Do you need to Search all the major Search Engines?

[General]

Irina had asked me since we always hear "one must search on all search
engines, otherwise one misses a lot", if I would run the same search on some of the major search engines and see what the results may be.

Since I was very curious, here is the search and each of their results:



SAP Netweaver people that configured, designed or implemented SAP Netweaver.



Here is a side-by-side comparison of the same search:

GOOGLE:

site:www.linkedin.com "sap netweaver" (configure | design |
implement) -intitle:directory -inurl:updates -inurl:jobid -
inurl:find -inurl:answers


Results: http://bit.ly/113Dz5 =1,000+ (approx 2,300 results)

YAHOO:

site:www.linkedin.com "sap netweaver" (configure OR design OR implement) -intitle:directory

Results: http://bit.ly/E8puQ (800+ results)


LIVE:

site:www.linkedin.com "sap netweaver" (configure | design |
implement) -intitle:directory -inurl:updates -inurl:jobid -
inurl:find -inurl:answers

Results: http://bit.ly/HwDcw = (135 results)

Seems to show that you would still be fine if you only used Google!

[Irina Shamaeva]

Thanks, Gary! This is a good example. I've been curious about trying this type of comparison for a while now.

I have looked at Gary's searches closer and have collected some statistics that should be interesting for many. That said, I was surprised that yahoo collected few results and I suspect it's "misbehaving" today; so I will try yahoo again at another time.

Here's statistics on unique people names in all three engines:

Google: 618 names; a clear winner
Live: 499 names;
Yahoo: 203 names (but maybe it's having a bad day)

Overlap:
Live and Google overlap: 223 people
Yahoo and Google overlap: 107 people.

By the way Google has the most number of results but fewer than 1,000; it corrected its guess after I went though all of the results.

Here are the tools I used for this analysis:

Broadlook Diver
MS Excel.

Other thoughts and experiments on search engine comparison are welcome!

[Manick Paul]

Hi Gary,

I always spend maximum time on Google only rather then wasting time on other engines. As I believe Google contents the maximum database compare to others in this market now. Its a good example for all who believes in searching multiple search engines.

Thanks for your Research...

Regards
Manick

[gary cozin]

Irina-

Thanks for figuring out the overlap or any dupliates on the search engines. Cleary shows Google has the greatest reach for searching-

Irina Shamaeva said:

Thanks, Gary! This is a good example. I've been curious about trying this type of comparison for a while now.

I have looked at Gary's searches closer and have collected some statistics that should be interesting for many. That said, I was surprised that yahoo collected few results and I suspect it's "misbehaving" today; so I will try yahoo again at another time.

Here's statistics on unique people names in all three engines:

Google: 618 names; a clear winner
Live: 499 names;
Yahoo: 203 names (but maybe it's having a bad day)

Overlap:
Live and Google overlap: 223 people
Yahoo and Google overlap: 107 people.

By the way Google has the most number of results but fewer than 1,000; it corrected its guess after I went though all of the results.

Here are the tools I used for this analysis:

Broadlook Diver
MS Excel.

Other thoughts and experiments on search engine comparison are welcome!

[Kelly]

A few questions about this comparison - first, the search strings are different... (apples to apples?)... Also, does Live once again take the inurl:command?

With something much more simple, and therefore eliminating Google's automatic stemming, typo recovery and synonym searching,etc., I nearly ALWAYS get the most results with Yahoo. Since Google looks for so much more than what you ask for, it would make sense the quantity is greater in Google, but when that is controllled for, Google routinely is last on my list.

For example, a search of my name, (in quotes), yields 5,400 in Google, 16,000 in Yahoo and 922,000 in Live (this is suspected in Live).
A search for the word "the" yeilds 17 billion in google, but 28 billion in Yahoo...



A great tool to compare the three right next to each other (number of results and all) is Jux2. For testing results QUANTITY it's a great site. As you know, keep the search simple,and use Boolean that all three sites understand in order to compare aples to apples. A search if the word "resume" for example shows 155mil in google vs. 678mil in Yahoo.

Either way, Google, Yahoo and Live only allow you to see the first 1000 results (so who cares if there are 4.9ba-gillion?), and since the ranking of those results will be different, I highly recommend using more than one search engine. User research shows that most people only look at the first few pages anyway, if they ever get off of the first page of results to begin with.

Don't let Google's quantity of results lead you to believe that they are the "biggest" - Google usually looks for what you aksed plus lots of other things it "thinks" you want as well. Elimiate that, and Google's quantity suddently doesn't look so "big".

