Mountain View based SearchMe has been around since 2005 and has raised $31 million from Sequoia, DAG Ventures and Lehman Brothers. But until last weekend when I met founders Randy Adams and John Holland I knew next to nothing about them. It now joins Mahalo as one of Sequoia’s big bets in search.
I say “next to nothing” because I actually did write about them before. In January 2007 the company launched a test product called WikiSeek that returned results only from Wikipedia and sites linked from Wikipedia. At the time Adams said WikiSeek was just a test product for the technology they developed. Now, over a year later, their ready to put up their main site.
SearchMe goes into private beta today - sign up on the home page for an account. The main difference between SearchMe and other search engines is that it returns results primarily in a visual format, via an image of the result site. The results are displayed in a way that is similar to browsing through albums in iTunes - see the following videos to get an idea of how it looks:
Today SearchMe has indexed just a tiny portion of the web - about 1 billion pages. But they are quickly adding to the index, and say that what they’ve grabbed so far is suitable to show off their technology. In the live demo I saw some results that were great. Other searches returned only so-so results.
Categorization and Vertical Search
SearchMe isn’t all flashy graphics. They are also auto-categorizing every page in the index to help users with disambiguation. A search for “Apple” can be done in the category “technology” to avoid results about fruit. And when you search, SearchMe places the categories it thinks are important at the top. In the screenshot above, the query “safari” shows categories including “companies,” “software,” wildlife,” “photography,” etc. The image to the right shows category suggestions for “blackberry.”
Update: Louis Gray noted SearchMe was indexing his site back in February, and said “Is it taking a graphical snapshot, in the same way that www.archive.org has done to show how Web sites looked over time? I’m not exactly sure.”
Hi,
I hail from Nigeria — can you please promptly send me the remaining $30.5M left after the R&D efforts shown above?
Even better, just ship it down the 101 - I’ll be ready and waiting. I’ll be the poor, war-torn startup with an actual idea of what it costs to build things, and what being frugal means.
Visual search engines have a *very* long history of failing. This is primarily because when people are searching for “things” those “things” are not always “products” - it’s often just generic categories of information.
Check out SpaceTime3D — another very cool app, but pretty pointless in terms of competing with a real search engine.
Even clustering approaches from data mining have not taken off. (Clutsy and gang.)
looks nice , but is it actually solving any problem in the search landscape ? Not really sure. let’s wait for cuill and see what they come up with.
Interesting search engine but long waiting for an invite…
C’mon Sequoia funded this? Unbelievable. How many times does the same idea get funded over and over.
Lets see, I will take the album art flipper idea of iTunes and combine that with the idea of crawling and indexing the web and presto I have something that has been done before. Hopefully they are using an API from Google or Yahoo and did not spend money trying to beat them on the indexing.
I remember Snap had this and killed it, I think redzee still has it in some form.. I am sure there are others still kicking
If it displays visual results, how do they filter out pron imagery? I’m no prude, but middle america is likely not ready for images of furniture s*x, if they type in the keyword ‘furniture’. And I’m sure a lot of parents would be concerned about their kids viewing lurid images.
bobbydang: from their FAQs -
“We make every effort to allow people to filter out adult material from their results. We’ve done
our best, and we’d like to think that we’re working at the same level as other major search
engines. It’s impossible to be perfect with today’s Internet, but we’ll continue to improve on this
as time goes on.”
a search engine to ‘judge a book by its cover’, so to speak?
So only if you like the site design layout, you’ll further proceed to check it out deeper? bs
that’s not what search is about, aesthetically-speaking. it’s about information, information in text…which is what Google does already so well.
Its an interesting take at searching and may appeal to some segment of users. But what is important that apart from coming up with relevant search results they need to come up with the results fast..i think thats one of the key reasons why google took off coz it was fast
hasn’t this kind of a search been done before by the others? where is a value for user here?
Mike: How about catering to your real audience and writing about true startups and entrepreneurs that are truly in the ‘arena’ battling all odds, and not some bogus search engine that has raised $31 million and has nothing to show for it. Shit if I had $31 million dollars I would build a fuckin empire. Your readers want to be inspired not discouraged by this nonsense. You should considering starting a new blog for real startups, and keep one for bullshit funded startups.
Luis Pereira
Yakov, see http://www.snap.com
Luis Pereira - really? You don’t want to hear about newly launched startups if they are well funded? seems like, you know, kind of what we do here.
