| This 31 message thread spans 2 pages: 31 (  2 ) > > || |
|High Links-High On-Page Optimization--Low Rank|
Re-Optimized and Dropped in Ranking
Looking for feedback and experiences with similar situation to mine:
I was and have been ranked #2 for a keyword term for some time. But this month a competitor outranked me and dropped me to #3. So, I did a keyword density analysis of our 2 sites, and raised my keyword density up to be in line with theirs. I dropped to #6.
~Here are other factors:
~My PR: 5
~My site backlinks: 497
~The other sites (other than #1, see below) backlinks now ahead of me: between 24-50
~PR of other sites now ahead of me: PR4, and one PR5
~allinanchor: I'm #2
~allintext, allintitle: I'm #3
Did I over-optimize when I increased my keyword density (which to match my competitors, I raised to around 10% and 8% for each word in the keyphrase)?
#1 for the keyword phrase has makes sense with:
~over 1,000 backlinks
~zero keyword density
~allinanchor, allintext (though nowhere on site--and clean site), and allintitle: #1
Today, I under-optimized on-page factors to see what happens on next crawl.
That's interesting. Right now all kinds of odd stuff happens with on-page it seems - or perhaps the true explanation is something else and the timing is just playing tricks on us. Some see zero effects and contribute all to links while others see positive effects. I've not seen any negative effects sofar, it will be very interesting to see if your reduction of on-page optimization will change your ranking, and in which direction.
Here we see again that on-page optimization is not as important as off-page.
|Here we see again that on-page optimization is not as important as off-page. |
I respectfully disagree with that. All WebMistress said about her/his onpage was that the keyword density increased- that's bogus. Keyword density is not a considerable factor anymore, and if that's the reason you seize upon then I have serious doubts about your on-page prowess (no disrespect intended).
Proximity to the top, bold, proper use of italics, h1, h2, proper use of css for unessential bold/italics, keywords used together and apart, location of links within the pages- all that goes into it.
If you're worrying about keyword density I would respectfully suggest to forget about it and to look into the other factors I named above instead.
It's like poker, you have to have the cards, onpage and offpage- different cards and combinations are worth more. PR doesn't matter as much anymore if you can hit the on-page factors in addition to the insite factors, and the link factors. Ok, it's a mouthful...
Martinibuster's post is bang on the mark.
Personally I am a fan of keyword density.. because it is something I can easily analyse and manipulate.
But here's the key if you seriously want to do it: never never ever base keyword density on what a competitor is using.
As one of around 100 factors, you cannot possibly clearly identify and equalise all other factors between two separate sites (one of which you don't control) to effectively analyse it.
You can, however, easily identify that all other factors are equal between two pages on your own site on, say, 'large blue widgets' and 'small cyan widgets', - keeping the pages as close to identical as possible, no of incoming links, positions in navigation, etc, but varying the density marginally... the one that comes out on top of the serps relative to the other one will be closer to the 'ideal' density.
It's that "all other things being equal" part that's important - the only way you can make sure of that is to look at your own site. Forget your competitors in this case.
>> that's bogus
These terms are perhaps a bit strong, especially as WebMistress did increase her KW-density after all ;)
Anyway, on-page link text, h1, h2, <b>, etc also affect keyword density, as you typically use your keywords in them (that is, if they were not there before), so it's a bit hard to isolate that factor.
As those other factors do seem to be more efficient in most cases, it's interesting if KW-density can influence ranking on it's own, even if negatively.
allinanchor: I'm #2 allintext, allintitle: I'm #3
Looks like it's a very close run thing between Nos. 2 and 3, your "drop" may be due to some new factor in the competitors site rather than anything you did.
G may have even turned down the link text knob slightly ;-)
Nothing has changed EXCEPT my keyword density...all h1, h2, bold, proximity issues stayed the same, as well as great anchor text backlinks (mine over 400, my competitors I speak of, below 50)....the ONLY change to my site or theirs (and I keep close tabs on theirs) is an increase in KW density to my site. I'm thinking that because I have so many backlinks, compared to my competitors, and I have perfect anchor text, that those two things added to high kw density throws me down...as if over-optimized. Clearly if you read what I wrote about my #1 competitor, who I am not even trying to beat, their site has the backlinks factor and zero kw density. I guess we'll only know when the results change. I'll let you all know what happens. It will be a valuable experiment at the very least for all of us.
