| 7:51 pm on Jul 22, 2004 (gmt 0)|
1. phrases in file names
2. alt text
One and two are pretty well discussed.
Could you expand on #3?
| 8:38 pm on Jul 22, 2004 (gmt 0)|
Sites without alt tags for non-text elements may break the accessibility requirements of DDA (Disability Descrimination) Act in the UK.
| 9:02 pm on Jul 22, 2004 (gmt 0)|
I drank a Russian beer one night and all my sites were ranked for all kinds of good terms.
Seriously you can't really say stuff like that there are just too many factors at play. I do all those things and rank real well with PR5 websites. Get the right backlinks quality can beat out quanity.
| 9:04 pm on Jul 22, 2004 (gmt 0)|
I too would like to know if you made any other substantial changes in addition to "phrases in filename & <alt> text".
I use both of these in my sites and G has never had a problem with giving me top rankings for competitive phrases.
So your situation makes me wonder if the improvement in rankings was coincidental, or due to some other changes you made.
Phrases in filenames can be helpful to users. They show up in the search results, and a user can see this as a potentially more relevant result to his search.
IMO I'd be surprised if G was penalizing these sites based on the two tactics you discussed.
Can you share more about your De-optimizing?
| 9:14 pm on Jul 22, 2004 (gmt 0)|
All that fun and games did was get those sites past the OOP filters. ;-)
| 9:25 pm on Jul 22, 2004 (gmt 0)|
3. the most notorious H1, H2 ...
| 9:43 pm on Jul 22, 2004 (gmt 0)|
How bout this, you changed the website and the engine saw it updated :) Updating a factor in ranking?
| 10:26 pm on Jul 22, 2004 (gmt 0)|
Phrases in filenames existed long before SEOs or even search engines came along. They made sense then (and continue to make sense now) because it's natural and convenient to name a page about jelly doughnuts "jelly-doughnuts.htm" or "jelly_doughnuts.htm." Google obviously agrees with me on this, because it often uses filenames like "guidelines.htm," "seo.htm," and "webalerts.htm."
It's also unlikely that Google would look askance at the use of H1 or H2 tags, because those are common (and recommended) elements in an HTML document's structure.
Google might be less forgiving of keywords in alt text, because alt text isn't there for SEO purposes--it exists for users with text browsers or screen readers. Still, I doubt if Google penalizes for alt="****x ****x xxxx" keyword packing; it's more likely that Google simply ignores such text.
Of course, if you're doing other things that might trigger an "excessive SEO"" with Google, behavior that isn't suspicious in itself (e.g., using descriptive filenames) might be regarded in a different light. It wouldn't be unreasonable for Google to think that the presence of the exact same multiple-word keyphrase in the anchor text, filename, title, and 20 other places on the page might be a little suspicious--especially if it's on a quadruple-hyphenated domain that's part of a questionable link network. Such a combination of factors, which is statistically unlikely to be "organic" or "natural," would be reason enough for Google to impose a penalty or (more likely) remove whatever ranking advantage some of those factors normally might bring to the page.
| 12:41 am on Jul 23, 2004 (gmt 0)|
|I removed it all and within TWO weeks, every single page that was edited was at least top 20 |
It would most likely have happened anyway. Many sites & serps do jump around from Page 1 to Page 100 and back again over the course of a few weeks
| 12:42 am on Jul 23, 2004 (gmt 0)|
|De-optimizing is a fast way to get better ranked. Try it, you will be pleased |
Is that the most unquantified thing I've read all day? Probably
| 1:15 am on Jul 23, 2004 (gmt 0)|
|it's more likely that Google simply ignores such text. |
I find that hard to believe. The text that shows on Google's search results are from the ALT on a lot of my sites.
De-optimizing works? Right and I just made more money than Bill Gates last night. WARNING If you don't have a site that's just for messing around with please don't try this at home!
| 2:11 am on Jul 23, 2004 (gmt 0)|
"Is that the most unquantified thing I've read all day?"
My neighbor's cat hacked up a furball, and I moved up two spots for widgets sitemap.
Am I in the running?
| 2:14 am on Jul 23, 2004 (gmt 0)|
Ogletree, can you be more specific about the Russian beer that improved your rankings? Have you done any research on other Eastern European brews, and whether lighter or darker varieties seem to be more effective?
| 2:48 am on Jul 23, 2004 (gmt 0)|
>>My neighbor's cat hacked up a furball, and I moved up two spots for widgets sitemap.
