Google Allintitle inconsistent results

I believe the allintitle is a guesstimate and not an accurate number. You never know until you click through all the pages...

Yes I think 209 is the accurate number. It's a guesstimate on google's part until you get through all the results. The computer power required to do a query like that is too performance intensive so it's a guesstimate until you force the right result - 209 in this case.

Sometimes Google will output different results if you leave a space between the : and "keyword"

That might be why you are seeing different results.

Did you maybe forget the quotes on the one search? That could make a big difference.

I have been using the number on the last page of google results a lot more to judge competition.

I notice that you can get different allintitle results from google even if you hit different google servers but they are not usually that much different.

Hello.

Yes, I have encountered this many many times.

I've actually asked this question to people pitching various keyword research tools asking them how they consider this to be useful when the numbers always increase after you perform such a search? ha ha

This is my own thinking on the matter and it may be entirely wrong.

I perform a search for what appears to be a good target keyword for seo purposes and I see less than 10,000 results (often less than 1,000 results).

Then a day or two later (sometimes just an hour later) I performed the same search just to confirm the earlier stats... and I see the 10,000 has changed to 57,000 or 761 results had changed to 31,000.

I am not certain... and this is where my opinion floats into this post... but I kind of looked at it this way: when I performed those searches and discovered a very low amount of results I think this scenario possibly triggered a sort of flag in Google's search engine system that caused them to launch a sort of mini process to locate more pages (options) to return to their users.

I just view this stuff from a software engineer's perspective and I have made systems do things like this in the past. When a user mined for data that very few results were available for... I'd trigger a backend processing job to go out and do a more thorough search, locate more records that were relevant for the search and tagged those additional records as being relevant. This way it is much more efficient because the system only does this work when it sees there is a need. It responds to demand. I think it is quite possible Google has implemented a similar mechanism.

Again, I do not know for sure that is what happens. I am just saying such an approach from a software system engineering perspective makes a lot of sense. Respond to the demand (an incoming search) by rebuilding that particular search index.

I rarely use keywords in quotes or the allintitle search these days for this reason. I figure every time I run such a search I am triggering a reassessment and therefore these numbers are also kind of useless although we are all possibly helping Google to fill in the gaps in their search engine indices.

Now, I just do a combination of two things: first looking at the pages that appear in the first three pages of results for the normal (no quotes) search for the keyword... those are the true competition... and if I really want to have a clue as to how many there are (which is not nearly as useful as knowing the strength of the pages ranking in the top 30 in my opinion) I just do the search in quotes and flip through the results until I reach the very last page which always says something along the lines of In order to show you the most relevant results, we have omitted some entries very similar to the 82 already displayed.

Anyway, all of that long ramble was simply to say... yep, I have experienced the same thing.

I wrote an article about this topic: Google Lies To You! - Intel by khadaji - Qondio United States. It's a good idea to jump to the last page if the number of competing websites is important for you to more accurately determine.

There's even a search parameter you can enter to get to the last page, I recall reading about it, but I don't have it. Anyone know it?

One thing I have found is that the allintitle: attribute can be inconsistent. Sometimes it works, sometimes it doesn't.

For example, if I search for:

allintitle:"build muscle" "fast ways"

... I may get different results than if I typed:

intitle:"build muscle" intitle:"fast ways"

This is just an example of course.

I personally like to force the intitle:attribute for every phrase I want to see in the title separately.

@Terry :

Thanks for the reply but allintitle without quotes will always give a larger number

Moreover, for each of the following cases, we will have to go to the last page

allintitle:hedge fund management 63,000 results

allintitle:"hedge fund management" 21,500 results

intitle:hedge fund management 1,490,000 results



@George:

Thanks for the reply but I got the following

allintitle:"build muscle" "fast ways" 90 results

intitle:"build muscle" intitle:"fast ways" 90 results same answer in both the cases :confused::confused:

Moreover, could you please tell how does the intitle search help us?

Maybe I am missing something :rolleyes:

thanks
Ankur

REALLY! I never thought of that. My program AIS does all this stuff but it doesn't do that. I am totally going to add this in.

It's what is on the first page that matters...and nothing else !