Skip to content

I Am An Internet Marketer

I don't feel like I'm an internet marketer, but I must be because a couple of days ago (17/01/2011) I sent an email newsletter to my "list" promoting a product.

I'd had a plan for the email, and had been dying to send it out for a while, but thought that it would probably be better to wait until the Christmas madness had subsided. Few people are going to want to buy what I'm putting in front of them in the midst of Christmas spending. I also thought it would be better suited to a Monday delivery.

Measuring Performance

I sent the newsletter to 826 subscribers. There is an image link at the start and a text link at the end to the product I'm promoting and so far this year (it's 17th January), I've had 6 sales from my websites. These 6 sales are not related to the promo email I just sent and won't be counted. It's more to give a benchmark. I know that I'm not selling many of these products via my websites (one every three days or there abouts), so if I see a massive surge in sales, I know it's from my new friend: the internet marketing email!

The first day was the worst. I couldn't help watching my getresponse stats throught the day and they were hugely disappointing. By the end of that first day, 826 emails had been sent, but none had been opened, 10 had bounced and there'd been one complaint! How someone complain about the email when none had been opened is beyond me...

At the start of the second day, still none had been opened!

Update (19/01/2011) : Oh for God's sake, I've just realised I didn't select the option to track open rates! No wonder it's telling me that nobody opened the damn email! Well there's a lesson.

The good news is that I made two sales over night worth $15 each. Hopefully those sales represent the few drips before the deluge. I think I'll let my little test run for a month before drawing conclusions about mailing a list.

Update (19/01/2011 - 21:00) : Woo hoo, another sale!! A total of 9 so far this year, with 3 probably attributable to my newsletter. OK, enough micro management.

Update (21/01/2011): 5 sales since the newsletter, totalling $75.

Update (24/01/2011): 7 sales since the newsletter, totalling $105.

Update (27/01/2011): oh well, it looks like the honeymoon is over. No further sales. That's a little disappointing, I was expecting more from my "highly targeted list". It didn't take me long to put th email together, so the ROI is pretty good.

Posted in articles.

Tagged with , , , .

No comments

By admin – January 19, 2011

PR5 Expired Domain

I let a PR5 expired domain slip through my hands in the Godaddy expired domain name auctions a couple of days ago. It was a really good catch, but when the bidder I was bidding against raised their bid to $1,315, I folded. If I'd not already bought another expired domain for $1,500 earlier in the month, then I might have bid more for this one.

The domain I lost was nortellearnit.org, and it has some impressive credentials:

  • PR5
  • Around 7,500 inbound links in total
  • A lot of .edu links
  • 4 .gov links
  • A wikipedia link
  • A good level of traffic, as indicated by SEMRush
  • Very nice keywords it ranks for in Google, also from SEMRush

Erstwhile Traffic

SEMRush can show you traffic levels and also the kind of keywords that the domain used to rank for. It looks like this domain was receiving at least 8,000 unique visitors / month via Google searches.

At Least 8,000 UV / month

I've found on my sites that actual Google search traffic is around 3 times greater than the totals reported by SEMRush, so this domain is getting very nice traffic levels. There are so many inbound links, that I think there's a good chance there may be a lot of referral traffic too, but SEMRush doesn't report that.

The keywords the domain ranks for are perfect generic terms. If you had won this domain, you would want to check out the search phrases the domain ranks for and create some content about those phrases. Ideal search phrases are "evergreen" ones that don't go out of date too quickly. If we take a look at the SEMRush keywords report (see link above), we see that the search phrase that brings the most traffic is "teachit". I think this simply means "teach IT", as in information technology. That's a good generic term that isn't tied into specific technology and is more enduring than phrases like "teach windows xp", for example. Phrases that are linked to specific technology risk having a finite shelf life. With "teach IT" the scope is wide, so you could write about a huge range of subjects that fall in the area of IT. And, of course, technology related topics tend to attract higher Adsense revenues, if that's how you wanted to monetise this site.

