Updating Question - no ranking movement despite on-page changes

[MrFewkes]

Hi,

Ive been making changes to sites now based on various sources of information. The objective of the changes is to acheive a better rank.

The changes (done separately - then wait for cache update) and monitored over time (about 4 weeks) have been made on 7 domains of mine.

The changes are split into 3 types.

1) I changed some sites to add the keywords to them inside various HTML tags which I was informed was helpful. So for example I added an address tag with my keywords in it (just an example).

2) I added certain words (not my pink widget words) to my pages which I was advised google required so that it could place my site in the correct "category" - a "Selling" site that is - for pink widget. Rather than an "information" site for pink widget.

3) I added my pink widget keywords to *new* anchor text on my index page which links to existing (and indexed) subpages containing the keywords.

All my changes are cached now - the index page of each site which is where all the changes took place across all the domains are all pr1 to pr3 (just to note)

I do believe that the above changes ARE ranking factors (please please please please correct me if you think im wrong).

The sites are all ranking between position 4 and 80 (so not under any penalties I can see).

Other info - no affiliate links on the sites, pure oscommerce + hand written index page (which im tweaking).

The main point is - non of the sites have moved in rank - AT ALL!

I feel like im flogging a dead horse. Ive spent weeks reading and reading and got nowhere.

Has google not yet factored in my changes maybe?

Cheers - any advice appreciated.

[tedster]

There are a number of possibilities - and one major factor would be how critical off-site factors are for that particular query term.

In a one query term economy, there may be a decent number of backlinks for the top ranked URLs, and then on-page changes may not move the needle in any noticeable way. For other query terms, on-page factors could be huge.

For analysis, I would encourage you to think in terms of three buckets, not two, for SEO factors:

on-page (title, headers, alt attributes, text content, and internal linking of various types)
on-site (internal backlinks to the target URL and their anchor text)
off-site (external backlinks and anchor text)

If you are waiting 4 weeks after apparent caching, I'd say that's plenty long enough to assume the changes are incorporated into Google's ranking analysis. One of the areas they talked about with their new Caffeine infrastructure was the ability to make this kind of change faster than it was historically -- updating all of their internal meta-data for any URL almost as soon as they get new crawl information.

[MrFewkes]

Tedster,

Thanks for the reply.

My aim is to take on the on-page with this I think - first of all. Im quite happy with my links at present for off-site.

Then - I will look at on-site factors aswell as you say.

I am certain at this time in this products life (its new) that backlinks arent really drowning out on-site or on-page factors just yet to any major extent. Most sites listed have few or spammy links and pretty much all page 1 serp sites are PR0 (not grey - pr0) so thats a fair indicator of its current standing in competetive terms.

G returns 1.9 million just for interest.

On one of the phrases, one guy has 8 domains in the top ten for the same phrase, all indexes - all pr0. So there is (in my mind) great scope for getting all the answers I need by analysing his pages and sites.

Am I on the right track - given that I know now from your post that G updates faster than I had been perceiving.

Cheers

[tedster]

The most important on-page element is the title, so make sure that one is nailed down. Then look at the rest of them, including anchor text as an on-page element.

[MrFewkes]

My title is this.
Pink Widget | Buy Pink Widget | Free Pink Widget Delivery

So many times I do hear that the title is the most important item - but I do struggle to understand exactly what to place there. The above structure seems to work for my competitor.

The following statement is true :-
"Include your keywords in your title"

But thats not enough is it? Im NEVER sure what else to put - as the keywords alone arent enough.

Am I on the right track?

I appreciate your input.

[tedster]

That particular title may even be too keyword stuffed (I'd limit it to 2 times) - but yes, that is the main idea.

[htdawg]

@tedster
for the pink widget title example MrFewkes gave, i can imagine that most of the serp results all have at least either the word widget or widgets in the title and or the description atleast a couple of times and you start to see a pattern in what pages google shows for that phrase which makes you wonder if that say 8 out of the ten results has my main phrase atleast a couple of times in the title and the rank well for it so they must be right. and sometimes keyword stuffing/density works but then again there are countless other factors that come into play for ranking.

[freejung]

I've also noticed a lack of movement lately for on-page tweaks that I would have expected to make a difference, say 5 years ago.

