| This 91 message thread spans 4 pages: < < 91 ( 1 2  4 ) > > || |
|Our Google Ranking Experiment|
In the past, people have been kind enough to share their experiments on WW and some have been quite helpful to us. Thought I'd post the results of our recent work. Roughly one year ago we created 2 sites targeting different sides of the same industry and worked them differently:
Site 1: Bought a brand new domain name that was brandable e.g. "widgets4you.co.uk" (the 4u and RUs thing is very popular in the UK!).
- We built around 70 doorway pages i.e. pages on topics related to the widget being sold and wrote keyword full (not stuffed) articles.
- We bought paid links from directories (e.g Yahoo) and other high PR sites.
- We built a reciprocal (not three way) link directory on site offering link exchanges to anyone in any industry.
- We hired 2 people and had them work full time finding reciprocal link exchanges on other people's sites that were mostly related to us ... though sometimes the connection was tenuous!
- We wrote 10 articles and submitted them to article sites.
- Anchor text was therefore controlled and total links as of last month was 7,151.
Site 2: Bought a 3-year aged domain name on Sedo. The domain was a keyphrase i.e. bluemetalwidgets.co.uk
- We started off adding around 3 pages of unique and informative content a day aimed at our market, any keywords in there were all naturally occurring, not aimed for. After the first 6 months we reduced this to 1-2 pages per day. Again, each article was on current affairs, tips etc. for the industry, totally unique and between 400-800 words.
- We developed press releases and submitted a total of 16 over the last 6 months through PRWeb, PRNewswire and BusinessWire.
- We developed a blog.
- We developed a video news channel now showing 22 videos.
- We started a twitter service of main headlines.
- We started an RSS/email subscription service to our articles.
- We had no link building work, we relied on natural link occurrance (4,374 links as of yesterday) and natural anchor text.
The idea being that one would use regular SEO techniques (reciprocal link building, doorway pages etc.) whilst the other would use actual business methods (press releases, informative user-oriented content etc.) to see which did best.
Site 1: after just 3 months it was skyrocketing past some pretty hefty competition with traffic increasing well each month. The site was making £10,000+ a month for the last six months we had it and just sold for a rather nice figure.
Site 2: has struggled to rank anywhere, even for it's own name, and traffic has been stagnant since the outset - it made a loss for the first 8 months and made just under £3000 in it's best month which was last month.
I don't really want to state conclusions but I hope the information alone is though-provoking ... not least because the reciprocal and paid-link site was ranked in the top 3 consistently by Google (even above the 10 year old and government websites!) whilst the natural linked site never even got off the ground.
I agree with the general ranting sentence wheel gave, but a comment relating to this topic would have been nice.
|What we really need to understand is that Google isn't going to do our marketing work for us. You want something done right, get off your glutes :). |
May be a great sound byte for another thread where lazy people are complaining that their 12 link exchanges haven't got them to the top for everything. But here we're discussing 7000+ forced, sculpted, reciprocal links from lower quality sites completely outwaying the 4000+ natural links from high quality sites.
|The site was probably parked and may have had a history specially being a kw based domain. Somebody probably spammed the crap out of it. |
I've seen this come up a few times in this thread. It was a parked domain and it was one of those with a few hundred keyword parking pages on it. I used WMT to remove all the parked pages from the index and contacted Google under the re-inclusion link about 6 months ago to highlight it's history just in case there was some black mark. As far as I'm aware, 6 months is plenty of time for Google to have responded and removed any filter if there was one?
Hey internetheaven, the Site2 has trust already.
Get some proper-anchor-quality-not-important links and it will be a killer in a few weeks.
I believe it will be much safer site than the one built solely on recips due to better backlink profile.
Anyway, it is all trust + anchor, ins't it?
wanted to say 3 things in regard of this thread and SEO in general:
1. Buy incoming links from high PR sites and pages (site wide links ARE OK) to promote your main pages.
2. For long tails write good content, and lots of it (many just buy the content as well).
3. Use smart and descriptive page titles.
Doing just these 2 things is 90% of all SEO work. All the other stuff (Code rewrite, W3C standarts, index+search listings, press releases, Meta tags, Hx tags, bold text etc..) Is very nice, and shows the client you actually do some work, but they count for very little as far as Google concerns.
Which to me, shows that Google is not much more than a bad joke. If you have money, you buy good links and originak content. If not, well, Google doesnt want you, he wants the Big companies with big bucks - relevance doenst matter (just for long tails no boddy competes for) just money, and for this you need a very simple algorythm....
Thanks for sharing this is a grade post and info :)
I built a site 3 years ago. Great content, actually better than great content - cool, unique, useful for research, interest to everyone in the business. Then I aggressively drove links to it. I got links from .edu and .gov levels. I got links from industry associations that don't link to anyone, ever (the content was that good). And the site ranked. Not quite to the top, but pretty good. Then I sold the site, there's been no SEO since.
