Slashdot Mirror
Google's Share of Searches Falling? Or Increasing?
prostoalex writes "Get two research companies in the room, and you'll likely end up with three opinions. Bear Sterns quotes ComScore Networks data, which says that Google's share of searches is slipping, down to 36.9% in June 2005. WebSideStory, a Web research company, on the other hand, claims that in June 2005 Google hit a new record as far as share of searches, hitting 52%, and leaving rivals far behind."
Statistics only benefit those who pay for them.
...while I google for the answer...
This actually happens far more than most people realize. Close to 37% of comparative statistics are in opposition to each other, while nigh on 79.3% of all statistics continue to be made up. There's also a a roughly 100% chance of this getting modded redundant if I don't hit "submit" in the next 9.3 seconds. We'll all know whether I made or not shortly...
If brevity is the soul of wit, then how does one explain Twitter?
IT's far more important _which_ searches they're getting. In my online advertising efforts, the quality of traffic coming from Google is FAR superior to what we're getting out of MSN, AOL, etc.
Well if the amount of searches google is getting is going down -- you have to account for where they are going.
Are the lost searches going to elsewhere to altavista, ask jeeves, dogpile?
How can you conduct research if you can't account for where the "lost searches" have gone to? How can they tell it's not an error in their study?
How could anyone get conclusive results without operating inside of google, and inside of yahoo? It seems to me like the *best* you can do is have a bunch of websites log what URL's they're coming from, but that's inconclusive; some users will use one search engine repeatedly just to find an article, while other's will search for the same thing across a lot of search engines.
I just don't see how anyone could come to a result that's completely objective.
"Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
Have Google searches replaced DNS for web name lookup? Not yet, you say? Then they aren't the most popular search engine yet!
Do you like Japanese imports?
Well, that's kind of silly. If one is legitimate and another backed by, say, Google or MSN, clearly they don't cancel each other out.
Studies can't really cancel each other out. If, in the presence of both studies, neither of them count, then they were probably both wrong to begin with.
And it is possible for both to be right, if they use different methods for sampling or measurement.
I find it interesting that MSN has 15.7%, which is entirely due to new installations of Windows setting IE's homepage to msn.com. Without that I'd bank their percentage would hover somewhere around the minus. IMO Google has no competitors in terms simplicity and results.
"Simplify, simplify, simplify!" Thoreau
The numbers -- for those interested.
Google's market share of US searches for June 2005 was at 36.9% compared with 37.5% in May 2005. Yahoo! had a 30.4% share, and MSN had a share of 15.7%. Bear Stearns noted that Google's query volume rose 36% YTY versus a 28% increase for the industry, outpacing Yahoo!'s 32% increase but trailing MSN's 42% increase. Month-to-month, Google's query volume declined 6%, which compares with a 4% decline for Yahoo!, a 1% decline for MSN, a 4% decline for AOL, and a 7% decline for Ask Jeeves. In Q2 2005 unique searchers versus Q2 2004 increased 31% for Google, 21% for Yahoo, and 14% for the industry, while the number of searches in Q2 2005 increased 38% for Google, 42% for Yahoo, and 31% for the industry.
Slashdotter observes that ComScore Networks gets a lot of its data from a piece of software called "Marketscore", which sure sounds like a form of spyware.
Slashdotter hypothesizes that the people who prefer Google (over MSN, Yahoo, AOL, and the various "search engines" that are installed by spyware companies) are less likely to tolerate the presence of crap like "Marketscore" on their boxen.
Slashdotter suggests that analyst from Bear Stearns ought to look closely at the source of his data and ask pointed questions as to whether or not there are variables that cannot be measured by ComScore Networks, and whether or not these variables are skewing the data he's paying for.
There are lies, damn lies and statistics. Personally I can't remember the last time I used something other than Google for a search and what other search engine has had it's name verbed in common usage? Even back in the day when AltaVista was the search engine of choice I never heard anyone say "AltaVista it".
"Oh, people can come up with statistics to prove anything, Kent. 14% of people know that."
In the end, both numbers are probably correct according to the methodology that was used. The disparity is likely due to using vastly different methodologies.
Perhaps they used user surveys of on-line behavior. Perhaps they used traffic reports from popular sites to see where search hits came from. There's countless ways to figure it, but seeing as I can't remember the last time I used anything but google to do a search, I'd tend to favor the larger number.
This sig has been temporarily disconnected or is no longer in service
Please, for the good of Humanity, vote Obama.
Google's new features are cool, no doubt, but perhaps it is losing popularity due to the ever-increasingly complex interface.
Sure it is.
Google on Dec 02, 1998
Google on Jul 21, 2005
-Valiss
Google conceptually figured out how to let humans do what they are good and, and let computers do what they are good at and use that as to generate search results.
Yahoo at the time, had hordes of people who browsed the web literally and categorized things by hand. A completely unscalable, and insanely expensive proposition.
Alta Vista was the king of meta-tag spamming. Sure it was all automated, but it sucked to use them for searching.
Now, you also fundamentally don't understand what makes google an innovator. First off, look at Ajax and how interactive and user friendly their websites are. Take their maps for instance. They are the first (as far as I know the only people) with whom I can scroll the map using a mouse. I don't have to use preset jumps. I don't remember ever seeing anyone who let me jump to intersections and show them to me. I don't know of anyone else who provides satallite and drawn maps both as seamlessly or for free. They make online applications that work nearly as well, and are almost as flexible as desktop application.
Plus it's innovative that they turned the entire business model on it's head. They are quality. Everything they do is based on quality. (That's a lot like Apple). They refuse to compromise quality to make an easy buck. It'd be easy for them to let people purchase "bonus pagerank". They haven't, and they won't. They ensure that everything that is paid for is clearly marked that way. Unlike any other search before it.
They don't have to make any money from the people who use their software. They make all their money from people who are interested in presenting information related to the information google is presenting.
Next, at the time of it's IPO, it was speculated that Google's SA scalability and parallel programming technology was worth more then the company as an advertising company ever was. The problem with that analysis is that Google's knowledge is buried in people's head, if they left the company it that value would flee the building with it.
Finally, how many companies do you know of who encourage (require?) you to spend 20% of your time being creative to work on personal ideas that could become conceptual products.
Apple isn't an innovator in terms of the iPod or digitial music. I've been listening to MP3's on a computer since 1995, and several people I know have had portable MP3 players since late 2000, or early 2001 at least. Apple was very late to the game. They just happen to own it now. What Apple did was change the business model to one that people are happier about. Remarkably like Google has with search.
As to Oracle, I'm not sure about their history. I'm fairly confident the concept of an RDBM's was widely known prior to Oracle's existance. (Conceptually I believe they have been around since the late 1970's, I though Ellison started on it in 1982 or so). Ellison surely wasn't the innovator, I believe that honor would go to Micheal Stonebreaker (Standard researcher who was the world's leading expert on RDMS for all of thei early stages).
Kirby
When I first read the report on the 19th (Google - 36.9%, Yahoo! - 30.4%, MSN - 15.7%), I called BS. Here's my reasoning. Just before Star Wars was released, I held the number one search spot for "Star Wars Trailer" (no quotes) on ALL three mentioned search engines for a week. I got a million hits in about 2 days. 69% were Google, 14% from Yahoo and 2% from MSN.
I later moved down to 4th in Google but held my place in Y!/MSN. The scores shifted very slightly but remained in favor of Google. Unless this proves that Star Wars geeks are primarily Google users, it proves the surveys innaccurate in favor of MSN and Yahoo!.
That's my 0.02.
Movie News - "Entertainment news, bitch!"