I think that once you establish a certain baseline level of relevance (that is, once Google understands clearly that your page is indeed selling Pink Widgets, which I'm sure is clear from your title and other obvious factors like body text), after that it's pretty much down to off-page factors. Do your on-page tweaks really make the page more obviously about selling pink widgets than it already is? Seems unlikely. The on-site stuff like information architecture matters, but I think especially for competitive terms where all of the top sites have at least simi-decent SEO, your ranking is going to be primarily determined by off-site factors.

[freejung]

Another way of putting this is that I think on-page SEO these days is sort of like "you must be at least this tall to ride."

You should make sure it's clear to search engines what terms you are targeting. That means KW in title, body, alt, anchor, yada yada. You should make sure your site is on-theme and content-rich, has no technical problems, and has sensible navigational structure.

If you don't do those things, you're not likely to rank in most cases. If you do those things, you might rank.. Where you rank will then be mostly determined by off-site factors (PR, Trust, Authority etc).

As tedster said, there may be some terms where on-site factors will matter - those are probably long-tail, low-competition terms where just being relevant at all is good enough to rank.

[MrFewkes]

K.

The competitor sites generally have pink widget 3 times or more in the title - so I "felt" i had to do the same - dont ask me why :)

His top ranking site however has pink widget in the delimited title 4 times - the first being on its own - like this.

Pink Widget | Buy Pink Widget | Pink Widget **word** | Pink Widget **word** | **word** Pink Widget - $xx.xx **word** **word**

So - whilst I totally 100% understand where Tedster is coming from in the stuffing angle - we can see by example and therefore proof that google does not appear to dislike this to the point where it would punish a site for doing it above all other factors. Not with this keyphrase anyway for now.

I suppose my question stands though but needs putting another way now.

Lets assume that a sites index page contains ZERO optimisation for a phrase by means of the following (on-page) :-

In title but....
Not in description
Not in alt
Not in Anchors
Not in H tags
Not in image names
Mentioned in body text say - two times for good measure.

Lets say that this situation was cached and had been for say 6 months.

Then - I come along one day and fill the pages image names, h tags, alt tags, anchors and description etc nicely with the key phrase mentioned only previously in the title and body.

For this example - can we assume that other sites have all 100% exact offsite and onsite factors equal to each other and they are equal to mine aswell.

So - all things equal - im saying that my pages are not moving when I add my words in (I think).

Have other people noticed this aswell?

Cheers

[tedster]

Yes - because these are very widely known factors that have been in play for years, and Google has moved far, far beyond this kind of information retrieval. To get a taste for what else may be in play and simply swamping out the new signals you were introducing, you can read some of the Google patents we've discussed here, especially:

Historical Data [webmasterworld.com]
Phrase-Based Indexing [webmasterworld.com]

The Phrase-Based Indexing patents (I believe there are now 8 of them) will give you a picture of how much things have changed since the text-match days that have gone by.

[MrFewkes]

Tedster - yes - I will read those - cheers.
I fear though that this will bring more questions even once ive understood them.
So I dont let my head run off in the wrong direction - does "Phrase Based" indexing have any connection with the old beastie known previously as "LSI"?

[tedster]

There's no direct connection between LSI and phrase-based indexing which was patented in 2006, although some of the same textual relationships certainly might be surfaced - since those relationships are inherent in the web documents themselves, and not dependent on the technology used for analysis. However, LSI is just too computationally intensive, even for Google's massive power.

But notice the word "phrase-based" and how it indicates a major step away from simple text matching. It identifies meaningful word groups across all the web content. These meaningful phrases or word groups are often called n-grams, as in 2-gram, 3-gram etc. 5-gram seems to be the current cutoff. If you have an appetite for data crunching, you might be interested in the raw n-gram data that Google made available to the public in 2006 as Google's 1 terrabyte n-gram corpus [googlesystem.blogspot.com].

One immediate take-away might be that this - the old concept of "stop words" is oversimplified and not as applicable anymore. If a phrase is identified as meaningful - such as the 3-gram "stars and stripes" - then the word "and" may no longer be thrown away as a stop word for related queries that include that phrase.

There's a whole lot to digest here, but it is technology that Google has been chewing on for over four years. If you are familiar with the concept of co-occurring semantics, note that there is a similar idea in these phrase-based indexing patents - what phrases tend to occur together in the same document, and to what degree of statistical significance. So if you hope to rank well for a given phrase, then the presence of a few related phrases on the page might help.

This would be phrases that are not merely stemmed versions of the original, but made up of completely different words. A page about "making a doctor's appointment" might well include "the nurse secretary" or "writing a prescription" - and if trait is shared across a significant number of pages, it can become a kind of relevance predictor. And if too many of the related phrases are all on the same page, then that fact may be a scraped content predictor.

