Slashdot Mirror
Bing More Effective Than Google?
Xiph1980 writes "Experian Hitwise claims Bing and Bing-powered search to be more effective than Google. The success rate for Bing searches in the U.S. in July was 80.04%, compared to 67.56% for Google. The market watcher defines 'success rate' as the percentage of search queries that result in a visit to a website. Searches made through sites owned by Yahoo, which farmed out search to Bing under a deal struck in 2009, were also more efficient than Google. Those searches yielded a success rate of 81.36%. The claims of Hitwise don't explain why I keep finding things like Microsoft service pack download pages better through google than through bing."
The claims of Hitwise don't explain why I keep finding things like Microsoft service pack download pages better through google than through bing.
That's because unlike Google, Bing doesn't favor its own services over others. Google favors their news service, maps, YouTube, shopping and every other service over others. Bing returns results objectively.
There are also differences in algorithms. Bing doesn't count so called junk-links while Google does. Bing prefers link inside good, relevant content. Google, on the other hand, counts all kinds of links. That's also why Google is full of shitty results, as SEO spammers game the system by spamming links to blog comments and every other place where they can get it. As Bing doesn't count those links almost at all it means their results are much more cleaner.
The problem Bing is facing is that they cannot get as much user data from searchers as Google. They miss a lot of long-tail keyword data that Google gets just because of their dominant market share. They also miss a lot of data of what result user thinks is relevant and good for the search query (both Google and Bing track which result user clicks on) and how much they spend on the site (both services again track if you return back from that result - if you come back quickly, it's obviously worthless result for the query). This is also the same reason why Bing toolbar gathers that data on users who use Google - the same thing that somehow got twisted in slashdotters heads as Bing scraping and stealing results from Google. The only thing they do is collect that click data.
Judging by the usual slashdot response of "but they should just improve their algorithms", people don't seem to get how immersively complex current search engines and their algorithms are. It's not just about following links on other websites - we have been past that for almost 10 years now. Algorithms are the base of the search engine, but they're almost worthless without all the keyword and usage data that really powers them. That is also why Google is so keen to collect every single piece of information they can get their hands on.
Microsoft has done a lot of things correctly with Bing. I would say their algorithms are even better than Google's, as they're able to compete with much smaller market share and data against Google and actually provide better results. It has come a long way from the Windows Live Search days.
Google+ vs. Facebook, and why Google+ will fail
Just become somebody clicks through to the site doesn't mean the search result was a success.
Google is my preferred search engine, but the results are noticeably geek slanted. That's perfect for me, I am a geek. However it is not what everyone wants. Bing I find does a better job giving what a non-technical user might be after. You have to remember that as a tech person, what you are interested in may not mesh with what non-tech people are interested in.
So for me, Google it is, but that may not be true for everyone.
Take calculator and define for example. I don't need to click anything after searching, because google tells me the answer directly. I would say that's more effective than making me click through for currency conversions and dictionary definitions.
The market watcher defines "success rate" as the percentage of search queries that result in a visit to a website.
These finding may be interesting and can be interpreted in many ways, but it's completely arbitrary to associate "success rate" with the percentage of queries that resulted in a visit to a website.
Just one example for an exception: maybe the "blurb" offered by Google gives you more information, sometimes even to the point of giving you the answer you were looking for. Search for "first apollo launch" on both Google and Bing. I'm getting more dates in the blurbs on Google than Bing. Now search for "barack obama age" -- Google actually answers the question: "Best guess for Barack Obama age is 50 years (August 4, 1961)".
There are plenty of other reasons for why queries don't lead to websites. This has practically nothing to do with "success rate".
Entomologically speaking, the spider is not a bug, it's a feature.
I use Bing for recipes and things that normal people search for. I use Google for anything technical since bing appears to be clueless about that stuff.
Google works if you already know how to use a search engine. My Mom doesn't know how to ask google good questions, so she needs to use Bing to find when the special church service is in her town.
I reckon this is because SEO's and link farm scum are throwing all their weight at gaming Google rather than Bing.
I still haven't forgiven Microsoft for pounding, and I mean pounding, a self-hosted (long story) site for a small retailer I worked for a few years ago. We got a nearly $1000 bill for excess bandwidth. I checked the logs and they were downloading entire directories of images over and over and over. Non-techy Boss NOT impressed.
If I'm looking up the name of something (via related criteria), or searching for a particular statistic, my ideal is to find it displayed in one of the website titles or excerpts without ever having to click anything.
Google also displays dictionary entries, etc. so that I can generally lookup words and get the definition right in the results.
Many times I consider a result "successful" when I don't find what I'm looking for--it was evident from the results that the object or information I wanted did not exist, so, while disappointing, Google did the job I wanted it to do.
I think a far better test is whether, after searching for something, small keyword alterations are made. Granted, many times there is a level of human refinement where people start off not knowing quite what they're looking for, but I think there is probably a much better correlation of people trying different words because they didn't find what they wanted than not-clicking anything. Basically, if people are coming away from Google and Bing equally satisfied, and Bing users click more, that means Bing is less effective and making its users do more work to get their info.
When things get complex, multiply by the complex conjugate.
So, marketing apparently CAN divide by zero?
What about searches that were a mistake and corrected.
I admit that sometimes I use google as a spell checker and never click through to a page. I'm sure other people do this.
I have consistently found google more effective. My suspicion is that this is because I am usually looking for information as opposed to products.
Maybe it's just that those who choose to use Bing as search engine are more likely to click links? or they are easier to please with the returned results set? Correlation does not imply causation. It will be difficult to make conclusions without having one study group using both.
Google cache
Agreed - I generally find my answers on the first page of a Google search.