Put me in the "use more than one" camp.

[Irina Shamaeva]

Kelly,

You are bringing up some good points and hopefully others will respond. It's good to compare apples to apples. Let's try to come up with more examples and statistics that will educate us.

As you are correctly pointing out search engines will not find more than 1,000 results. Any number that is higher (especially if it is much higher) is just a guess and cannot be used for comparison . Therefore, if one wants to compare numbers of results one needs to create an example where the number of results is less than 1,000. Searches on "the" and your name have more results than 1,000 so the numbers of results do not tell us any reliable info.

Since you say that Yahoo brings the most results in your sourcing practice, perhaps you could prove that by giving us an example of a search? The numbers of results must be below 1,000; also "repeat search with the omitted results included" will need to be used for proper testing.

This comparison is also a good way for all of us to learn the syntax and operators of search engines. The Live search help is located here.


Kelly said:

A few questions about this comparison - first, the search strings are different... (apples to apples?)... Also, does Live once again take the inurl:command?

With something much more simple, and therefore eliminating Google's automatic stemming, typo recovery and synonym searching,etc., I nearly ALWAYS get the most results with Yahoo. Since Google looks for so much more than what you ask for, it would make sense the quantity is greater in Google, but when that is controllled for, Google routinely is last on my list.

For example, a search of my name, (in quotes), yields 5,400 in Google, 16,000 in Yahoo and 922,000 in Live (this is suspected in Live).
A search for the word "the" yeilds 17 billion in google, but 28 billion in Yahoo...



A great tool to compare the three right next to each other (number of results and all) is Jux2. For testing results QUANTITY it's a great site. As you know, keep the search simple,and use Boolean that all three sites understand in order to compare aples to apples. A search if the word "resume" for example shows 155mil in google vs. 678mil in Yahoo.

Either way, Google, Yahoo and Live only allow you to see the first 1000 results (so who cares if there are 4.9ba-gillion?), and since the ranking of those results will be different, I highly recommend using more than one search engine. User research shows that most people only look at the first few pages anyway, if they ever get off of the first page of results to begin with.

Don't let Google's quantity of results lead you to believe that they are the "biggest" - Google usually looks for what you aksed plus lots of other things it "thinks" you want as well. Elimiate that, and Google's quantity suddently doesn't look so "big".

Put me in the "use more than one" camp.

[Shannon Myers]

These is a great discussion for Quantity as Kelly mentions but I think this needs to be looked at as Quality as well. A LinkedIn is only going to give me a specific person once where Google could give me that same person 1000 different linked ways but it's still just once specific person.

[Irina Shamaeva]

I totally agree. I did my statistics, posted in this discussion, based on unique person's names.

(Of course it could have eliminated different people who have the same first-last names from the comparison but I don't think it's important if we are just comparing the engines.)

Shannon Myers said:

These is a great discussion for Quantity as Kelly mentions but I think this needs to be looked at as Quality as well. A LinkedIn is only going to give me a specific person once where Google could give me that same person 1000 different linked ways but it's still just once specific person.

[Katie McAuliffe]

Hi!
How about trying your searches on Twingine.com and then you will really see the Yahoo and Google results side by side on one page?

[Kelly]

Here's a few more search examples to prove the point that Yahoo returns more results than Google. These searches are simplistic, but as per Irina's request yeild less than 1000 results. Again, demonstrating QUANTITY not quality in these searches. (Clearly, if I were going for quality, I would make some refinements to these.)

(intitle:resume OR inurl:resume) "clinical research" (pa OR pennsylvania) (215 OR 610 OR 484 OR 267)

57 results in Google, 143 in Yahoo.

site:myspace.com (fluent OR bilingual) spanish "marketing manager" -intitle:jobs

14 results in Google, 29 In Yahoo.

(intitle:speakers OR inurl:speakers OR intitle:keynotes OR inurl:keynotes) "director of human resources"

395 in Google, and 719 in Yahoo.

If my previous extremely simplistic searches did not prove the "quantity" point that started this discussion, these searches clearly demonstrate that Yahoo, not Google seems to have the larget number of pages indexed. The results are typical of what I see day in a day out - Yahoo has about twice the number of pages as Google. Google addicts BEWARE!!! (Wink!)