And you know how many good quality links that site has attracted since I stopped? 1.
My main site slipped last October to spot #2. Beaten by the dastardly cowards using directory submissions. My site again is all top quality links - 100% hand developed. Yet I'm being beaten by directories. My content is better, but not my rankings. And nobody's linking to my unasked. So how am I going to get back to #1? Not complain about directory submissions. I'm going to throw on some more interesting content and aggressively go after new links.
Your marketing is your link development. And Google counts your marketing. Don't do anything, then you eventually won't rank.
Yes, quality links work. But it's not a guarantee. It only means you generally need less links and we hope it's more stable (which has been my experience). If someone comes along with 10,000 directory links, against your 500 quality links, they're going to outrank you. You have to go out and get another 500 quality links. Waiting for those 500 links to develop on their own is futile. Claiming that you have 500 links and that should be enough is futile. At some point high volumeXlow quality will outrank low volumeXhigh quality.
Very very interesting IH, thanks!
I'm just interested how you managed the reciprocal link building. Did you have a number of links pages i.e. links1.html, links2.html, links3.html pages or did you cater for them throughout your website?
internetheaven, excellent side note about 2 more sites you didn't disclose. Did they make money, and if yes more or less than the 2 disclosed?
|I built a site 3 years ago. |
wheel: ah now! this is where my tin foil hat shines brightly! I have been building sites since 2001 and yes, three years ago both of these sites would have done really, really, really well. Sites I built then, like yours, need very little to keep them at the top.
That's why I posted the results of this experiment of two brand new sites sites started in 2008. That's a whole different game.
|Adsense is for MFAs, article syndication sites and parked domains. |
That comment is kind of silly (and if you believe it is true, then, bluntly put, you don't know what you're talking about).
Internet heaven, that's a very good point.
I currently play around with some edgier stuff to see what ranks. And there is some relatvely easy ways to get rankings these days. Nevertheless, old school white hat on old domains work.
But would I start today with a brand new site and white hat it all the way? Huh. I'm not so sure about that. I think I'd probably not. In fact, I'm releasing a new product in my niche and I'm promoting it two ways - on my existing, old WH site and on a seperate, unrelated more grey hat site. The idea of buying a brand new domain and building it from scratch didn't even cross my mind.
Not sure what we can make of that, when we're at the point where starting from scratch isn't where I would even think of going.
Now that experiment I posted in the supporters section, that is in fact on a new domain, starting from scratch, to be built all white hat in a competitive field. But notice I'm 'experimenting' with it.
Wheel what do you have against directory submissions. Google even encourages submitting to directories. There are a lot of directories that have to approve your link and follow all the white hat rules that Google gives for a directory. It is not black hat at all.
Agree with all that this is a great post.
However, somethings missing here.
The purchased domain (site #2) got both: LINKS and CONTENT, and still did not rank. Site #2 got links naturally. It has more than 1/2 the amount ONE-Way natural links vs. number of exchanged links to Site #1. Site #2 was packed with content, (approximately 750 pages, 600 words per page average, updated daily.) vs. site #1 which only got it's links pages updated.
Remember the Successful Site in 12 Months [webmasterworld.com] classic? Site #2 followed more rules than site #1.
So I further agree with apauto (and others) that the purchased domain (site #2) may have been blacklisted.
Thanks for the discussion. As numerous posts have mentioned, this was not a clearly defined test. With all the variables, it's hard to say what you were testing for. The major difference, IMO, probably comes down to the two full-time people finding links....
|We hired 2 people and had them work full time finding reciprocal link exchanges... |
|We had no link building work, we relied on natural link occurrance... |
Suppose instead, everything else was the same, but this was the link building strategy tested....
|We hired 2 people and had them work full time finding reciprocal link exchanges... |
|We hired 2 people and had them work full time finding one-way inbound links, good quality directory listings, and link exchanges with high-quality sites where the links would benefit users of both sites. |
Probably stacked the other way.
Robert Charlton - "With all the variables, it's hard to say what you were testing for"
Oh come on, it is clear what the strategy was:
One site with heavy aggressive "on the edge" SEO
One site natural, organic and social (albeit quite agressive)
Yes, there are elements that don't make them exactly comparable - the domain issue being one of them, but I am sure this experiment wasn't done to be entered into the science journal 2009 to be quoted by scientists r us "et al".
This was a pretty good long term experiment that was shared to all of us, of which we should be grateful. I am sure the idea would be to see just how different the results would be - that is the key in my opinion.