[MrFewkes]

Hi Tedster,

I am with you as far as the post above goes. It may seem simple - but I looked at LSI with kind of a simple mind - like this....
"If a page wants to rank for the phrase ALBERT EINSTEIN - then it better mention relativity, energy, mass, light etc."

I have a program which crawls sites and provides me with those 4 words (for example that is) - as I dont know them without prior knowledge of einstein for example again.

Then the program plots a graph of page usage of those words for each site in the top twenty. The graph thins out as the words appear less frequently in pages down the serps - the graph showed that the higher pages almost always contained a high percentage of the words which were common to the top 20 docs in the serps.

I think I am correct in assuming therefore that google couldnt implement this on a large scale (as you say due to computational constraints) - having written the code to do it against a few serps I can understand why.

I came to the conclusion that (in theory at least) if I created enough documents wanting to rank for "Computer Screen" - and on each of those documents I placed the word "Treehouse" (unrelated word in natural documents) - then effectively I could make google only ever rank pages for computer screen which contained also the word treehouse. More to it but thats the general gist of it.

So, anyways, back to it, they have done something lesser which is phrase based - that would require less cross-referencing for certain on a massive scale - once a "fixed" DB was acheived.

Hmm - im waffling now.

"So if you hope to rank well for a given phrase, then the presence of a few related phrases on the page might help."

Tedster - I hope so - because in this instance "might help" is one thing - but as a result of this thread - I am actually relying on the implementation of phrases totally stomping on the old style stuff in order to give me a movement.....as as I say - my pages just seem "stuck".

:)
Cheers

[MrFewkes]

Ok - ive read some now.

I need to write a crawler which creates lists of ngrams from the pages returned in the serps for a given phrase.

If I do this - then in theory - I will be able to view common ngrams appearing in the top 20 (say) - note various stats (which I wont publish here) and then based on the information got from actual serps, I should be able to select certain ngrams for inclusion in my own pages.

Sound good?

[tedster]

I mentioned phrase-based indexing as just one example of how complex things have become, and not as "use this idea to automate your writing". I'd probably go more with natural expansion of the vocabulary on the page, rather than looking for an automated approach.

Just read your on-page content with this concept in mind and it can remove the shackles that came with old-school "text match" ideas about keywords and ranking. You are free to move about the entire English language, so write naturally what your visitors need to know, move them to respond with your words, and don't have much concern about finding a keyword formula.

[edited by: tedster at 6:36 pm (utc) on Jul 19, 2010]

[freejung]

MrFewkes, forgive me, but I think you may be massively overthinking this.

If you were to put the same amount of effort it would take to write such a crawler as you describe into finding a better way to get links, I bet you'd have orders of magnitude better results.

If you've already established significant relevance for a term (because your page is about that term, right), the kinds of tweaking you're talking about are going to have some very minor impact on your relevance score, when it's entirely possible that you will be unable to move up in the SERPS with any amount of improvement to your relevance, without improvement in your offsite factors.

I would take the lack of movement as an indication that you're barking up the wrong tree with on-page optimization, and focus more on other factors.

[MrFewkes]

Freejung - noone wants to share links anymore - Im not certain any automated program could persuade someone to exchange or give a link. Its just down to hard graft (and to be frank - utterly boring and soul destroying).

My competitor ranks with 1 pages sites and pr0 against my pr3. Whilst that doesnt mean much - it does tell me (or at least indicate) that ive got some good links somewhere.

Im waiting on a google visit to see if a few changes ive made to a site on the "old stuff" as it were makes any difference. If I can honestly say to myself that Ive completed my research down the onpage, onsite stuff - then Im ready to take on linking.

Im fully aware of all the BH software out there - X.u..., S.N... etc etc - id rather get some links in from my competitors related sites - like they are falling over themselves to give me links LOL.

Tedster "I'd probably go more with natural expansion of the vocabulary on the page, rather than looking for an automated approach"

I understand what you are saying - but what I am finding (sometimes) is that google will "choose" a phrase or word which I am not familiar with and almost certainly do not know that google associated it with another phrase. A program such as Ive stated above would kind of "point" me in that direction. I would then mesh the phrases into my pages manually.

This is to "be on the safe side" to ensure Im not missing out on the inclusion of a potentially heavy hitting association.

Im going to try a BH version aswell on all this - given that google have driven it mainly down to links (well it would seem that way) - then links they can have.

