|Convera Survey : Only One in Ten Search Engine Users are Satisfied|
|Twenty-one percent of professionals feel search engines understand their queries, and only one in 10 find exactly what they want on the first attempt. |
On the second try, 90 percent will try the same search engine using a slightly different term, while 60 percent will try another popular search engine. More than half will turn to a topic-specific search engine.
| A commanding 95 percent of professionals find Internet search engines |
an aid for conducting work, yet only 40 percent say they are very
satisfied with the results.
Only 21 percent feel that their search query is always understood.
Less than 25 percent of professionals are very confident that when
using popular Internet search engines they've looked everywhere to
|On the second try, 90 percent will try the same search engine using a slightly different term, while 60 percent will try another popular search engine. |
Pardon my confusion, but isn't that 150 percent?
|Pardon my confusion, but isn't that 150 percent? |
treat both of the questions as separate ones , as a part of questionaire submitted they might have answered and percentange has been calculated
It never ceases to amaze me that the big search engines do not invest in educating users in how to use a search engine.
Maybe surveys like this will point them in the right direction.
Educated users = lower PPC revenues. Conspiracy? Probably not. Economic reality? Bingo.
I do have to give Google kudos (which is something I don't often do...) for Google Scholar. A friend of mine who is a Materials Scientist can't praise them too much for it.
At the same time, the regular Google search (which he still needs for other things, such as product information) frustrates him. His biggest complaint is the irrelevant MFA sites that come up on nearly every search.
He's not a computer geek, so adding a -inurl: with the common offenders to every search is something that would never occur to him. (Though I've now showed him the how to do this.)
|Educated users = lower PPC revenues. |
But according to the statistic posted PPC isn't cutting it either:-
|Less than 25 percent of professionals are very confident that when using popular Internet search engines they've looked everywhere to find answers. |
Whether you're targetting adverts or organic results the technology of providing topical results is the same.
|Whether you're targetting adverts or organic results the technology of providing topical results is the same. |
Not exactly. There are similarities, but advert targeting involves the human element, both in the fact that advertisers are selecting keywords to advertise on, as well as the involvement of market forces in selecting which ads will be seen.
That said, the search engines currently give advertisers terrible tools for targeting their ads. It is simply crude, without even the ability to use any kind of shorthand such as regular expressions to make it easy to focus precisely and narrowly. Advertisers have to exhaustively list phrases that might be matched, even though it would be easy for the search companies to provide much more powerful tools. Advertisers are also "locked out" of less popular search terms by high prices apparently meant to discourage the use of less popular terms, I am assuming because of the cost of adding bins to count hits on these terms.
Nothing like putting out a slightly biased bit of press, eh?
I tried their results on their demo site govmine.com and was completely dissatisfied for the topics I tend to search most often, so they obviously have some experience with dissatisfied search engine users.
It's also amusing with the timing that they are in the news now as their crawler just recently started trying to pound on my site in mid-November and took another attempt today.
|184.108.40.206 [7-9745.san2.attens.net.] requested 680 pages as "ConveraCrawler/0.9d (+http://www.authoritativeweb.com/crawl)" |
[edited by: incrediBILL at 9:51 pm (utc) on Dec. 28, 2006]
Some search engines have done a better job than others. For example, google displays tips, and has a decent help section on their site. Not to mention all the cash invested in UI.
Plus, if you go to a major bookstore, you'll find enough books out there on how to use google
In my experience most un-educated Internet users will pick PPC ads for a commercial search query as it is almost always a better match.
Whilst that is good for me it shows how far we have come when commercial queries are better served by ads than natural results - which are by the way pure crap for any worthy commercial query.
But that is just from my perspective - I do not care if professionals find search results any good at all, and this study is of no value to anyone that does not sell to professionals?
Academic search is close to an ideal win-win situation. In this case, ads don't compete with search, but complement it, and there is a clear seperation. (Search results are academic papers. Ads, if any, are products somehow related to the research.) There isn't the problem of spammy sites, MFAs, etc.
I find it funny (pecular, not haha, as my Grandmother used to say...) what a good job Google does of Academic search, where the inherent conflict of interest with sposored results is practically nonexistent, when they (IMO) continue to do such a terrible job of regular search.
From the press release: Convera, a provider of search technologies,
Perhaps I am getting too cynical about scrapers when they try to claim they are doing it for the betterment of mankind etc.
Convera == scraper?
|Convera == scraper? |
That's basically it. Google allows public access to the collected data and is therefore a more public service than Convera. Convera is in the business of scraping sites without allowing a reciprocal searching of its index unless you pay them (from a brief look at its site). One is an essential part of the web ecology as it helps it grow - the other is a parasite. Harsh interpretation but it is accurate I think.
These questions were carefully set up so as to yield the answers required by the press release.