One of the kitties I was foster mothering for a while decided to use the dining room floor for a catbox, but unfortunately it didn't help me at all. Maybe I shouldn't have brushed her as often, it seems hairballs work better than potty.
>>Ogletree, can you be more specific about the Russian beer that improved your rankings?
Personally I prefer Belgian dark ale rogerd, it's the nectar of the gods - but there are many beverages that work equally as well.
I had some hot toddies with tea, honey and brandy for medicinal purposes when I had bronchitis, and cut down on the excessive number of occurrences of the same identical keyword phrase on some pages. There's a magic number that only NFFC and I know but won't ever tell anyone, but if a page sounds repetitive and spammy when read aloud, plus has the phrase in things not in visible page copy, like outbound anchor text and alt attributes, it's OK for Yahoo but not for Google.
Thanks for sharing it with us funandgames; I don't know about specifically calling it an OOP filter, but caveman, as usual, has got it.
You have to watch word order and proximity - and the number of occurrences of both the individual words and the words used in exact phrases. That's on-page. Then there's excessive use of identical anchor text in inbound links, but maybe we shouldn't go there. ;)
| 3:36 am on Jul 23, 2004 (gmt 0)|
| 6:29 am on Jul 23, 2004 (gmt 0)|
Sometimes I wonder why I post tips here. Seems like the flames are getting hotter each time. Perhaps I should sneak out before I get burned........
| 6:53 am on Jul 23, 2004 (gmt 0)|
funandgames, posting ideas or suggestions is great, but when you post something as declarative as "de-optimizing is a fast way to get better ranked" with "evidence" that anyone can contradict in one second with a contrary example, then you should expect people to clearly disagree with you.
| 6:57 am on Jul 23, 2004 (gmt 0)|
>>> Sometimes I wonder why I post tips here. Seems like the flames are getting hotter each time. Perhaps I should sneak out before I get burned........
funandgames, that should not discourage you actually...it is a sort of open discussion; many people will believe you, while many will not.
What you posts does make sense, so there are more arguments! That's a reaction, friend...
| 7:14 am on Jul 23, 2004 (gmt 0)|
Well done on your success and thankyou for sharing it with us. A lot of critical posts have a lot less qualified comment than yours... after all, you're rankings improved which means your actions may have done something right. Who are they to be so arrogant? They may be getting good rankings despite their poor on page seo, for a number of other reasons. There must be something you did which helped, I don't believe in luck, you did something right.
The reality is that the mix of tactics is the key. Pages that rank well are probably there because they have got it 50% right. Others seem to assume that because their pages are high ranking anything others do, which they don't, must be wrong.
You may have got it absolutely right, you may have accidently got it right purely by refreshing the file date on the pages, you may be wrong and it was pure chance, tough this is least likely. Who knows? The google algo is not known by any of the critical posters, their ego's are massaged with clever jibes, but ultimately they contribute nothing to the forum or their own potential by being so dismissive.
Your post may just be another part of the jigsaw, and with constructive debate we may be able to dig deeper. There will be many who have logged your experience and benefit in putting the jigsaw together. Thanks for posting.
|troels nybo nielsen|
| 7:52 am on Jul 23, 2004 (gmt 0)|
I wonder if my understanding of the English language is even worse than I thought. In Danish "optimal" means something like "exactly right", not "as much as I believe or at least hope that I can get away with".
Since yesterday I had a significant drop for one of my vanity words. It may be "excessive use of identical anchor text in inbound links". It may be Google still getting temporarily confused by the changes I have made this year with several hundred pages getting new URLs, many of them on new domains. (And thousands of links getting changed accordingly.) It may be my forgetting to light a candle last night. Or it may be something that I have not thought about. I don't know.
But what I do know is that I will not change anything because of this drop. (That may be temporary.) I will keep doing what I have done for a long time: Create new content. Improve old content. Make small tweaks in linking structure, page design and all the other technicalities where it is possible to make my websites more user friendly.
Banalities? Yes, indeed. But that is the kind of reality that I have chosen to live in as a webmaster. Others may choose differently.
| 7:59 am on Jul 23, 2004 (gmt 0)|
It's the derisive sarcasm that was uncalled for.
| 8:09 am on Jul 23, 2004 (gmt 0)|
I doubt that chaging the things you mentioned really had an impact.
I will try the Russian beer later :)
| 8:51 am on Jul 23, 2004 (gmt 0)|
Well said Patrick.