What a mess.
:(

[gouri]

If you've already established significant relevance for a term (because your page is about that term, right), the kinds of tweaking you're talking about are going to have some very minor impact on your relevance score, when it's entirely possible that you will be unable to move up in the SERPS with any amount of improvement to your relevance, without improvement in your offsite factors.

On a website, if making on-page changes cannot help a lot in the rankings if the relevance of a page has already been established, can adding pages to a website that are related to the theme help in the rankings?

[tedster]

Sometimes building out an entire section of good quality, on-theme content is a great way to go. You are building a better resource for the query and that's a good thing - especially if you take care to have good interlinking for the new section.

[freejung]

gouri, yeah, that's not the kind of changes we were talking about, if I understood the conversation correctly -- I thought we were talking about one-page tweaks like adding links with kw in anchor text. I'm sure tedster is right about adding new pages, plus you get the opportunity to optimize for related keywords.

As for being outranked by one-page PR0 sites, unless the one-page site has an exact keyword match in the domain name, I suspect that the PR0 is not an accurate representation of their backlink profile.

Yeah, getting links is hard and shouldn't be automated -- but then that's true of most things worth doing, IMO.

[MrFewkes]

I put another page up the other day and linked to it from the index (which is the page I want to rank). I called it "keyword1-keyword2.html" and its anchor text was "keyword1 keyword2" on the index page linking to it. Googles got it - still no change in ranking.

I will put another page up.

Yes freejung - the other sites name is keyword1keyword2anotherword.com - mine is the same aswell.

Tedster - I am going to do what you say there about building a better resource section about the product.

The site is currently an index page + my new page, plus a load of product pages made from OScommerce (not sure if you are familiar with OSC but its got all the SEO pluggins like metas and seo friendly URLs etc).

Tedster - can you expand for me just a little regarding the "good interlinking" for the new section I do.

My goal is to get the index page to rank for keyword1 keyword2. Im prepared to work hard at this - extremely hard as its a vendetta now! (not really a big money maker - more like about $200 a month profit to be honest).

If I cant get this to number 1 though what chance have I got on other phrases?

All advice appreciated - I would like to keep this thread alive until I rank. Then I will share of course.

Cheers

[tedster]

can you expand for me just a little regarding the "good interlinking" for the new section I do.

I mean that pages should clearly form a "section", where it's easy for the user to click from one to the other in a step or two.

[gouri]

MrFewkes,

Does the new page that you put up have a cache version?

Sometimes I have seen that to see the effect that a page has on your rankings, you have to wait for a cache version of it to be available.

[MrFewkes]

Tedster - I will write a subsection of 4 pages in addition to the one page I have done which is linked to from the main page. That one I have done - I will link to 4 more pages - and each of those pages will link to
a) All the other 3 pages in the section
b) The parent page (the one I have done)
c) The index page.

Does this sound like the right track?

[MrFewkes]

Gouri - yes - the cache is indeed updated and there has been no ranking change.

I am starting to firmly believe that google does not shift rankings around much at all.

[MrFewkes]

Well - an update for all. I made a plain well known flat out onpage change to a site of mine which was at position 38. After the changes were updated and cached - the site is now sitting pretty at number 4. Still a bit to go but thats a great jump.
I have today made the same change accordingly on 20 domains and noted the current position if each in the serps. This should be good - some of them are at 15 to 20 so if things work out they should easily jump to page 1 and high up. (especially considering the utter drivel that google currently ranks!)

A couple of changes phrase based and including a few "LSI" type words (but not LSI is it ;) should see that first site pip the post to number 1.

Ill let you know.

[tedster]

Thanks a lot for getting back to this thread with your good news. I've also seen some jumps from page 3-4 up to top 5 based on a nice light but strategic touch with semantic relationships (and no it's not the same as LSI).

It sounds really good.

[1script]

I made a plain well known flat out onpage change to a site of mine

Not sure if you'd be willing to discuss if you are talking about actual content changes / internal link changes or other template based changes but I may be little less at loss about what you are referring to if you can clarify whether you are talking about SERP movements of one particular page or "site average" (if that's even a valid unit of measurement)

[gouri]

MrFewkes,

On July 24, you had a post in this thread about adding a new page to your website but seeing no change in rankings.

Did you see any changes in rankings based on this change later on?

I know you mention on your August 19 post that an on-page change that you made helped you in the rankings but I was wondering if the addition of the new page helped you?