Forum Moderators: open
As part of my investigations I checked all athe various allin searches and found it in the top 10 for all. My PR is higher than many of the sites above me so what is the missing factor.
Mine isn't a new site, it's got more genuine local interest links than many above me (which are nationally run db driven sites) so I don't really get it.
Is it possible that you can do well in all the seperate allin searches but together your site is regarded as over SEO'd?
Or is it possible the allin searches just don't work and I should be looking elsewhere.
Also I still have good rankings in Yahoo and MSN if this makes any difference (over SEO'd?)
I'm sure there have threads about this in the past - I just didn't pay much attention to them. I guess I need to do some searches for old threads about it.
Try all the following searches
allintext:christmas
allintitle:christmas
allinanchor:christmas
any other allin search that you want for the word "christmas"
The results are all pulled from exactly the same ranking list with the sites removed that do not include the word "christmas" in the specified area.
If you rank well on this search list, and you have "christmas" in only one case of anchor text, you would still rank high on allinanchor.
allinanchor does not rank according to how much anchor text you have fore that word, it never has, google has never claimed that it has, and anyone that tells you otherwise is wrong.
So what's the diagnosis of site that comes near the top in nearly all the allin filters, isn't new, has good PR but flounders in the SERPS!?
Since you are doing well in the raning that is used by the allin commands for that term, which PR appears to play a part in, I would guess that you are doing something that Google specifically does not like at the moment to reduce your position in the final, general SERPs.
OK, then allinanchor measures exactly what?
Until we know exactly what an allin search rank on, it measures exactly nothing.
It simply tells you whether a page meets the criteria to pass the filter. The ranking in that list is is determined by the ranking criteria which we do not know.
I would guess that you are doing something that Google specifically does not like at the moment to reduce your position in the final, general SERPs.
That makes sense. But they must really hate it to put all those completely irrelevant sites above mine!
i.e. allinanchor:christmas means that the site that is listed #1 has the most anchor text containing "christmas"?
Obviously that is not true. It measures something that we don't know. Just because we don't know it doesn't make it not exist.
If you rank well for allin commands but poorly in the regular search, you have a problem. For most new sites this is an easy way to see the sandbox they are in.
Isn't allinanchor a measure of your inbound links (IBL) using the keyword?i.e. allinanchor:christmas means that the site that is listed #1 has the most anchor text containing "christmas"?
Nope. As explained above, and if you run the tests I suggested and look at the results you will see that. Change the search term you use and it will aways work the same way.
It is a filter, not a method of ranking.
Obviously that is not true. It measures something that we don't know. Just because we don't know it doesn't make it not exist
It is obviously true from the perspective of anyone outside google. it is the same as saying "he measures 17" The measurement is meaningless without the units and perspective.
So for all of us, it measures "exactly" nothing. But it does tell us some interesting things about how we rank in what appears to be a somewhat simplified algorithm
If you rank well for allin commands but poorly in the regular search, you have a problem. For most new sites this is an easy way to see the sandbox they are in.
I agree on the first part. As for the second, I have never experienced the sandbox, and almost never do allin searches, so I wouldn't know.
Nope. As explained above, and if you run the tests I suggested and look at the results you will see that. Change the search term you use and it will aways work the same way.
Riddle me this BigDave:
I do a regular Google search for KW1. SITE1 comes up 1st and SITE2 comes up 85th.
Now, I do an ALLINANCHOR:KW1, SITE2 comes up 1st, and SITE1 comes up 14th.
How can these results come from the same ranking list?
If you rank well for allin commands but poorly in the regular search, you have a problem.
I guess I have a problem then. Mines in the top 10 for all 3 allins, but on page 11 for a regular search. Although searching with a different term for the same page, I'm on page 3. I don't know if that means I can conclude that any problems I might have are not due to something in the page itself but rather related to just specific keyword phrase seaches on that page or not.
How can these results come from the same ranking list?
I would suspect that your issue is that you have a reading comprehension problem.
What I wrote was:
Try all the following searches
allintext:christmas
allintitle:christmas
allinanchor:christmas
any other allin search that you want for the word "christmas"
Where in that list of searches do I suggest that you compare any of them to a regular search? I suggested that you compare the list of results for all the different allin commands with each other.
Allin command coorelate to each other, not necessarily to the regular results, and that is why threads like this show up on such a regular basis.
Complete nonsense. Just because someone speaks Swahili in your face doesn't mean it means nothing. It just means you don't understand something that could mean a lot.
"[M]easure" is a different word than "understand". Without the units, it is not, by definition of the word "measure", a "measure".
If there is a context where you have the rest of the information, the combination of that information measures it. One portion of that information does not. We only have a portion of that information.
"How far is it to Seattle?"
"Miles" is not a measure.
"52" is not a measure.
"52 miles" is.
I guess I have a problem then. Mines in the top 10 for all 3 allins, but on page 11 for a regular search. Although searching with a different term for the same page, I'm on page 3. I don't know if that means I can conclude that any problems I might have are not due to something in the page itself but rather related to just specific keyword phrase seaches on that page or not.
I believe this shows a common miscalculation of the importance of semantics when ranking in the serps.
You have assumed that your kw is sufficiently powerful to control the serp ranking. It might be in some rare cases, but in most cases a set of words/phrases make up the semantic determining the rank reflected in the serps.
So your allin checks are on specific kw's that are part of the relevancy algo, but not all. Your actual search reflects a ranking based on all of the algo. That they do not agree is not surprising. If you knew the algo, you could run allin checks on all of the relevant kws, put it all together according to that algo, add linking/PR factors, and match the SERPs exactly. Of course if you could do that, you wouldn't be chatting here at WebmasterWorld :-)
Of course if you could do that, you wouldn't be chatting here at WebmasterWorld :-)
I might still hang out for fun - but if I figured out the algo I would be retiring shortly afterwards!