@AD
I agree. Speed, coverage and freshness of information is going to be a key to the success of this. If this is really as fast as it looks in the demo (it looks pretty slick), then I think people will use it.
Just imagine how resource intensive it would be to do this on a massive scale, say on 10 million searches a day (which is not even that many compared to Google). I wouldn’t want to have to pay for that server farm.
Mike - They been around since 2005 and have raised $31 million from Sequoia, DAG Ventures and Lehman Brothers. Is that what you ‘define’ a startup?
Luis - well, yeah. they just launched, you know, today.
Mike - ok, great. thanks for the news. keep up the good work.
I think this sounds promising. This can help filter out the results very easily by actually looking at the snap shot. From small text or headlines, its very difficult to judge a useful resource. I also think this can be better than snap.com, simply because, if it can provide smooth interface.
People have just too much money to spend on sites like that.
My sense is that this search interface is really good for certain kind of searches and not for others. What seems most interesting to me is not the iPod Touch-like cover flow, but the disambiguation feature presented in icons. In their demo, its “query term labrador = the province in Canada or the dog”. Once the user has helped refine the search, then the cover flow allows you to find an interesting site.
But it seems much less interesting for more detailed research. I agree that its nice to see a preview of what is on the page, but I feel that can be done in a much more efficient manner than showing the entire page. That is what Google is doing with Universal Search, bringing in more of the page onto the Google search results. I’m working on a travel specific search engine and we do the same thing. In the flow of research, I don’t know if you want to go to a “cover flow” or “snap” experience before you have decided what page to go.
On the other hand, it’s great for browsing and discovery like Stumble Upon. Maybe these guys should look at being visual Stumble Upon instead of visual Google.
I like it. I can see the bias against site design vs. site content, but having a preview of the site is so much better. I wish them success.
Ask.com already does this with their binoculars feature although the screenshots are smaller.
But, do you WANT to see visual images of websites like this? Personally I don’t and I can’t see it taking off. The amount of funding is ridiculous IMHO. Besides isn’t it all supposed to be about localisation now? Can’t see any hint of it.
Galactic sized fail.
How come I’m supposed to filter information quickly by looking at a whole webpage instead of snippet that filters information to it’s rawest form and has a descriptive title (which is what Google does)?
Man, with $31M, you’d think they could at least hire some voice actors
I think one of the biggest problems new search engines face is building a large enough back end to actually crawl the web and have meaningful result sets.
Yea thats fine and dandy that you have a nice front end, but how many results do you have, and how old are they? To make search work, you need to crawl google/yahoo amount of data, and keep it updated. The backend to do this would have to cost millions and millions.
Most of the traffic through search engines come through long tail searches (3-4 terms) that require HUGE amounts of data to return meaningful results. This is why I think searches like Mahalo fail. They target one or two word searches. I haven’t see searchme, but I bet they don’t have near the data yahoo/google indexes.
agree with that other guy, ask.com binoculars are a ton more useful than this. At least there you get to see a whole lot more information
Wikiseek is a massive fail as well.
http://siteanalytics.compete.c.....?metric=uv
I think these 3D flipping UIs are not made for static content. It might be useful for scanning through multiple TV channels or something, but not the web stuff.
Man what a bunch of armchair haters this lot is.
If they pull this off then there is going to be a market for something like this. And the huge upside is a lot of great advertising real estate. How many people do a search in Google and then open up the various results in a new tab - you go to the tab - do a quick visual and then either read it or close the tab. This rolls it all nicely into one - quickly flip through all the various pages and then go - seems mint to me.
As for the amount of investment - hey go big or go home - if you are going to do something like this you are going to need some serious scratch to do it right.
Mike, thanks for the update and what do you want for your birthday?
I wonder how well this’ll scale?
I actually like it.
Though this is not new, as Ask (binouculars), Girafa.com and Snap.com are doing it for some time now - i hope they will manage and give a fight to Google.
Good luck guys!
I agree that there are a lot of armchair haters hanging out here at TechCrunch. I gathered my thoughts further on SearchMe and posted about it here: http://snurl.com/21i9c.
My advice to SearchMe is determine what search queries you really want to serve well and focus on them. My opinion is that its general, discovery oriented searches. I think it would be wonderful mashed up with StumbleUpon to become StumbleUpon 2.0. Good luck and the good news is that you have plenty of cash to experiment until you find a winning solution!
lol @ #10… your god complex is probably why no-one has given you $31M dollars.