Trademark Issues

The second biggest traffic puller, however, is the search phrase "nortel", and this triggers alarm bells for me. This shows that people are currently searching for the Nortel company and this domain name is linked with that company. The domain name itself is Nortel Learn IT. A quick check of the WHOIS information for this domain shows that Nortel Networks Limited owns it.

Google provides some clues even as we start typing in our search for "Nortel Networks Limited":

Nortel Networks BankruptcyOh oh! A bit of digging around tells us that Nortel Networks Limited went bankrupt in early 2009 and is still in the process of selling its assets. It's likely that the domain name in question, nortellearnit.org, got forgotten in the chaos and it expired. If only the domain name didn't have any branding in it!

I don't know much about trademark law, but I got a little nervous about the prospect of buying this domain. I didn't like the thought of spending lots of money on it, only to find that somebody had the right to take it off me at a later date. If the bidding in the expired domain auction had only gone as high as $500, it would definitely have been worth the risk to me. As it was, I bid up to $1,300 when the heat of the moment swept me away, but then regained my common sense when I was again outbid!

Another thought is this. You could buy this domain, knowing full well that the new owners of Nortel Networks Limited (or owners of a subsidiary company) are likely to want the domain back, but rely on the fact that it might take years for the sluggish wheels of the legal process to turn. I don't know how long these things take, but suppose you have two years before you have to give the domain back. Nortellearnit.org is a domain that seems to have a lot of authority in the search engines, so a link from it to other sites you own would be worth a lot. Adsense revenues on a high traffic domain like this, in a technology niche as it is, would amount to a fair chunk of money over two years too. And then there's the old link selling route. My guess is that you would probably be able to more than recover your investment over a two year period, if you picked the domain up for under $2,000.

I'll be interested to learn what the new owner does with this domain, as there is so much potential for making a really good site on it. I'll be very surprised (and disapointed) if all they do is park it, as many expired domainers do.

Posted in articles, auctions.

Tagged with .

2 comments

By admin – December 29, 2010

Keyword Research Using SEMRush

My favourite analysis tool of the month is... SEMRush. Actually, I'm not a big user of tools, so there won't be a regular tool of the month award. But in SEMRush I've found something truly wonderful: the ability to spy on the keywords other websites (OK - competitors!) rank for. And the ability to export them to Excel. And then resequence them by search volume. And filter out the high competition phrases.

In a nutshell, SEMRush allows you to identify all the easy keywords your competitors rank for that have decent search volumes. More Lazy Man's (or lady's) Keyword Research!

Let's run through the process of identifying these keywords.

Using SEMRush For Keyword Research

The process can be summarised as follows:

  1. Use SEMRush to list all the keywords a competitor site ranks for.
  2. Export to Excel.
  3. Sort the spreadsheet by Search Volume.
  4. For each eligible search term, conduct some rudimentary competitive analysis.
  5. Filter out those terms that are too hot.

We'll go over those steps in more details here.

SEMRush Google Keywords Report

Type in the domain name of the site you're investigating and click search. For some reason, it doesn't work if I leave the "http://www." bit in, so I usually just strip that out. Next, click on the Google Keywords link on the left.

Google Keywords In SEMRush

The Google Keywords report displays all search phrases that the site ranks in the top 20 for.That list is gold. However, we need to be able to manipulate the data so let's export it to Excel. In the top right of the page you should see some buttons to export the data into a spreadsheet. I usually use the one shown below:

SEMRush - Export To Excel

Working On The Keywords InExcel

Once the data is in the spreadsheet, we need to resequence this list of phrases by their respective monthly search counts. Select the Search Volume column and click the Home tab > Editing group > Sort & Filter. Select Sort Largest to Smallest and in the window that appears, select Expand the selection and click OK.

SEMRush Sort Search Volume

We now have all the phrases with high search counts at the top of the spreadsheet. We're going to set a threshold search count for phrases below which we aren't interested in. I usually don't look at phrases that get less than a reported 1,000 searches per month. If you want, you can delete these from the spreadsheet.

Quick And Dirty Competition Research

Here's a slapdash but quick approach to performing research on how much competition there is for a particular search phrase. In Google, type allintitle:"big hairy butts". This command shows us all pages in Google's index that have "big hairy butts" in the title. If a webmaster is going to optimise their page for a particular phrase they are going to put the phrase in the <title> tag.