To the point that was made about quality of results being more important than quantity I agree, and would build the searches to drive a happy balance of quality and quantity. For the quality arguement, I still vote for Yahoo!, as Google is now "interpreting" your queries and providing additional results (again, making them appear bigger than they really are!). For example, if you search for the word MARYLAND in Google, Google automatically looks for the abbreviation, MD, (even though you didn't ask for it in your query) which pulls false results that include MD as in "Medical Doctor", when what you really wanted was the word MARYLAND. So you can't go on pure number of results anyway, as Google's results include not just MARYLAND, but MD as well. MD as the abbreviation for Maryland probably is an ok result for your search, but MD as in "Medical doctor" is a false positive, therefore increaing QUANTITY and decreasing QUALITY. (And just for the record, even including Google's interpretations and additional search terms, a search for MARYLAND yeilds 154 million in Google, and 659 million in Yahoo!)

So for quality/purity of searches, and well as quantity, I strongly vote YaHOOOOOO!!! (Insert the Yahoo! yodel here.)

Love these discussions!





Irina Shamaeva said:

Kelly,

You are bringing up some good points and hopefully others will respond. It's good to compare apples to apples. Let's try to come up with more examples and statistics that will educate us.

As you are correctly pointing out search engines will not find more than 1,000 results. Any number that is higher (especially if it is much higher) is just a guess and cannot be used for comparison . Therefore, if one wants to compare numbers of results one needs to create an example where the number of results is less than 1,000. Searches on "the" and your name have more results than 1,000 so the numbers of results do not tell us any reliable info.

Since you say that Yahoo brings the most results in your sourcing practice, perhaps you could prove that by giving us an example of a search? The numbers of results must be below 1,000; also "repeat search with the omitted results included" will need to be used for proper testing.

This comparison is also a good way for all of us to learn the syntax and operators of search engines. The Live search help is located here.


Kelly said:

A few questions about this comparison - first, the search strings are different... (apples to apples?)... Also, does Live once again take the inurl:command?

With something much more simple, and therefore eliminating Google's automatic stemming, typo recovery and synonym searching,etc., I nearly ALWAYS get the most results with Yahoo. Since Google looks for so much more than what you ask for, it would make sense the quantity is greater in Google, but when that is controllled for, Google routinely is last on my list.

For example, a search of my name, (in quotes), yields 5,400 in Google, 16,000 in Yahoo and 922,000 in Live (this is suspected in Live).
A search for the word "the" yeilds 17 billion in google, but 28 billion in Yahoo...



A great tool to compare the three right next to each other (number of results and all) is Jux2. For testing results QUANTITY it's a great site. As you know, keep the search simple,and use Boolean that all three sites understand in order to compare aples to apples. A search if the word "resume" for example shows 155mil in google vs. 678mil in Yahoo.

Either way, Google, Yahoo and Live only allow you to see the first 1000 results (so who cares if there are 4.9ba-gillion?), and since the ranking of those results will be different, I highly recommend using more than one search engine. User research shows that most people only look at the first few pages anyway, if they ever get off of the first page of results to begin with.

Don't let Google's quantity of results lead you to believe that they are the "biggest" - Google usually looks for what you aksed plus lots of other things it "thinks" you want as well. Elimiate that, and Google's quantity suddently doesn't look so "big".

Put me in the "use more than one" camp.

[Irina Shamaeva]

Hi Kelly,

I tried one of your examples just now,
intitle:speakers OR inurl:speakers OR intitle:keynotes OR inurl:keynotes) "director of human resources"
and in fact Google finds 343 results, Yahoo finds 306 results. To make a true comparison, you need to "repeat the searches with the omitted results included" which also gives the engines a chance to review and finalize the list of results.

I agree that these discussions are fun!

Kelly said:

Here's a few more search examples to prove the point that Yahoo returns more results than Google. ...

[Rithesh Nair - Sourcing guy]

1)Searches will be different with 100+datacenters google has ...:)
2)thumbshots.org is another way to find the comparison bet'n sites searches
3)multiple engines all the way..:))
regards
Rithesh

Irina Shamaeva said:

Hi Kelly,

I tried one of your examples just now,
intitle:speakers OR inurl:speakers OR intitle:keynotes OR inurl:keynotes) "director of human resources"
and in fact Google finds 343 results, Yahoo finds 306 results. To make a true comparison, you need to "repeat the searches with the omitted results included" which also gives the engines a chance to review and finalize the list of results.

I agree that these discussions are fun!

Kelly said:

Here's a few more search examples to prove the point that Yahoo returns more results than Google. ...