The only surprise to me is that they supposedly got more than half of respondents to admit that when they fail to find what they're looking for it's because "they do not have the skills to find it". Even if it was true (which it isn't), you couldn't drag an admission like that out of me with wild horses.
...BTW, I agree with their concept that subject-specific or industry-specific search engines can add enormous value. But in the government site mentioned above (which is clearly NOT a scraper despite what has been alleged), they don't do a great job of leveraging the industry-specific angle IMO.
The difference is in the symbiotic relationship that Google has with the web and the parasitical relationship that ops like Convera have with the web. From what I can see, Convera is in the business of taking web data and repackaging it for profit without public access. This is different to Google's action of taking web data and making it searchable for web users, essentially for free.
|I agree with their concept that subject-specific or industry-specific search engines can add enormous value. But in the government site mentioned above (which is clearly NOT a scraper despite what has been alleged), they don't do a great job of leveraging the industry-specific angle IMO. |
Industry and niche specific search engines have their place but for the operators of large websites, these scrapers and maggots are a daily problem and do not send sufficient traffic to justify allowing them access.
Swanson is I believe on something here:
|Whilst that is good for me it shows how far we have come when commercial queries are better served by ads than natural results - which are by the way pure crap for any worthy commercial query. |
If this is not well thought by any SE so what is it?
Good results out of the blue :) for once I may literally use LOL!
Now about education: I disagree; users do not need to be educated
Why do we always think Aspirin while disregarding the cause
And here the cause lies possibly in SE interfaces and their range of search offerings.
These facts really do not suprise me much at all. Most people are lazy and only type in one or two words the first time around and are not that specific in their search.
I remember a few years ago when it became public knowledge that a woman had been in a long-term relationship with a well known married man. The mistress stated she went along with it only because she thought he might hurt her career if she stopped.
She worked part-time as a cheerleading coach at a small middle school.
Since then, when someone speaks of having a "career", I always wonder where the dividing line is between a career and a job? If I go to work making fries at McDonald's, is it OK if I call that a career?
Articles like this bring up the same basic question. What criteria establishes these people as "professionals?"
|It never ceases to amaze me that the big search engines do not invest in educating users in how to use a search engine. |
|Most people are lazy and only type in one or two words the first time around and are not that specific in their search. |
I second these, and have held this view for almost a decade now, since the days of NorthernLight and Excite and InfoSeek and Lycos, and what not.
This also reminds me of a section that the PC Magazine used to publish occasionally, in late '80s and early '90s, if memory serves me right. They used to interview the Tech Support people from major hardware/software companies to pump them for stories of dumb and funny sorts of calls.
They once reported that a large percentage of software support people used to get daily calls, more or less of this nature:
Customer: I can't get this ^&*$% to work.
Tech Support: What does it say on your monitor?
Customer: It says 'Strike any key to continue'
Tech Support: Please do what it says
Customer: I can't find the darned $*&^# 'any' key on the keyboard
Can the general awareness level of average search engine users be compared to this? For example, I wonder what percentage of users would know that they can use simple +, - and " " operators to get (at least somewhat) more precise results?
[edited by: Web_Savvy at 5:47 pm (utc) on Dec. 29, 2006]
|The difference is in the symbiotic relationship that Google has with the web and the parasitical relationship that ops like Convera have with the web. From what I can see, Convera is in the business of taking web data and repackaging it for profit without public access. This is different to Google's action of taking web data and making it searchable for web users, essentially for free. |
You haven't looked too carefully of Convera's description of what they do, then.
From the company's website, I see they have a wide range of search products, primarily intended for INTERNAL search. When they talk about "professional communities", I think they are actually talking about indexing a specific site with the cooperation of that site.
GovMine, though, is an Internet search and is free.
I tried it, and it is pretty spiffy. It isn't actually specific to government sites, but I guess government-related topics. I did a fairly vague search, for "San Diego zoning". I am encouraged that there seems to be some understanding of topic. Part of the returned results are identified topics, in this case, "San diego => "city",zoning => "land use", and "san diego => San Diego County" => Topic. You can narrow by "commercial area", "industrial area", "open space", etc. or broaden as well. In "advanced setup" I really like that you can select categories to include or exclude in your search.
Now, can somebody show me just where Convera is "scraping" and not making the result available to the public for free? First of all, "scraping" to me means taking CONTENT and presenting it as original results without a link back to the source. Does Convera do this?
Sounds like scraping to me...
|Imagine being able to deliver a perfect slice of the Internet to your professional community, providing continually deep, compelling content complete with industry specific terminology. |
If that didn't say it all, try this...
|...TrueKnowledge for Web is designed for: |
* Information aggregators looking to extend their offerings with more authoritative Web content, providing their customers with a single source for professional research.
Where do you get this information that you aggregate in the first place?
You scrape it, that's where...