Generally, if a phrase has less than 50,000 results returned by the allintitle: search, I see it as a possible target. You'll have to set your own threshold for this, and that will depend on your experience and also the authority (and everything else SEO!) of your site.

If you work down your spreadsheet, performing allintitle: searches on each phrase and omitting those with more than 50,000 competing pages, you will be left with high search count/low competition phrases. Target those and the distinct possibility that you'll soon be a millionaire is very apparent. It's a good idea to create a new spreadsheet at this point, to hold all the eligible keyphrases. Typically, I have 3 columns in this new spreadsheet:

  1. Phrase
  2. Search Volume
  3. Allintitle

Simply copy the phrase and search volume from the original spreadsheet to the new one, and add values for the intitle: searches.

Excel Keyword ResearchThis new spreadsheet makes it easy to pinpoint hot phrases that will be easy to rank for.

Rinse And Repeat

There is usually more than one site in a particular niche. This means that you can now pick another site in your chosen niche and repeat the above process. Keep adding to your phrase/search volume/allintitle spreadsheet and after a few iterations, you'll have a monolithic spreadsheet of really good keywords that will keep you busy 'til you die.

Now that you have a list of keyphrases, target them by writing new pages on your site about them.

Posted in articles.

Tagged with .

No comments

By admin – December 13, 2010

Find Connected Sites

Thanks to Experienced People for listing some very useful tools for due diligence on websites.

When researching a site you're thinking of buying, it's helpful to know what other sites have the same Google Analytics code or Adsense code embedded on them. Unscrupulous sellers will often use the same Google Analytics code on multiple sites to create the impression that the site they are selling gets more traffic than it really does.

It's the same concept with Adsense. Displaying the same Adsense code on multiple sites can lead the prospective buyer to believe that the site being sold is generating more Adsense revenue than it actually is.

The average seller will not volunteer this information, so we need to discover it for ourselves. The following tools identify sites that sport the same codes. They also report other data too.

  • Spy On Web - reports sites with the same Adsense publisher ID and Google Analytics code as the one you're researching. Also lists the IP address of the host server and all the domains hosted there. The DNS servers are reported, too.
  • Serversiders - in addition to reporting Adsense and Google Analytics connections, serversiders actually does a bit of analysis itself by statings things like "Domains that appear to have the same administrator or owner as this domain include....". So the tool must be looking at the administrator in the WHOIS data and finding matches on other domains. Smart. Additionally, this tool picked up the fact that I was using Chitika and Adsense on one of my sites.
  • Domain Crawler - presented the same Adsense/Analytics cross checking info as the previous tool, but did not find as many connections.
  • AdShadow.de - showed no results for one of my bigger sites.
  • SameOwner - reports similar data to previous tools, but requires registration to see the data. Adsense connections missed a site that the others picked up. The cheeky bastards are charging for subscriptions to get more than the briefest details! I'll pass on this one. Serversiders seems to be the best tool so far, and it's free.
  • w3who - provided inaccurate results for the Adsense connections, but was OK on the Analytics.
  • ReverseInternet - another site requiring registration to view the complete details. But, boy is this site thorough! Using it I've just found a site that redirects to one of mine. It's listed as having the same Adsense publisher ID so the crawler must arrive at the URL, get redirected to my site, but still think it's on the redirected URL. For "thorough", read "over-zealous", as some unconnected sites are getting listed too. That's OK, I think the fact that it found the redirected sites makes up for the over zealous reporting and I like teh way the information is presented using this tool too. This one's in the lead. Oh, now this is good: links from related IP addresses will help us to find links from other sites the seller may own. There's also a section that displays links to domains on related IPs, but this is less important for the site being researched. Oh wow - it has what I always wanted: Clickbank cross referencing!
  • Webrota - more or less the same info as previous tools. though the section showing other Adsense sites with teh same publisher ID missed some when I tested one of my sites. One thing I found useful here was the similar sites section. I'm always on the lookout for other sites in my niche (to buy!) and I got a few pointers here.
  • Hostnology - again, the Adsense publisher ID cross referencing wasn't as good as other tools (Reverse Internet, for example) and some sites were missed. At this point, I'm using Reverse Internet as the benchmark all tools have to beat! There is no information here that is not reported by other tools.
  • SiteGrep - the information presented by this tool looks just like that presented by webrota.
  • WebSiteLog - on the site I checked, it missed another that shared its GA account (2 different profiles on the same account, not dubious practices!!). This tool is absolutely (over)loaded with Adsense, so I doubt that accurate information is the owner's prime motivation. It missed other sites that shared the same Adsense account. Waste of time, this one.
  • DomPing - doesn't provide Google Analytics cross references and displays incomplete cross references of Adsense accounts. I've been spoilt by the rigour of that other tool I keep harping on about. Reverse Internet, wasn't it?
  • WebsiteInDepth - one useful feature of this tool is its ability to tell you how many letters are in the domain. WTF?! Counting wasn't my strong point, so thanks! Jeez, this pathetic tool couldn't even determine the age of the domain I tested. The visuals on the site are wonderful, which just goes to reinforce teh theory that you need more than a nice theme to make a useful site. Reverse Internet, on the other hand, is very bland in comparison. But at least it f*cking works.