Again, Robert Charlton - yes of course with a different or more balanced link strategy then the results may be different - hell, combine the 2 strategies and it would be even better. But that is all in the public domain, and rule 1 of standard seo anyway - this was a real long term practical example of what could happen if you follow very different strategies.
The point of this (in my opinion) is to see just how radically different the actual results are, by taking two radically different approaches.
And shouldn't the real issue be to anyone who plays clean, conservative and completely white hat (hate that term, but I mean it as a "stick to your principles" type of thing) that by looking at results of this experiment they could be scratching the surface of the potential for their earnings.
I for one am guilty of never sharing our experiments - but that is exactly what we do, create 2 types of site in many verticals with different strategies and then create hybrids based on the results. I have been doing this for 7 years and my results are identical to the findings of this experiment.
The mainstream view is that breaking the rules in some way gives you less peace of mind, however nobody knows the real rules - Google won't tell you exactly what you should do because it is simply a mathematical algorithm.
"Quality" is determined by maths (according to search engine algos) and of course that can never match what humans see as quality - and once you get your head around that then you will accelerate past the people that say "oh don't do anything that might get you in trouble" like there is someone looking over your shoulder - like the kids at school that daren't play out anywhere away from their house!
This experiment should be important to everyone because it shows that using techniques that are frowned upon - the site succeeded for a long time without being penalised at all, quite the opposite. It is also significant that it outperformed a natural site that also had an aggressive strategy - in a big way. I mean how many people actually spend that many resources on a natural site - I bet not many, and I therefore bet that not many of you are making much cash compared to what you could be making.
That should be giving you food for thought, and ways in which you can utilise that with your own site - sites (create more maybe?)
Anyway what a fantastic post and thanks very much for sharing in so much detail.
|We hired 2 people and had them work full time finding one-way inbound links |
I know of one business that did exactly this. It was slow work - they averaged 20 new one-ways per month between the two of them. That's around one link every two days per full-time person!
But it wasn't all tedium - one side benefit was the networking. Another was the feedback received about the site and how to step it up another notch so it became even more link-worthy.
The results for the site itself were absolutely stellar. It achieved sitelinks on three different generic search terms. It had a top three ranking for a huge number of good terms and a flood of natural traffic. It outranked competitors with 1o times the raw number of backlinks.
And finally, the business itself eventually was sold for a very healthy profit.
The point of our test seems to be being questioned. I wanted to see if Google were at the stage that they could honour their promise that natural is the best way to go. Whether we would be rewarded for allowing our site to develop "naturally" as opposed to "forcing" our rankings. Whether cheap, sculpted linking was still better than free-range quality links.
|So I further agree with apauto (and others) that the purchased domain (site #2) may have been blacklisted. |
That was a thought, but like I said in previous posts, we contacted Google Webmaster Tools (twice, 4 months apart) and our situation has never changed. If domain was the only problem then surely the results would have altered? As far as I've read, Google do read and act on re-inclusion requests ...
|The results for the site itself were absolutely stellar. |
tedster Please could you quote the age of the site and when it started link building in this manner. Also please confirm that they did not buy paid links as part of their campaign.
|It outranked competitors with 1o times the raw number of backlinks. |
That's what is above me, stuff with 0 backlinks but part of a much older site than mine and/or with 100% sculpted anchor text.
The problem with the failing site (I thought) was that the free-range anchor text was "click here", "check it out", "example.com", http://www.example.com/folder", images with no alt text and so on. That outwayed/devalued the rest of the anchor text that contained the keywords. The theory being that:
- 50 proper anchor texts plus 50 "http://www.example.com" anchor texts
can be beaten by
- 10 proper anchor texts
as each non-specific linking phrase actually damages the anchor text profile.
The site was 5 years old when they hired the one-way link staff - they were already doing pretty well. The results I mention were what they achieved after 8 months of actively networking for one-way links. The only link that cost them any money was the Yahoo Directory, but that was in place before the two employees started work.
Good story. I'd like to know what did the sites giving out the one-way links get? The link team must be pretty talented to get so many one-way links? Where do you find this talent?
wheel's competitor also found great success from one-way links via directory submissions.
After reading your and wheel's stories, I am torn between hiring one-way link builders for non-directory links and buying directory links. Paying somebody to get one-way links by networking isn't too different from paying directory editors for links.
I am wondering about paying the same average amount of directory listings to a non-directory site. Paying $50 for one link from an editor of a directory isn't very different if at all from paying a webmaster $50 for a link. There is human review with discretion and payment in both cases.
When the debate first came up between paid directories and buying links, the comments from Cutts/Google were basically saying paid directories are okay if they are quality directories.
The webmaster of a quality site is going to be careful about putting any links on that site just like a quality directory owner.