Ok, here's how it turned out:
When I brought down my keyword density to the lowest I have ever had it, I went to the top!
So, for my site at least, it seems because I am highly optimized in off-page factors, I do better with lower on-page optimization. Overall Over-optimization did, for me it seems, drop me in the rankings.
Hope my experiment helps someone else.
Nice find. I've been expirimenting with this same thing myself.
You changed position within one day of tweaking content? Report back in a few days. I have my suspicions your boost was due to what is perceived as fresh content, and not any on-page seo magic.
Were you watching for googlebot during this time period?
I changed within a couple of days when I upped the keyword density, and then I changed within a couple of days when I lowered it back down. I am still stable at the top. I have no doubts that my keyword density changes were the reason for the changes, as no other changes were made. I have seen re-optimized internal pages change within days as well and remain stable for weeks so far.
Google visits my site daily, so, no I didn't wait for him to visit as I know he visits everyday.
Oh, and to address the freshtag issue...I have remained stable at #1 both with the freshtags and when those freshtags disappear, and then when they reappear.
So can you do the same experiment again? Put the old page back up, and drop to 6, then swap back again?
All experiments have to be repeatible. If you can do that I will be convinced :)
So many factors involved, it is often difficult to see.
George - Be convinced. I too noticed this same thing on an important page within my site a little over a month ago and decided to de-optimize my on page factors. (keyword density) I had it up to 22% and was in #3 for a long time until a sudden drop to #17. Brought density down to 10% on my target phrase. Two days later - up to #5. I have repeated this experiment three times in the last three weeks with the same results. The drop will vary a couple places depending on fresh results for other sites, but always returns to #5 despite other fresh results or even my own. I watch my log files all the time for googlebot who is as faithful as mans best friend.
Thanks Rogue for supplying the results with a repeat experiement. I personally do not want to repeat the experiment because I have no doubt I will drop again, and it affects my income. So, I'll not be repeating it :)
cripes. I truly truly hate it when people get stuck on semantics and obscure the topic with debate over a word, but in this case I think it matters.
Folks, increasing keyword density is NOT 'optimising'.
Moving closer to Google's* ideal density is 'optimising'.
* Insert SE of choice.
If your density is higher than Google's ideal, then REDUCING your keyword density is optimising it. If your density is lower, then increasing is optimising it. And yes, we can only guess at what 'ideal' is, but think 8-12% for a starting point.
Please, please get it out of your heads that to increase any factor is necessarily "optimising". It will make a huge difference to the way you look at your optimisation work.
Point noted. But at the time we make these changes we are, or at least we think we are, optimizing. It doesn't always turn out to have optimized though, does it? But I think we all understand when we say "optimizing" that we are merely trying to optimize. And how else to really optimize except to take risks such as lowering or raising keyword density, and tweaking a whole bunch of other factors. In the end, although I had to play with KW density in both directions, I reached my goal--I OPTIMIZED. So, I guess you could say optimizing is the process, and optimized is when you reach the goal.
So what are you suggesting WebMistress?
I have a similar problem, I didn't fall in rankings but the whole effort of changing eveything went waste as I didn't make to the top tho I am way far away from my competition.
- what is the conclusion?
Any one hinting that - what we were doing till now "as optimization" is simply ignored by the Google now?
funny: page jumps after letting it rest for 3 weeks
I was trying very hard to get to the top for a quite competitive one-word-phrase which is related to my site.
The second day after I created the page I jumped to around 50 in about 40000. And then I could try everything in the book (though only reading white pages) - it would stay there. Fresh tags come and go, position stays the same.
And now - about 3 weeks after I almost gave in and didn't change anything on the page, it jumped to #6 (no fresh tag).
Would you attribute this to an algo change, or did they implement something like a "timer"?
When I read this topic, I started looking around a bit, and could find no proof of your theory WebMistress. Not that I do not believe you, I just seem to find the opposite for everything I researched.
