a beautiful quotable gem from Mr. Kawasaki who ends his post with the following that completely struck a nerve with me.
“Create something great, sow fields (not window boxes), let a hundred flowers blossom”
I’ve always believed one has to “shoot for the stars to hit the moon”, Guy has reminded me that I should be shooting for a hundred stars to hit a dozen moons.
Google Toolbar Broken & Why I’m keeping it.
Andy Beard wrote today: Google Toolbar Is Sick - Now Removed. In it he posts a capture of the current upcoming pages in the Google dice. Whilst I wasn’t surprised to see some overlap - we’re both web developers / SEO / SEM types, and I’m pleased to say that we’re mutual friends on both Stumbleupon and Digg - but even so I was shocked by just how much correlation there was.
Comparing just the first of Andy’s captures to mine. I have 26 links, and see 20 matching titles (grey in my grab), and 1 that probable match (coloured pink in mine). Just five differed for me - two football links, two SEO pages and a mod_rewrite article. I had all but Keyword Discovery of Andy’s list.
I always knew the Google Dice feature was flawed in comparison to the StumbleUpon toolbar, it would have to be, I visit many pages on Google that I don’t “vote” for. With SU I implicitly tell them what I like and what to give me more of. With Google they are simply looking at both my search and web history, much of which will be largely irrelevant when creating a suggestions list. During the normal course of a day I will visit pages from Digg that I aren’t necessarily things that I’m truly interested in. I’m sure if I thumbed up every I see in SU, the pages delivered by that would be just as poor.
However, despite Andy’s concerns on both the relevancy of the algorithm & that it is “unsafe for family consumption”, the well documented privacy issues and now seeing the fact that the Dice suggestions are extremely generic, I’m keeping the toolbar installed.
Why?
I suspect that the toolbar’s web history data is used within Google’s search calculations as it’s been mentioned before about other influencing factors in the results other than on-page and linkage data so if I can have a little influence then I’m prepared for the downside of using their tools.
If I were Google, as well as looking at search & web history, I’d also look at how pages are discovered using links sent via IM, are bookmarked within the toolbar, and at links sent (and clicked) from within gmail. I’d probably look at the top pages with social media sites and analyse traffic data too. I’m not too concerned about the privacy issue - as long the data remains private between Google & I - Google probably have many other mechanisms for keeping track of what little-ole-me gets up to anyway. If they can improve search results and tailor results better suited to me, especially with less spam, then I’m happy for that.
I usually don’t have to worry about my browser being used by anyone else, so the main reason for Andy’s abandonment of the toolbar being that it is “unsafe for family consumption” isn’t currently a concern for me so I’ll live with it’s big-brotherness, the poor relevancy of the Dice (which I seldom use) and try to remain aware of the implications of having an active toolbar on a browser in a shared environment. In fact, that’s something that any computer user should remain alert to in a variety of work and family circumstances and not just when discussing the relevancy of Google’s suggestion tools.
Why Start A Blog?
I’ve actually been blogging in one way or another for 5 or so years, going back to some geocities spaces, blogger and Stumbleupon.
With those blogs I’d never taken a niche or specific topic and instead just had them as a collection of links with commentary or the occasional online diary. I used the existing Liam Delahunty Tips for most of my web/tech focused writing, which otherwise may have become a blog.
I’ve been thinking for some time about turning the tips into more of a blog, mainly for the tags/categories but have decided to work this domain first and see what kind of readership I earn and comment more on matters that don’t fit easily into the tips section.
As it stands, I intend for this to be more of blog on SEO, PPC, social networking and perhaps the always popular “making money online”. Linux and web development will likely remain over on liamdelahunty.com for now.
Whilst I find my feet here I’m going to start of reviewing (or at least commenting) some of my favourite blogs and personalities in these verticals as well as posting some original material. I really would appreciate your comments and ask your indulgence whilst I doubtless play with the templates and layouts as I’m getting things together.
Starting a blog now may seem like too little too late, but the fact is, still 99% of people online don’t currently blog and I’d estimate despite the plethora of blogs within the known SEO/SEM community that 90% of SEOs don’t blog and even fewer web developers do; as such I’m a pioneer, albeit one following a more developed path than forging ahead on my own!