Conclusion

There's only one Reverse Internet! It does a lot and it's accurate. It reported all the sites (22 of them!) that shared the same Adsense account and both sites that shared the same Google Analytics account. Additionally, it found domains I didn't own that edirected to my site. Amazing - I've actually been looking for just such a tool and Reverse Internet does this without even meaning to.

In addition to the usual sites that share the same IP address information that most tools display, RI also displays Clickbank cross references. Yummy. This means that you can find other sites that use the same Click bank tracking ID. Very useful indeed when used in your site buying due diligence. Get this: it even shows you other sites that promote the same CB product as the site you're investigating! I think I just wet myself.

Posted in articles.

Tagged with , .

2 comments

By admin – November 28, 2010

KISSinsights Results

If you read my earlier post about the KISSinsights survey, you're undoubtedly chomping at the bit to find out what the results were. Well, the results are in. And they're as barmy as I feared. In my experience, any forum for self expression tends to attract nutters.

Just look at blogs :)

Surveys are no different, and one line surveys are like nutter magnets!

I put KISSinsights on a different site to the one you're reading, and here are some of the more interesting answers to the question: "What other information would you like to see on this page?".

  • Author Details and contact information - so surveys attract stalkers!
  • sex - I must confess that the topic for the site is as far removed from the "s" word as you can get. This answer puzzles me.
  • where to find xxx - OMG, this is gold! xxx is a product I can sell. Thank you, thank you, thank you for an insight so good I could kiss it. This demonstrates the value of surveys. There was another later response asking for the same thing too. This just reinforces the idea that selling xxx on the site is a Good Thing.
  • Setting xxx - this is a request for information that is very useful for guiding new content creation. I don't have any content explaining how to "set xxx" but now I know that at least one person would be interested in reading about it. It's on my to do list. In fact, the next five respondents wanted some "how to" information, too, on varying subjects. This is fast becoming a "to do" list for content creation. More on methods of content creation.
  • When will the software be released? - oh dear. Apparently I have done a poor job of communicating release dates as the product has been available since March!

One advantage that this survey has over the 4q iPerceptions survey I'm using on yet another site is that with KISSinsights, you get feedback at page level. If someone says they need to see X on page Y, you know exactly where to add that new material.

How To Misinterpret Statistics

The respondents to the above KISSinsights survey numbered 15. Out of those 15 people, 1 wanted sex. That means 6% (1/15, and I rounded down for modesty) of respondents came looking for sex. The site I added the survey to gets 15,000 visitors / month. 6% of 15,000 is 900 people. That's a lot of sex hungry visitors.

So, 900 feisty people come to my site looking for a bit of 'ows yer father. Oddly, the same proportion want to know what my contact details are! They're obviously the same people. Maybe they saw a photo of me or something. Honestly, good looks can be a curse...

Stay tuned for more ludicrous interpretations of statistics.