For example, let's take the highly competitive keywords "search engine optimisation". When I'm looking at the top result, I can't find any cloaking or sneaky JS-techniques, so I assume the page is ranked high on links and content alone. PR is only 6.
They have about 500 backward links, most of which have NOTHING to do with SEO (so a possible thematical bonus can be excluded), and none of them have spectacular PR.
The keyword density is about 30%, which is extremely high, and possibly the reason they're #1.
Looking at the second site, they have about 200 more backlinks and a PR of 7.
Now why isn't the 2nd site on top? Imo this is because of on-page factors, and I think the 30% keyword density has A LOT to do with that, especially since the 2nd site has a WAY lower density.
Or am I missing something? If not, I would assume a high keyword density is still very valuable.
|I would assume a high keyword density is still very valuable. |
Yes you got it right sir, if I fall below 18% I have sleepless nights, I have to maintain 20% to stay on the first page.
|Now why isn't the 2nd site on top? |
I am sure that keyword density is a factor in general, but for this example the main reason that site is at #1 is anchor text backlinks (for which it has more than anyone else for that keyphrase).
>> Yes you got it right sir, if I fall below 18% I have sleepless nights, I have to maintain 20% to stay on the first page.
Personally, I think >15% is a tad high, even if it really helps your rankings. In general, even topical white papers never have a keyword density higher than 5%!
I imagine G punishing a site one day if it contains more than a certain percentage key phrase density.
>> I am sure that keyword density is a factor in general, but for this example the main reason that site is at #1 is anchor text backlinks (for which it has more than anyone else for that keyphrase).
You're right, I overlooked that. But still, it sure has a lot of links containing "search engine optimization", but haven't seen much containing "optimisation".
Do you think G will recognise them as being almost the same as the search phrase, therefor ranking the page higher for the phrase?
|Do you think G will recognise them as being almost the same as the search phrase, therefor ranking the page higher for the phrase? |
Google finds a high occurance of the phrase "search engine" in the anchor text, and a high occurance of the word "optimisation" on page.
It could be that which is doing it, or you could be right - google recognises differences in UK and US English.
Not sure which. I'm curious to know if you're right though, if anyone can confirm.
Site 1 is in the Google directory which I think is very important.
As for anchor text containing 'optimisation'. I'm not going to check as AlltheWeb reports 16000 and 83000 backlinks for sites 1 & 2 respecively.
I think there are so many factors obviously at play when it comes to ranking. I just know I had been at #1, then a site took #1, and so I brought my keyword up to their level, and I dropped to lower than I had been since I first got indexed in February. I let it stay that way for a couple of weeks thinking it would bounce back up, but it didn't. I lowered the keyword density and it shot to #1. You'd think since I have over 400 backlinks with great anchor text that this other site with only 40 backlinks couldn't overtake me, but they did, and they did it with high kw density. So, that is proof that high kw density will take you a long way in ranking. However, at some point, it appears, if kw density is high and all other optimization factors are high, the high kw density may hurt a site. I really don't know what else explains the effect raising the kw density had on my site. But I thought it was worth sharing. I'm not trying to prove any theory; I'm just sharing my experience for everyone's benefit, and especially for anyone else who is experiencing something similar.
I think if all on and off page factors are all high then G thinks the webmaster is too high and knocks them down a peg or two.
Thanks for this thread - it set me in the right direction on something that I am working on at the moment.
I have a new area on one of my sites and am targetting a two word and and a secondary, less competitive three word phrase on one of the main pages. I immediately jumped into the index at about 35 for the three word phrase and 502 for the main two word phrase. All OK so time to get some links - after getting a few good links three word phrase goes to 25 and two word phrase goes to 581. More links added and three word phrase at 19 two word phrase goes to 620 and then 681. On friday after reading this thread I reduced the keyword density of the two word phrase and hey presto have shot up to 379. Finally moving in the right direction...
|Finally moving in the right direction... |
Out of curiousity, after you tweaked the keyword density, did the page also read better?
| This 31 message thread spans 2 pages: 31 (  2 ) > > |