Or perhaps I have a somewhat different interpretation of what they do. :)
|You haven't looked too carefully of Convera's description of what they do, then. |
Well they never bothered contacting me about crawling my main website. And my site happens to be an Irish specific one. Perhaps it comes under the "Relevant, meaningful information wherever it resides" aspiration on the Convera website.
|From the company's website, I see they have a wide range of search products, primarily intended for INTERNAL search. When they talk about "professional communities", I think they are actually talking about indexing a specific site with the cooperation of that site. |
It goes back to the bandwidth:business equation. GovMine is irrelevant for my market and I have to pay for my bandwidth. Therefore like many other webmasters, I have to decide whether allowing some repackager like Convera to spider my site is worth it in the number of users that it would send this way.
|GovMine, though, is an Internet search and is free. |
To me it means taking the data and repackaging it for sale. Perhaps it is a wider definition. But in any case, it is my bandwidth that these guys are using and that's why they've been banned.
|First of all, "scraping" to me means taking CONTENT and presenting it as original results without a link back to the source. Does Convera do this? |
|To me it means taking the data and repackaging it for sale |
If that's your definition, then I'd ban Google as well as they repackage your data for sale as an advertising medium.
I define scraping as a one-sided relationship that results in my data being taken with absolutely no tangible benefits for myself. Google, Yahoo and MSN are symbiotic in nature and return tangible benefits, aka free traffic, therefore I let them crawl.
I don't foresee a product aimed at professional research providing sufficient traffic to justify crawling my site.
Somewhere down the road all of these crawlers are going to have to form some sort of cache proxy cooperative where our pages will be crawled one time and then shared by a multitude of crawlers from a central location without each crawler individually hammering on our servers.
Of course by then pigs will have wings and it will be ski season in Hell.
It has been a long day. I should have remembered what I wrote earlier in the thread. :)
|If that's your definition, then I'd ban Google as well as they repackage your data for sale as an advertising medium. |
Exactly - the bandwith:business equation. Convera and the other scrapers and maggot ops don't get this crucial point.
|I define scraping as a one-sided relationship that results in my data being taken with absolutely no tangible benefits for myself. Google, Yahoo and MSN are symbiotic in nature and return tangible benefits, aka free traffic, therefore I let them crawl. |
Again, Convera is NOT a "scraper" by any conventional definition (never mind a maggot).
Scrapers are parasitic. Taking incrediBILL's definition a bit further, I'd say that as soon as you're legitimately trying to add value to the Web, to make it easier to navigate, you're not a scraper.
It's of no consequence whether you're a for-profit company or not. It's irrelevant whether you have API that allow others to build applications with the data you've collected. It doesn't even matter if some website owners end up paying out more in bandwidth costs then they reap in benefits. Those factors simply have nothing to do with whether you're a scraper.
|I'd say that as soon as you're legitimately trying to add value to the Web, to make it easier to navigate, you're not a scraper. |
I don't draw any distinction between the scrapers that build an MFA site with the content or some company that charges professionals to perform research, as neither are beneficial to me, therefore I treat them all the same and lump them in the same group: BANNED.
However, Convera doed identify themselves as a bot, and claim to follow robots.txt, so we can call them a crawler or spider if that makes you feel better.
Still no benefit for me to allow them to crawl.
Back to the topic, so what if SE users are satisfied with the results as the ads are typically more relevant to what the SE users want which is why Google is making a ton of money from advertisers. Convera is just taking a different approach and charging professionals for more targeted results, or so they claim as my sample searches on Govmine left me cold, but the net result is the same in that someone gets paid as the result of the search.
[edited by: incrediBILL at 11:40 pm (utc) on Dec. 29, 2006]
This isn't an engine problem, it is a user problem. I search no less than 100 times in any given week. That is a slow week. On average I search more like a hundred times a day. Suffice it to say, I search ALOT as part of my business.
I find that most times the answer I need is in the top 3, almost always the top 1, and my question is answered without even clicking the listing link, whether I search in the course of my business, or search for personal reasons.
The times I have to search twice, or go onto page two of search results is less than 1%... virtulally nil. I take is as a personal insult if I can't draw the answer I want in the #1 listing, and a pain in my neck. I work quickly, and poor results take up valuable time. So, I have a great technique for getting my answer in #1.
Yet, almost everyone I know, can never find what they are looking for. I can sit down at their machine, and produce an instant answer, and it always amazed them.
It is a user problem, not an engine problem. People have no clue how to do proper research. When I look through my visitor logs for the some 30 sites I own, I am amazed at the stupidity of searchers. I can figure out what they were likely looking for, but because they had a search term that was incorrectly phrased, they get one of my sites, instead of what they were looking for.
It doesn't say much for the intelligence of the average internet user. But, then I have done millions of searches in 14 years of internet use. I have to say, the last three years have been the best for doing searches. Not only I have developed a great technique for getting answers, search technology has come a long way.
What engine do I use? Google every time. I wouldn't waste my time on any other.
[edited by: MsHuggys at 11:17 am (utc) on Dec. 31, 2006]