Yahoo! Giving Up on Search? Hey, Wait Just A Gosh Darn Minute!

So every morning one of the first sites that I visit is Memeorandum. I won't try to over-hype Memeorandum here, as we're all pretty familar with it at this point, but for those of you who have been hiding in the desert under a rock for the past two years check it out. It's a great place to monitor the pulse of what's going on in the tech world each day. And what is going on today? The top story: "Yahoo! Gives Up Quest for Search Dominance." Say it ain't so Jerry Yang, or at least say it's part of a secret trick double secret agent plan to make everyone *think* this is in fact the case while you knock Google off their game with the old keep your eye on the birdy punch.

Yahoo! give up on search? You've got to be joking. From the Seattle PI: "We don't think it's reasonable to assume we're going to gain a lot of share from Google," Chief Financial Officer Susan Decker said in an interview. "It's not our goal to be No. 1 in Internet search. We would be very happy to maintain our market share." Or as the author of the article for Bloomberg Jonathan Thaw put it: "Yahoo! Inc., one of the first Internet search companies, has capitulated to Google Inc. in the battle for market dominance." Capitulated. Interesting choice of words there.

Equally pointing to Yahoo's demise is a stat from ComScore Networks pointing to their lack of success over the last year in search: "Yahoo! handled 19 percent of global Internet searches in November, a drop from 27 percent a year earlier."

Steve Rubel broke the news in the blogosphere first and wrote, "that's it, I am no longer using Yahoo Search. I have no interest in using a product that the company doesn't aspire to make best of breed."

Ok, It was a rough 2nd inning but we still have 7 innings left to play and here is how Yahoo! can and hopefully will regain it's footing in search.

1. Combine the Flickr/Yahoo! Image Search algorithms into one. AND DO THIS NOW! Actually, you can keep two algorithms but just give anything that has any decent rank at Flickr priority over anything the Yahoo! Image Search algorithm will fetch for you. I've written on this a number of times in the past. Flickr's search by most interesting produces *vastly* superior results to either Google or Yahoo! Image Search. It boggles my mind that this still has not been done and the only plausible reason in my mind why it hasn't is that Flickr has been growing so fast and focusing on their infrastructure and getting this in tune before taking this next step. Step it up. Give Flickr the resources to handle the additional traffic and turn on the switch.

The superiority of Flickr's images is just so plain to see. Try this example. Google Image Search for San Diego, Yahoo! Image Search for San Diego (not a single Flickr image on the entire first page results), heck just for fun I'll throw in Ask Jeeves Image Search for San Diego (suprisingly better than both Google and Yahoo), and now look at Flickr. Amazing how much superior. There is a single reason why this is the case. Image quality is hard to distill based on text and Flickr relies on an algorithm with mass human input and review -- get it, social search? Flickr is just the start.

2. Buy Digg, again, NOW. Obviously it's not that easy and it's easy to armchair quarterback these things when A. I don't know any of the underlying details or issues and B. It's not my money. Still, Digg can be to news search what Flickr is to image search. Although Digg has not implemented an interestingness or news rank feature similar to Flickr's interestingness it would be *very simple to do so*. And when Digg does and you go searching for a news story on TiVo and you compare what Digg fetches with what Google News fetches, it will not even be close. Yahoo will most likely need to pony up $30 million to buy Digg but they can not afford not to. Digg is only going to get more expensive as they broaden their news beyond tech and if you don't hurry up, CNET or Microsoft (I seriously doubt it), or even Google is going to beat you to the punch. Yahoo! try this for an experiment. You can track your incoming links. Start testing some Yahoo! news and see how relevancy looks when ranked by incoming Digg hits to those stories. You will find the relevancy high.

3. Integrate Del.icio.us (I hate putting all those periods in there, it’s so much easier to just say delicious), into the web search, news search, etc. algorithm. If something's got 50 bookmarks it's probably good. It probably should get a higher rank in your algorithm. 50 people like it, bump it's relevancy up. This is easy enough to tweak.

4. And here comes the most controversial part. Assemble a team of 25 editors right now to begin fixing highly rated items in the background now. What do I mean by this? I'll use a fairly easy example (and my apologies for the self promotion here). Do an image search on Flickr for San Francisco. Ok, far better than Google or Yahoo, right? Technically speaking even at the most basic levels the photos are technically much superior. Now the problem with this search is that some things come up for a San Francisco search that shouldn't because we've mistagged them. The Merlin Mann photo while very interesting from a historical web perspective, is not very interesting in a San Francisco perspective. Similarly, my highly rated photo of a snake. Damn good photo of a snake if I don't say so myself, but not of San Francisco. I just happened to take the photo in San Francisco at the Academy of Sciences. A human editor needs to go through the most searched image, news and web searches and modify these tags. You don't kill them, you leave them on, but you neutralize their rank for San Francisco (and you make sure it's tagged snake). The problem is that this will cost lots of money to manually do this. It doesn't matter. To start you only need to do this with your most popular searches and you will begin seeing an effect immediately.