Posted in articles.

Tagged with , .

No comments

By admin – November 17, 2010

Kiss Insights

Encouraged by the success I found with the 4Q iPerceptions free survey that I added to one of my sites, I decided to try out a new style of survey. Behold the Kiss Insights survey! This survey is in the style of Twitter as the site visitor has only limited space to give feedback and is perfect for ADHD sufferers and those who have to make good their escape before the rozzers arrive. They get a short, punchy question that pops up in a discrete area of the site where they can enter a similarly punchy response.

It's not intrusive at all.

I added a survey to another of my sites and also to the one you're reading, too. By the time you've read to this point, you should have noticed a little panel sliding up in the bottom right of the screen (please work, please work!). It should look something like this:

Kiss Insights

How cool is that?

There are a lot of pre built surveys that have their own category of question, including the all important "why did you abandon my shopping cart process", but the one I found most useful for my sites is the following one. "What other information would you like to see on this page" is such a subtle, yet open invitation for the respondent to give constructive (maybe negative) feedback about the site.

I'm hoping that because it's not a full blown survey with lots of questions, it'll get more respondents. And because it's such a small and unobtrusive little thing, my guess is that it can appear on every single page without interfering too much with the user experience. Actually, it does appear on every single page, so I hope that is the case!

I can see a potential problem immediately with the timing of the popup widget. It appears very soon after the page is loaded and before the reader has had chance to read the whole page. Is the temptation to send a response straight away, before the whole page has been read, too great for the average visitor? If you've ever studied the responses to a survey you've implemented, you're probably familiar with the nonsensical and damn weird comments that some people leave. I just know that some people are going to reply to the question "What other information would you like to see on this page" with the precise description of the page they're currently on!

We shall see.

Posted in articles.

Tagged with , , , .

1 comment

By admin – November 9, 2010

Due Diligence On Websites Using SEMRush

This auction for a Power Of Mind Control website caught my eye, mostly because of the decent traffic statistics. 28,345 unique visitors and 870,600 page views / month is quite good in my book. Each visitor is looking at an average of 30 pages!

Is the seller exercising some kind of mind control on his visitors?

The first port of call is SEMRush, where we plug in the URL and get the following:

Porn Terms In SEMRush

Click to enlarge

Oh my word! Look at those shocking search terms that brought visitors to this site (highlighted, and under the "Keyword" heading). If you can't speak Francaise, I'll do a bit of translation for you:

  • zoophilie gratuite - free bestiality. Ye Gods, I swoon!
  • sensitive pornograph ova - no idea but the second word looks like porn, and I don't care whether it's sensitive or not!
  • foto scopate - swept photos? What with the earlier porn references, I'm thinking that any kind of photos on this site will be dodgy.
  • adolescentes nues - nude teens. I'm getting the picture. No, I mean figuratively!
  • voyeurfrance - I don't think we need Google Translate for this one...

SEMRush is good in that it displays the URL that gets returned in the Google SERPs when someone searches for a particular term:

Investigate The URL

I'll tell you why it's good that the URLs are reported here. It's because the owner of the site has actually removed those pages that show up for those search terms, or they are restricted viewing (see later). Whatever the reason, we can't now see the pages on this site that rank for those terms.

Let's click on the first URL in the list. This is the page that visitors find when they search for "zoophilie gratuite". SEMRush helpfully inserts the URL into the search box at the top, where we can copy it and paste it into our browser's address bar. We can see straight away that the URL is for a page on a forum.

URL Is A Forum

When I try and load this page into my browser, I discover that I need to register on the forum to view the page. So I do.

However, the thread I requested is off limits.

Thread Is Off Limits

This could be because it's protected within a restricted area of the forum, or the owner simply removed the thread. Either way, there used to be several threads on this forum that attracted traffic from people searching for porn terms. In France. Those pervy frogs are at it again. I can't see anything on the forum other than the usual paid to post threads that website sellers usually splash out on, and the inevitable spam.