The true definition of the word 'optimise' and 'de-optimise' is a debate we have had before and is a total red herring. Lets move on and accept it refers to the reduction of previously used tactics, whether you agree with the theory or not. The phrase 'de-optimise' refers to removing a combination of seo tactics which may, if done in unison, trigger an algo. The theory is that Google is looking for 'seo'd' pages. They won't be offended by any one tactic, but in combination this warrents a closer look at the page. Thus all the previous seo tactics still work, but too many together is asking for trouble.
The comparison may be made to 'circumstantial evidence'. No one peice of evidence is proof, but in combination a case is made. I believe google is doing this. They may have made the observation that pages with w, x, y and z tactics are 90% produced by seo webmasters and are often low quality pages. If further investigation proves their links in are also low quality, the page is dumped. This may be a side effect of hilltop rather than a specific filter. The combination of seo tactics ranks the page initially high. Hilltop is applied to the top ranking pages and they fail the test and are dumped. Meanwhile, the lower pages which appeared in the initial search, and are still relevant but did not rank highly through clever seo, rise up and take their place.
In other words, use too many seo tricks and you will have hilltop to answer to. If you survive hilltop, your page is genuinely relevant, even if you have used all the tricks in the book. If you fail hilltop you are out. If your page is not seo'd but still appears in the initial search, all be it lower down, you can achieve good rankings because others above you will fail the hilltop test. This is because google accepts that relevant pages do not always rank well and despite being good quality they are pushed down by seo'd sites of lesser quality. Therefore, subject the high ranking sites to hilltop and remove the pages that know more about seo than the subject they are about! By 'de-optimising', you lower your ranking in the initial search google performs and are thus not subject to hilltop. You will ultimately not rank as well as heavily seo'd sites that also pass the hilltop test, but the net result is a possible higher ranking than if you had been subject to the hilltop filter.
I know the above is only theory, but maybe it sparks some ideas more constructive than thoughts of cats being sick :) Thats the point of this forum... isn't it?
| 9:09 am on Jul 23, 2004 (gmt 0)|
lets not get too thin skinned.I am not sure that funandgames was too insulted.His reply suggest that he was behaving in the spirit of fun and games.
| 9:32 am on Jul 23, 2004 (gmt 0)|
>>It's the derisive sarcasm that was uncalled for.
Hence the deliberate retorts that followed. It's all about respect, especially respect for good intentions. This is about sharing, it isn't about the Poo-bahs vs. the Peons. I don't know about anyone else, but poobahs don't impress me, especially self-proclaimed ones; intentions do.
Moving along back to the subject in hand, there *were* certain things that could be added and lessened that helped after Florida, and we're not done with all that yet.
It may be circumstantial, but when two sites that got hit have done the same things and neither is touched until long after all the others came back (and those didn't) - and then one is fixed and comes back and the other is still in Google_hell - anecdotal is not as worthless as some would like to make it out to be.
A lot of what we continue to do is based on anecdotal evidence, and if we keep doing it and it keeps working - we'd be nuts not to continue to do it without having done "scientifically proven" research.
Every tip helps - and thanks again for being willing to share.
| 11:13 am on Jul 23, 2004 (gmt 0)|
i would be very interested in the post you made earlier about optimal keyword density, proximity to other elements etc. Is this something you have had sucess in changing?
| 11:44 am on Jul 23, 2004 (gmt 0)|
I have a sub-domain with 40 pages. 36 of the pages are number one for their phrase. Each page is named keyword-keyword.html up to a maximum of 5 keywords.
| 11:47 am on Jul 23, 2004 (gmt 0)|
jsnow, I've participated in discussions but don't recall doing a specific post. But funandgames posted something on number of occurrences not too long ago
| 4:40 pm on Jul 23, 2004 (gmt 0)|
I think what bothered people and caused some of the derision was this line "I decided to take a look and saw several 'uneccessary' tactics such as phrases in filenames, <alt> text etc."
Both of these things can be abused, but he didn't say that he "fixed the abuse", instead he made it sound like those were bad things that must never be done and that Google specifically penalizes for those things.
In fact, if you use ALT text properly for accessability purposes, it should go a long way towards keeping you clear of the unnatural content filter.
Descriptive filenames are also a good thing, but it is also probably considered in any code that looks for unnatural content.
I think he would have done much better by stating something like "tactics such as keyword stuffed alt text and filenames."
| This 103 message thread spans 4 pages: 103 (  2 3 4 ) > > |