Matt Cutts said [mattcutts.com] the big issue is: "As someone working on quality and relevance at Google, my bottom-line concern is clean and relevant search results on Google..."
This goal is met paying good sites for links.
Q: Hey, as long as we’re talking about directories, can you talk about the role of directories, some of whom charge for a reviewer to evaluate them?
A: I’ll try to give a few rules of thumb to think about when looking at a directory. When considering submitting to a directory, I’d ask questions like:
- Does the directory reject urls? If every url passes a review, the directory gets closer to just a list of links or a free-for-all link site.
- What is the quality of urls in the directory? Suppose a site rejects 25% of submissions, but the urls that are accepted/listed are still quite low-quality or spammy. That doesn’t speak well to the quality of the directory.
- If there is a fee, what’s the purpose of the fee? For a high-quality directory, the fee is primarily for the time/effort for someone to do a genuine evaluation of a url or site.
Those are a few factors I’d consider. If you put on your user hat and ask “Does this seem like a high-quality directory to me?” you can usually get a pretty good sense as well, or ask a few friends for their take on a particular directory.
If anyone can beat wheel with paid directories, they must be pretty valuable.
[edited by: potentialgeek at 4:50 am (utc) on April 22, 2009]
Essentially they got a great resource to recommend to their visitors - something they could not easily duplicate. I know, it sounds a bit trite, but that's the way it went down. That's how a lot of really great websites got their backlinks - by offering something incredibly "link-worthy".
Sometimes, the linking site also got free SEO advice, since the link team was really doing their homework ahead of time and could offer some solid recommendations. Yes, this link team was very astute, well paid and well trained. They were on the PHONE a lot, not pumping out tons of emails.
this post has raised a great response and may highlight the importance of controlling anchor text over letting your links point naturally. nonetheless, he had more optimized links.
>>>> Anchor text was therefore controlled and total links as of last month was 7,151.
I would of thought adding 7,000+ links on a new site over a 3 months period would trigger some kind of penalty.
internetheaven - great topic.
I would be very interested to see how a 3 month campaign of link-building affects the second site (the one that had used natural links plus content only)
My experiences in the past 9 years seems to also indicate that - at this point at least - planned link building gets better results than anything else. On my older sites, up til about 3 years ago, content and natural link building led to fantastic growth. I have several newer sites. On 2 of them, there was a deliberate campaign of planned link-building. Two other comparable sites (in terms of amount of content, size, etc) did not have this (budget issues) and the numbers seem to have suffered as a result, even though new content is added to all the sites regularly.
It's possible that with a planned campaign of link building, you control the exact name of the links, how many for each page, etc, whereas natural links can lead to 404s when people write in the wrong url, overuse of terms, etc.
Anyway, thanks for the post, Please let us know about future trials.
internetheaven - great topic ..
as for the revenue/profit ... these varies on the keyword I think if you use contextual advertising ...
I need to know which one was giving stable profit ? Site 1 or Site 2
|That's how a lot of really great websites got their backlinks - by offering something incredibly "link-worthy". |
Sometimes, the linking site also got free SEO advice, since the link team was really doing their homework ahead of time and could offer some solid recommendations.
Sure it paid off, if you visit their site and tell them where they can improve, not just "I visited your great website", they would also get the feeling that linking to your site would help them improve...
IH, was there a major difference in user behavior on either site? Did the site doing better have higher pageviews, time on site and lower bounce rates?
Pageviews were pretty much the same on both sites -- 2.1/2.46 average.
Time on site was slightly different -- 1.03/1.46 -- but most did not result in a click back to results, they moved to an untracked page on our site (payment page or secure admin area).
The bounce rate was quite different -- 62%/41% -- but surely the bounce rate is not such a dramatic algorithm factor that it creates the difference between fantastic success and total failure?
"Build it and they will come" hehe
[edited by: Makaveli2007 at 11:25 am (utc) on June 11, 2009]
"Build it well, get linked to naturally and some will come ... but Google won't ..."
"Build an MFA, get reciprocal links and Google will come ... which means everyone else will too ..."
When you say the first site made £10,000 per month, do you mean purely from Adsense ads?
|When you say the first site made £10,000 per month, do you mean purely from Adsense ads? |
I'm not sure what that has to do with the experiment, but "no". Adsense was not used as the revenue generator on either site --- if I had it would probably have been £100 per month instead of £10,000!
Internetheaven nice controlled work. Makes total sense too. I know of a business person in a predominately US genre that follows this same format in a very competitive genre. However, he leans very heavily on the paid links. After 4-5 websites and I am sure lots of earnings it really makes you rethink about the effectiveness of the algorithm as a whole, and how some smaller mom and pop websites get tossed out from time to time with the bathwater.
| This 91 message thread spans 4 pages: < < 91 ( 1 2  4 ) > > |