Look, Yahoo! lost how much in market cap on this announcement per the Seattle PI? 13 percent? Excuse the over simplification that this 13% drop was entirely due to failing so badly in the search contest but that's how much money lost in a single day? Isn't that about 7.4 *billion* dollars? I think that they can afford to put 25 editors on the case to start and at least try this idea. The editors shouldn't be expensive but they should be bright. Recent college grads would make the best candidates and you could start them off at $50,000 per year. Some would shine, excel and move to other areas in Yahoo!, others would drop out after a year or two of it, but you'd always be hiring.

5. Buy TiVo. Someone please tell me why no one will buy TiVo. Is there too much potential liability with ad skipping? I mean Microsoft's doing it with Media Center and it seems to be firmly entrenched in the general view that it's fair use and just a VCR on steroids. Are folks just hoping for the company to fail and buy the IP in bankruptcy court? What am I missing here? This, by the way as an aside, would give their new Go TV initiative a needed boost for a lacking feature PVR/TV.

Integrate television into web and news search. You can't give the content but you can give the description. If I'm doing a search for Lyle Lovett on the internet, there's no reason why one of the first results back should not be a link letting me know that he will be on David Letterman tonight and oh, by the way, just right click here to tell my TiVo to record it for me. Even more ambitious of course will be an attempt to get all closed captioned text indexed and then use this with TiVo to track to shows and have TV transcripts show up in search. Yes, there are all kinds of copyright problems with this and you're going to have to fight Hollywood, etc. etc., but if that's not what Terry Semel is there for then why did you hire the guy. Be the first to integrate TV listings into search.

This is a vast untapped archive of information just waiting to give some needy search company and advantage and a leg up.

6. Similar to boosting rank for image search, bookmark search, news search and TV search that rank high by social networks, use your latest acquisition of Upcoming.org to do it with events. If I I'm searching for Jeff Tweedy, this should definitely come up. I'm probably a fan and would want to know about this concert and oh yeah, Yahoo! already owns Upcoming.org where this info is coming from. Put another way, if I type a search in Yahoo for "Jeff Tweedy" Chicago. Then this should *definitely* be coming up. Why does this not come up at all even when Yahoo! owns upcoming.org? Why instead do I get a link to some lame generic page with a short bio on Jeff Tweedy? Why doesn't the upcoming.org page (a highly relevant page about a concert by the band I'm searching for in the city I'm searching for who happen to be playing *tomorrow* night) even show up anywhere on the first page results? I mean, come on, Yahoo! bought upcoming.org for a reason. If I'm searching for a band in a city that is listed on upcoming.org this should be my top choice. It would be as simple as automatically increasing page rank for popular items on upcoming.org. Oh, and also, keep your eye on Eventful.

Look, Yahoo!'s got all the pieces (well all the pieces except TiVo and Digg yet). There is no way that they should capitulate to anyone -- especially Google. Search is a billion dollar game, Yahoo! just needs someone there with some power, vision and authority to tie it all together.

The only thing that I can possibly speculate on why it isn't being done today is that they lack the corporate visionary to use the chess pieces that they have in play. The rook is sitting over there in the corner and not being used. Whether because of internal politics or Yahoo! fiefdoms or whatever all of their best pieces are not being played. That hot new player that you just paid way too much money for is sitting on the bench while the coach plays the tired old veteran player who is no longer on his game.

Whatever the analogy, Yahoo! needs a strong visionary in place who can tie all of the pieces of social search together, make a few more acquisitions, and turn this ship around. They need to be empowered from the top to make changes where necessary and they can in fact get back into the search game.

Winning the search game is largely about relevancy. Google beat Yahoo! because they produced and still produce more relevant results. Once Yahoo! integrates the various social search pieces into their algorithm, they need to begin an aggressive marketing campaign to prove to people that their results are in fact better. Bloggers, the press, blind taste test type marketing stations at places like CES, etc. should all be employed. They can in fact not only hold their ground in search but take some back if done right. If they continue to lag though and under-utilize their finest tools then they will not only not regain lost ground in search, but they will continue to lose more.

Update: I just received an email back from Danny Sullivan (certainly an expert) from Search Engine Watch and he made the following good points: "I suspect you've got Decker trying to respond to investors who want to see Yahoo's share of searches oust Google -- and that's a huge challenge. If she says they aren't trying to do that,they reset their expectations and perhaps don't drive the stock down.

It's also one thing to say you don't expect to be tops in search share and another to say you don't want to be tops in search at all. I don't read directly her comments to say Yahoo's giving up on search -- just that they don't think they can bump Google for search mindshare."

I'd have to agree with him on that and as is usual for me I let the technology blind my view on the business realities of the situation. Irrespective of how Yahoo! does with search going forward, by lowering expectations, they set themselves up with analysts not to take a pounding the next time. The wind is already out of their stock sails so to speak, move guidance down and then if you come in ahead of it next time you get a bump up in the stock. I also did not read that they were not interested in search. The "capitulate" word may be a little strong for the writer to have used in my opinion.

Tags: Flickr, Digg, Delicious, Yahoo, Google, TiVo, Search