SEMRush displays at the top of their keyword list the search terms that bring the most traffic. If all the top search terms are porn related, then the danger is that there won't be much traffic left when the new owner gets rid of the dodgy pages. The new owner would have to get rid of those pages, because they are not "money pages". No products get sold on them, they don't lead to signups and there probably isn't much ad revenue. The only benefit they give is to artificially inflate traffic stats. Which is good when you are trying to sell a site.

Flippa's Crappy "Due Diligence Data"

Oddly, the SEMRush data that Flippa displays for this site is incomplete, and misleading. Here are the keywords that Flippa reports via SEMRush:

No Rude Words Here, Move Along, Move Along

Click to enlarge

Where is the porn?

This is really weird. When I click on the More detailed information at SEMRush link, the porn keywords are presented, but Flippa is not showing them in their mistitled Due Diligence Statistics.

That's piss poor, if you ask me.

Where are those watered down keywords that Flippa displays coming from? Ah, I see: Flippa only gathers SEMRush data at time of listing. Like I say, that's piss poor. And why would the keywords change so much in the space of 24 hours, anyway?

My Favourite Search Term

My favourite search term that SEMRush reports is "mind control lesbian". I just have this weird image of an army of lesbians surrendering their minds to a powerful hypnotic figure.

Come, disturbing army of lesbians, and do my bidding. There is much evil to be done. You can click on these ads for a start. Mwuhahah

The seller is a real gem. After seeing this post, he threatened to steal all the traffic from all the sites in the world to prove his mind control power. "Behold the loss of traffic in all Google Analytics accounts for November 2 2010, for it shall be mine! It'll be a piece of cake using my zombie lesbians.", he says. What a loser.

Posted in auctions, Danger Corner.

Tagged with , , .

4 comments

By admin – November 2, 2010

How To Ruin A Good Website

Check out this auction for a PowerPoint tutorial site. If you check the previous auction listing, you can see that it was bought in March by someone who has self destructive tendencies.

According to the original listing, this site used to be a PowerPoint search engine. That is, you could use it to search for PowerPoint presentations. In the original listing, the seller boasted of lots of high rankings for "PowerPoint Search" related terms, and 2,800 unique visitors / month. Not too shabby.

Fast forward to the current listing. Here are the salient points:

  • traffic has come down to 1,136 visitors / month (no proof given).
  • the site has been repurposed to publish PowerPoint tutorials. It is no longer a search engine.

Weird Logic

The site in its previous guise of PowerPoint presentation search engine seemed to be faring OK, but the previous owner obviously saw more success to be had in the PowerPoint tutorial game. Fair enough, but why didn't they add to the existing site instead of replacing it? They could quite easily have kept the site as it was and add the new blog for the tutorials in a new folder. Doing so would have kept returning visitors happy, and wouldn't have disturbed search engine rankings so much.

The problems with repurposing a site can be summarised as follows:

  • inbound links may no longer be relevant. This can have several consequences. Links that aren't as relevant won't help the site's pages rank as well as they did. Also, webmasters who originally linked to a search engine site might remove their links now that the site publishes tutorials. Additionally, visitors that arrive wanting a search engine are going to be disappointed. Disappointed visitors = well, it's just bad!
  • disappearing URLs. I can't tell whether vanishing URLs were an issue in this case, but often when you change the content on a site, the URLs change too. Changing URLs may lead to valuable inbound links pointing to non existent pages. Also bad.

One of the problems with the new content on this particular site is that it's all copied from elsewhere. Great, another site that simply replicates tutorials that can be found somewhere else. It doesn't provide anything of value to the user!

Don't Trust Anything A Seller Tells You

Look at the comments in the original auction. Specifically:

Bidder 1: How much are you paying for advertising?

Seller:Nothing for paid advertising. The site attracted free reviews from top blogs itself.

Bidder 1: Most of your 12K links are from www.pakblogging.com. I've been talking to them and they rent out advertising space for $30 USD / month on their site.

Seller: Pakblogging.com is being operated by my associate, and I kind of own it. And yeah, we're going to keep the link up after transferring to the new owner. Thanks for the query.