This is an awesome post, a really cool startup (i hate search startups, just ask Richard M. @ rww… i’m a sour-puss about this search stuff… there is Google and there is Not-Google…) but this is really a cool (and NEW) idea… it has not been done before… (thanks for bringing us an actual contender in the Search space Mike, instead of trying to cram 100 a month down our throats like that even comes close to be useful… it does not. You do it right.
)
The videos, however… ugh… the narration was insulting. I’m not in preschool, thank you. “I’m going to type in Dee Eye Ay Em Oh En Dee Bee Ay See Kay, diamondback.”…. come on asshole, I know my abc’s.
Also, that 2 person banter on Video 1… stop that. It sounds fake, and really makes an otherwise awesome demo really really lame, and not the funny-campy-mars-attacks-lame… i mean actually just plain lame. hehehe. Just my 2 cents.
@ Matt.. your probably right
I like rooting for the underdog ‘bootstrapping’ entrepreneur who is actually trying to build something literally out of nothing against all odds. Anyone who is given $31 million dollars to build something has a huge advantage is no longer an underdog or true startup IMHO.
Nice visual idea, but not sure how useful it is.
What’s most interesting is how well the categorization will work and what are the principles behind it. If it works automatically, then they are onto something. Maybe that’s where they’ve spent those 31 million
This one has an EPIC FAIL written all over it.
PS: They COMPLETELY ripped off Apple’s Coverflow. Shameless! For $31MM, couldn’t they at least hire their own designer???
Similar to 36 I’m amazed how they can get away with this and this is considered appropriate - they completely store Coverflow’s Core UI which surely has some design patents on it. Apple might not sue but they very well could and one would think that Sequoia would want to be careful about that and that startups would be too.
IMHO I think y’all gripe a lot.
I think people with the guts to try something new in search (as #29 says) and actually sell it enough to get substantial backing are those who may actually change how painful it can be for the average user to try to find something out there in cyberspace. I’m thrilled to see some progress in this area and can’t wait to start using the thing.
Indeed, a shameless copy of Apple’s Coverflow UI!!! Shame on searchme.
This isn’t “something new in search”.
This is an R&D project that everyone directly involved already knows google will be the buyer. Sequoia has lots of insider info and pull within google.
It will take more than $31MM to compete.
Rumor is that Google already has this in the labs. They have a public beta of something visual at http://www.searchmash.com/flas.....:strollers
Snag
if you want to use a powerful visual search engine, try http://www.searchcrystal.com
It looks great with the visually oriented pages — as in searching for a stroller. But what about searching for things like: “Using nofollow with Amazon links.”
I just did a search on Google for “wedding photographer blogs.” That would seem a solid use of this service. But not combing result after result for some obscure phrase. That is painful enough on Google.
It’s starts to look more and more like dumb luck that Sequoia backed Google. They show no evidence of understanding the nature of that success. A company like searchme attempts to enter the search market using gimmickry. Search is hard, and you can only succeed by doing something really useful. This ain’t it.
How come I’m supposed to filter information quickly by looking at a whole web pages instead of snippet that filters information? i don’t think that it is good idea.
Michael,
Don’t you think this is a huge copyright lawsuit in the waiting? This is definitely not a transformative use of the work - its a replication of the authors art stored on a remote server. I doubt this is covered by fair use like thumbnails.
Nice, It looks great with the visually oriented pages.
Visual search engine, visual bookmarking … web3 is becoming visual
Have a look at http://web2wave.com which isn’t a search engine but a visual bookmarking Page. You can store, find and organize all your favorite web content via tabs, Web2Wave will theb displays website coverflows, Simply double-click on any coverflow image to make that site active.
sorry “www” is missing http://www.web2wave.com
which was literally just reproduced in under 30 minutes (yahoo search api, django)
its funny how quickly people forgot why categorizing search content didnt work in and scale in the first place. am i that old where i remember when Yahoo was just a categorized phone book?
If the founders or the investors are reading this post, I think everybody here would be interested to know what they think is the selling point of SearchMe
- Do they think they have a better search engine (I know a lot about search engines and I’d be very surprised if that is the case)
- Is it the UI which shows the page like snap.com or the categorization? (well, so many other people have done both these things, either explicitly or implicitly)