The allin commands measure something significant, not "nothing". They offer significant information that should not just be armwaved away because a person doesn't know the exactitude of the information.
Just because you don't know how far it is to Seattle doesn't mean the measurement, many measurements in fact, don't exist.
I never said that the measurement does not exist. It certainly does.
But it is not a measurement if it does not contain all the information necessary for it to be a measurement. It is a measurement ONLY when it is combined with the required information to make it a measurement. "52" is a number. "52 miles" is a measurement. We are only seeing the number.
Let me remind you of exactly what I said:
Until we know exactly what an allin search rank on, it measures exactly nothing.
Without that precious piece of information "exactly what an allin search rank[s] on", it is not a measurement. Just like 52 is not the distance to Seattle without knowing that it is "miles", "minutes", "kilometers" or some other unit.
When that information is added, it becomes a measurement. Look up the definition in a dictionary. It involves a quantitative relationship to a standard. If you do not know what the standard is, it is not a measurement.
On its own, without the standard, it is not measure. It is a subset of the required pieces of a measure.
It does not matter that *someone* knows the standard that completes it. Until that standard is combined with the quantitative value, it is not a measure.
How many more ways do I have to try and explain it to you? Or are you just going to stick with it stubbornly and I should give up?
The allin commands measure something significant, not "nothing". They offer significant information that should not just be armwaved away because a person doesn't know the exactitude of the information.
And where did I say that it did not provide any information?
Just because it "measures exactly nothing" doesn't mean that there is no use in it.
I also said, in the same post that you got so riled up about:
Since you are doing well in the raning that is used by the allin commands for that term, which PR appears to play a part in, I would guess that you are doing something that Google specifically does not like at the moment to reduce your position in the final, general SERPs.
which certainly acknowledges that there is some value to it. It looks like I said that if you rank well in that list, but not in the SERPs, that there is some other factor keeping you down. But gosh, you didn't seem to notice that part. You were too riled up about that other sentence.
The reality is that there is little value in what they return. It is useful for certain minor considerations and research, but there are plenty of other ways to get similar information.
If you know of somthing that those allin commands really tell you about the ranking of your site, please let us in on it. You've asserted this before, but you never have revealed the secret. Are you still insisting that allintitle actually ranks you on how well google ranks your particular title for a search phrase?
bigdave makes an interesting conceptualization of the allin searches when he calls them filters. It sounds reasonable to me to call the allin searches filters, just like the other searches where you can filter by document type or domain. But steveb makes some good points, too.
Some people assert there is a diagnostic value to the allin command and I can meet them half way on that. I myself don't rely on the allins as an SEO tool.
That's just my opinion and not in any way a fact. It's still debatable.
does anybody have an actual use for the allin searches?
Research.
BigDave is essentially right. These tools IMO, just give us public access to one aspect of a multitude of algos in force.
(My sneakin impression is that they're more geared towards Google employees researchin/detectin spam tactics plus other areas of internal interest and the likes)
paybacksa makes a good point RE: semantics too.
bigdave makes an interesting conceptualization of the allin searches when he calls them filters.
That is basically what google calls them. They do not use the word "filter", but that is exactly how they describe them.
And it is far beyond a conseptualization. It has now gone through repeatable rigorous testing. And I have demonstrated how others can repeat the testing.
I am waiting for someone to run the tests and demonstrate that it is wrong.
If you include [inurl:] in your query, Google will restrict the results to documents containing that word in the url.
I really think that they were intended for those rare cases where the *users*, not webmasters or google employees, are looking for a site where they remember something specific about it. At least inurl has been useful to me in the past when I remembered something about a site, and I remembered something about an url, but no matter how much I tried, I could not remember the domain name.
And my girlfriend is a college research librarian. She knows more tricks to weeding out information on the web using advanced operators and different search engines that just about anyone here. Remeber, google has a definite leaning towards appealing to the scholarly information crowd.
There aren't a lot of users that would ever use the advanced operators, but there are some, and they are ones that search engines build their reputations on.
I do think the allin commands have a use for SEO, but it is much more of a market research tool than anything else. You get to find out what your competitors are concentrating on, and what pieces of the puzzle they are missing out on.
It is just like the link: command. There are legitimate end user uses for it, but it tends to be used and abused by webmasters a lot more. Though I actually find link to be a far more useful tool than any of the allin commands. In fact, I find the current version to be a lot more useful than the older version that everyone seems to bemoan the loss of.
The allin commands measure something and it makes no difference at all if Big dave understands it. That is self-absorbtion to an infinite degree. Your insistence that just because you don't know the standard makes it non-existent is truly bizarre.
Just because we don't know by what standard the woman across the bar measures a man, that doesn't mean that she is measuring exactly nothing. And in fact, it is pretty darn dumb to think that.
Which to get back to reality and the point of the thread, yes you can rank well for allin commands and like crap for regular search. Then also, the criteria for the allin commands sometimes changes. That just means they have their own algorithm. Ranking good for allin and bad for regular search is a sign of a problem, although what the problem is varies from page to page.
I might still hang out for fun - but if I figured out the algo I would be retiring shortly afterwards!
Yeah, for a day or two until Google "adjusts" it ;-)
The most rewarding research I do on Google is keyword set research. Oddly, even though Google was the linking se, I pay more atention to backlinks and anchor text on Yahoo! than on Google.