LOL @ "I kind of own it". When the seller says "attracted free reviews from top blogs", he actually meant "I linked to it from my other site". So those 12K links are from another site the seller owns. The risks here are obvious: those links may disappear at the drop of a hat when the site changes hands. Any SEO/traffic benefit that those links give will just evaporate.

Bidder 2:For how long are you going to keep the link ... is it going to be permanent ?

Seller: Permanent. I want to link forever :)

Thank the lord that the seller gave us that assurance. We can all sleep easily knowing that the seller intends to keep those 12k links in place.

By the way, those links are no longer there.

It may be that the seller lied and removed those links himself, despite his assurances. It may also be that the seller actually sold the second site and the new owner removed those 12k links. Either way, this demostrates why it doesn't matter what the seller says in the auction: current links from other sites tha seller owns may not remain.

Posted in auctions.

Tagged with , .

4 comments

By admin – November 1, 2010

Using Google Webmaster Tools To Determine Anchor Text

I'd contacted a webmaster in May about getting a link on their site to my site, and they finally got back to me today with a "yes". I knew what page on their site I wanted the link on, and I'd already figured out what page on my site I wanted the link to point to. What I hadn't yet decided was what anchor text I wanted.

If you know anything about SEO (search engine optimisation), you'll appreciate the importance of anchor text in ranking. In a nutshell, using the anchor text "red widgets" in a link to your page gives that page a better chance of ranking in the search engines for "red widgets".

My thinking was that if my site is already ranking at position 1 in the search engines for "red widgets", I don't really need more links with "red widgets" in the anchor text. Hey, let someone else have those links! What I wanted to do was find some phrases I was getting decent traffic for, for which I was positioned further down the page than 1st, 2nd or 3rd. Then my link would help me rank higher for a phrase that I knew would bring me much more traffic. So, how do we find such a phrase: a phrase that would lead to the greatest traffic growth if we started ranking higher for it?

Holy different stats for different rankings batman - this sounds like a job for Google Webmaster Tools!

Using Google Webmaster Tools, you can see just how much traffic you get depending on what position in the SERPs (Search Engine Results Page) your page is listed at. Your pages will move around teh SERPs quite naturally, so one day a particular page might rank #2 and the following day it might rank #5, for example. That page will receive different numbers of clicks for each different ranking. Obviously, rthe higher ranked pages will attract more links.

To get to this useful data, go into Google Webmaster Tools and access the dashboard for the site you're researching. In the Search queries section, click More >> at the bottom. Find the phrase you're interested in (this is the potential anchor text that you're investigating) and click on it. Ignore the data at the top; what we're interested in is listed at the bottom and has the heading Position in search results.

Investigate Anchor Text Using Google Webmaster Tools

Click to enlarge

For each ranking position in the SERPs, GWT displays the number of impressions your site got and the number of clicks. So when we look at the image above, we see that my site appeared at position 2 for 2,900 times, position 3 for 6,600 times and so on. But the most significant part of this report is the bit that says mys site appeared in positions 6 - 10 for 33,100 times! That means that for most searches of this particular search term, my site ranks 6 - 10. My site would definitely benefit from getting a link with this phrase as the anchor text.

Let's work some numbers. The CTR for my site when it appears at position 2 is 6%. Let's use position 2 and 6% CTR as the goal that we try to achieve by getting this new link. Suppose that after getting this link, our 33,100 impressions are moved from positions 6-10 to position 2. In fact, let's pretend that all the impressions from positions 3, 4, and 5 move to position 2, too. That means that we will have 49,100 impressions at position 2. With a CTR of 6%, those 49,100 impressions translate to 2,946 visitors. The screenshot above shows a total of 1,300 visitors (the sum of the clicks) for these positions. So if the new link leads to my site ranking at 2 for all the times it currently ranks 3, 4, 5 and 6-10, then my traffic will roughly double for that search phrase.

Let's get the link!

Dodgy mathematics and wild assumptions brought to your courtesy of spgazette.com.

Posted in articles.

Tagged with .

No comments

By admin – October 26, 2010

Dodgy Traffic

Check out this auction.