But, I can't get past the definition for "success" in the summary. There are times when I Google something, and the answer appears in the summary - no need to click any links.
If you're measuring "success" in terms of dollars and cents changing hands somewhere, yeah, Bing is probably a success. If you're measuring "success" in terms of searchers finding the data they are looking for, I'll put my money on Google.
"Windows is like the faint smell of piss in a subway: it's there, and there's nothing you can do about it." - Charlie Br
Who funded the research?
It does often tend to skew the results in the favour of the person who commissioned the report.
I've noticed lately that google isn't nearly as sharp at finding the results I want. If I search for terms 'x', 'y', and 'z', google will sometimes give me a page with terms 'x' and 'y' but not 'z'. 'z' is on pages that link to the results, but google doesn't tell me this. If there are no pages with 'x', 'y', and 'z' on them then so be it, but don't give me pages that I don't want.
rant over.
The market watcher defines 'success rate' as the percentage of search queries that result in a visit to a website.
What matters is the much harder to measure percentage of search queries that result in a visit to a website that actually contains what you want.
Let's ask two popular search engines the same simple question:
"Who's the black private dick who's a sex machine to all the chicks?"
Seriously. Try it on Bing, then try it on Google.
Game over.
---------------------------------------
Rotate the pod, please, HAL....
So when I google for "google" with instant search, will it count as a search for:
-g
-go
-goo
-goog
-googl
-google
That's 6 searches, and I may click on none, realizing I'm already at the page that I was looking for.
But, but., why don't you have your Windows systems set to auto-brick^h^h^h^h^hupdate?
Seriously, I use Google to make the Microsoft VS help usable. VS help is reasonably useful for specific syntax for a supported method/function. It is utterly useless, in my experience, to decide which method/function to use in the first place, whereas Google usually has an answer located within the first 20 links.
IMO, there are serious deficiencies in Google (word1-word2, as a hyphenated string, for example), but I think, once I get the hang of custom searches associated with my gmail account, it will be usable for a wider range of queries.
Showing results for spelling. Search instead for speling.
"The market watcher defines 'success rate' as the percentage of search queries that result in a visit to a website."
From that, I would assume Bing just makes the results a lot less clear.
While Google may control the lion's share of the search market, queries made through Microsoft's Bing search engine lead users to click on a Web page at a significantly higher rate than queries made through Google, according to data released Thursday.
The success rate for Bing searches in the U.S. in July was 80.04%, compared to 67.56% for Google, according to Experian Hitwise. The market watcher defines "success rate" as the percentage of search queries that result in a visit to a website.
Okay, cowboy. I'll give you the explaination. It's "instant search".
The definition is completely flawed, probably it's intentionally flawed. In order to give Bing better results.
Sent as ripples into the electromagnetic field. No single photon has been harmed in the process.
In fact if you use a hosts file and adblock+ bing.com doesn't even load properly. If you use the bing search bar in FF or Chrome it works but formatting is all screwed up.
Lately it is *too* smart. When I'm searching for a specific term that happens to be a bit uncommon, I keep having to do my query like this:
"relatively_uncommon_word" -"common_word_with_similar_spelling"
Because it keeps guessing incorrectly that I actually want the common word instead of the one that I entered that is spelled similarly. I'm fine with the "Did you mean ... whatever" suggestions, but when Google uses those suggestions automatically in searches it gets really annoying. It means I keep entering my search, cursing at the wrong answer, and re-entering the search with the "-" on the common words before clicking on any links. Usually it only takes one failed search like that, but there are some occasions where I have had to eliminate two or even three words before I get what I want. I don't know what Bing does in this regard, but I'm wondering if a stupider search engine would be better in some ways compared to the way Google currently is.
If you let Bing be your search of choice you probably don't discern.
That's a baseless statement.
I have tried Bing on many occasions because I'm tired of Google's brokenness and new "features" it keeps rolling out*. Unfortunately, Bing still frequently returns things that I'm not interested in. Conversely, I rarely end up with a Google search that doesn't send me to what I want to find.
*I am completely fed up with Google's hijacking of my search terms -- Google used to predict what you wanted to search for and suggest it to you. Now it just takes you to where it thinks you want to go, and you're lucky if it'll spit out a "did you mean?" More troublesome is that frequently, where it thinks I want to go is completely ridiculous and nonsensical. Here's a real scenario: I searched for "united states weather radar". Google returned "Showing results for "unted states weather ra". Search instead for "united states weather radar". Who searches for "weather ra"?? This happens several times a day to me.
I get better results through Google. I am just one user of a very limited though active demographic group. Bing might deliver more of what the average joe wants, but I wouldn't know that because I am more of an average "cecil" than an average joe.
Who searches for "weather ra"??
Mortal,
That search may be made by someone confused about Egyptian mythology. The answer is: I handle the Sun and Creation. Weather is handled by my son, Set.
You're welcome,
Ra
Heliopolis
"Showing results for "unted states weather ra"
That's just their AJAX unable to keep up with your typing. It registers the \n before it finishes registering the last few characters of your query.
It happens a lot to me. I don't use the search form on the page anymore. Instead, I use the search box in Firefox.
Google's taken a wrong turn in its user interface. It's one thing to include results from other relevant features in the search results itself, but a completely other thing to lay out all of their services all over their page (top and side) even when they're completely irrelevant to your current search. It needs to go back to the clean, uncluttered look and feel it used to have. Otherwise, people will (and I imagine already have) start looking for alternatives.
"If a nation expects to be ignorant and free in a state of civilization, it expects what never was and never will be."
Mod parent up. Notice the fifth link on Bing: "Can Openoffice.org rival Microsoft with Libre Office?". I am no Microsoft hater, actually quite the contrary, but that link is not there on the first page of Google, and rightfully so.