Now check out the Google Analytics traffic stats for the site being sold. They make interesting reading from a "piecing together a jigsaw puzzle" perspective. I thought they looked rather good at surface level, but then I looked a little deeper and found myself asking the following questions:

  • Traffic sources - why are most of the visitors direct traffic? Direct traffic arrives because the visitors already know about the site. They type in the URL of the site into their browser's address bar, or they click on a link in an email, or they click on a bookmark. So they already know about the site before they arrive. So if most of the traffic (73.29%) already knows about the site beforehand, why does Google Analytics report that traffic as being 98.39% New Visits? That's nearly all! How did they get to know about this site? If what springs to your mind is traffic schemes, then we're on the same wavelength.
  • Average Time On Site = 11 seconds - why is it so low? 11 seconds isn't enough time to do anything other than think "WTF, what is this site, I'm outta here!". Coupling the time on site with the fact that most of the traffic is direct just deepens the mystery. Except that this is no mystery, right? Why would any subscribers, for example, come and visit and leave in 11 seconds? If you're a subscriber you already know and like the content!
  • Bounce rate = 97% - I have sites that have a bounce rate of ~ 80% that I need to worry about. A bounce rate of 97% + an average time on site of 11 seconds + direct traffic of 73.29% = Fail. We don't need simultaneous equations here!

Other Weird Stuff

Although the seller claims no revenue in the Claimed Financials section of his Flippa listing, he does mention the following:

Currently making about $70/monthly from Adsense

Currently making about $100/monthly from Affiliate Commissions

There is "proof" of revenue from Clickbank, but of course there is nothing there that identifies the site being sold as the source of that revenue. The screenshot doesn't even tell us what product is being sold. For all the unsuspecting buyer knows, that Clickbank revenue could have been generated on a number of sites the seller owns. The seller might have another site selling the Gargantuan Anal Intruder Version 3 and could simply have presented the revenue stats for that site.Nobody would know.

No proof of Adsense revenue is given.

I find it hard to believe that direct, first time visitors spend so little time on the site because they are rushing to buy the product being promoted there.

When I asked about how it could be that most of the direct traffic was from new visitors, I got the following reply:

I have my website syndicating backlinks from several different types of platforms, for example Press Releases. It's my own personal system that I use to get high traffic and authority baclinks to the site. Which in turns helps to get a new domain name indexed and ranked in the search engines faster.

Plus I use some paid services for traffic (Banner Ads, Press Releases, etc) to help the website build a huge email list for future auto-pilot profits. I spend about $25 per week.

Now that's a bit cryptic, but I would hazard that the seller's "own personal system" to get high traffic will no longer be used when the website changes hands. That traffic appears to be useless anyway. And what about the $25 / week cost that has suddenly materialised. That wasn't advertised in the listing, was it!

The direct traffic/new visits/11 secs per visit combination simply points to worthless paid traffic.

I Can't Read My Google Analytics Stats

The seller has taken the trouble to add Google Analytics to the site and then post Flippa verified stats, but he then spoils it by making stuff up about those stats. Here, they say:

Traffic Details
200+ Unique Visitors Daily - 7,000+ Monthly
90% Organic Free Search Traffic (Google, Yahoo, Bing)
2000+ Backlinks from high authority sites

But the GA stats themselves say 24.54% of traffic is from the search engines. Not 90%.

Additional Lessons

As a matter of course, I checked this site in spyonweb and a variety of other cross checking sites to find other sites displaying the same Adsense code or Google Analytics code. Unfortunately, because the GA code was put on the site only recently, these cross checking tools haven't had chance to crawl the site and compile cross checking data about it.

This discovery raises the possibility of dodgy sellers putting Google Analytics tracking code on a number of their sites only a short time before auction. Doing this will:

  • inflate the traffic stats as they will be reporting traffic numbers for multiple sites and not just the one being sold.
  • make it hard to trace this trickery as the cross checking tools won't have had time to identify other sites with the same Analytics code.

I'm not saying this kind of skullduggery has taken place here. There's enough of that already without the Google Analytics on multiple sites trick being played! It's just one possibility to bear in mind when investigating sites that have only recently had GA added.

Posted in Danger Corner.

3 comments

By admin – October 19, 2010