17 ways to beat Google - Search innovations and who is doing them

Nitin Karandikar sent me a link to his post on "Top 17 Search Innovations outside of Google" which reviews 17 different approaches to web search innovation. He also lists some of the companies leading the way in each area. Web search is a very popular topic for writers and bloggers. Charles Knight maintains a list of the Top 100 Alternative Search Engines and updates it monthly.

Web search is big business. Microsoft's Live Search is third in market share with about 10% of all web searches and it generates hundreds of millions of dollars in revenue. It is no wonder that VCs and entrepreneurs are investing lots of time and money to discover The Next Big Thing in search.

Of all the approaches described by Nitin and listed by Charles Knight, I think three are promising; Natural Language Processing, Local Search, and Cell Phone Voice Search.

Natural Language Processing (NLP) has been around for a while. The idea is to get users to ask questions in conversational style, and then parse that question using NLP techniques to divine meaning and context. The problem is that most users only enter two words in a search...sometimes three words. No matter how sophisticated your NLP technology is you aren't going to gain a lot more context from two or three words. The real secret power of NLP is realized by applying it to the search index...not the search term. Meaning, analyze all the web pages in the index with the NLP and add lots of meta tags, meaning, and context to the index. Then when search queries come in you can match the keywords to the much richer index. Powerset is doing some ground breaking work in this area.

Local Search is a huge market. The easy part of local search has already been done. The big search engines have developed crawlers to crawl all web pages, parse and extract local identifying information, and cross reference it to other online data. The problem is that over 50% of small local businesses don't have web sites. A great local search experience requires that ALL local businesses are included and that the index be updated regularly. CitySquares Online is a Boston based startup attacking this problem.

Cell phone voice based search is an obvious opportunity, but very hard to do. It requires great voice recognition technology, a well organized local search index, a high quality voice response system, and special screen rendering technology to display the results on the cell phone screen. Parts of the technology exist today, but no one has combined all the pieces into a great user experience. This is a huge opportunity with no clear leader.

OK, now back to Nitin's list of 17 search innovation areas. Please go to his site to get more detail on each one. Unless you are a search geek like me you may find this stuff boring, but here are the categories;

1. Natural Language Processing
2. Personalization
3. Vertical Search
4. Multimedia Search
5. Restricted Data Sources
6. Domain Specific Search
7. Parametric Search
8. Social network filtering
9. Human directed search
10. Semantic search
11. Discovery linked search
12. Classification, Tags, Clustering approaches
13. Results visualization
14. Results refinement and filters
15. Specialized search platforms
16. Related searches
17. Search agents

What do you think? Who will be the next Google? Which approach to search will provide the most value?  Place your bets and you could be the next search billionaire.

Subscribe - To get an automatic feed of all future posts subscribe here, or to receive them via email go here and enter your email address in the box in the right column.

Comments

[zdavatz]

Well speaking about Search and Innovation I believe you should not forget InfoCodex:

The thing I do not like about Google and Yahoo is, that they do not recognize documents with similar content. It happens often on the Web that a post or document is spread out over more then 50 websites. Now that is great for the author but not for the searcher because it blows up your search result unnecessarily. With InfoCodex this will not happen because the linguistical database recognizes similar documents and puts them into groups. This does not blow up your search result unnecessarily.

http://www.ywesee.com/pmwiki.php/Ywesee/InfoCodexProcedure

Three things Google should do:

1. Automatically classify a document according to its content.
2. Automatically generate an abstract of a document.
3. Generate a Heat-Map of the Contents of a Search Result.

http://www.ywesee.com/uploads/Main/InfoCodex_22.2.2007.pdf

[Mukund Mohan]

Don
The "next Microsoft" was Google. The "next Google" will not be in search is my personal opinion, just like google was not the next desktop personal computing paradigm.

Beating Google in doing "better search" is interesting but the next frontier has to be about a technology and business model change than paid text ads.

[Yihong Ding]

Don,

I would partially buy Mukund's vision. From one side, Mukund pointed out one important thing: "next Google" would not be built on the current model, similar to that this Google is not built on the older model. On the other hand, it is not obviously why the next Google will not be in the realm of web search. In fact, web search is a spirit of WWW. Really novel innovations on web search may definitely lead to the rise of another Google. The problem is, however, which type of innovations on web search can really be addressed as novel?

Most recently, I am doing a research about the evolution of World Wide Web. The most recent post about this topic can be found at here: http://www.deg.byu.edu/ding/WebEvolution/evolution-dream.html

Part of this article I discussed the future of web search. Though Google is looked like unbeatable at this moment, it certainly has its unavoidable problem that may leads to its sunset. But I doubt anybody could beat Google in web search if they try to follow Google's way to search the web. Even if we have listed all these 17 different ways, they still can hardly beat Google. With its money and human resources, Google can easily beat any competitor in any of these 17 ways in the global scale. Indeed, I doubt any company, including Microsoft, might beat Google in web search if these companies still stick on the so-called "oracle-based" web search model. Google has created the greatest oracle ever and barely anyone could build an even greater one.

To beat Google, we must overthrow the current oracle-based web search model. From AltaVista to Yahoo to Google, we have witnessed the rise of greater and greater oracles on the web. We humans are dreaming of building the "God of WWW" who knows everything on the web and can answer any questions about the web. The success of Google has pushed this God-making progress to a height that no one else has ever reached. The problem is, can this God-making progress be continued further? Is it possible to build a real "semantic God (Google)" that knows everything on the web not only by keywords but also by meaning? Honestly, I don't believe so. If our ancestors failed at building the Tower of Babel, we either could not be successed in building a new Tower of Babel.

Google will continue expanding its power by building this Tower of Babel. This ambitious path will not succeed to the end. For the ones who really want to beat Google, they should not follow the same way. In contrast, they should look at the history of mankind after the Tower of Babel. This history provides the hint for the future web search.

I have presented a detailed discussion about "collaborative web search model" in my article. The spirit of this search model is to allow individuals sharing their search results instead of relying on instructions from an oracle. This model is fundamentally different from the current oracle-based web search model. From one side, it is a simulation of search within the real human soceity when WWW had not been invented. On the other side, it is also motivated by two very new technologies---semantic desktop and semantic search.

Despites the detail of this search model, which could be found in the article, I would like to point out that Microsoft is a great company that may lead this transition. That I would suggest this to Microsoft is not due to its great power, but because of its domination on the current desktop operating system. Based on my vision, the future of web search will be back to the basis of desktops. Is it ironic? Yes, it is. If Microsoft stick on the current MSN search, even if it also merge with the Yahoo search, this combination would still not beat Google. But if Microsoft may take the advantage of its desktop computation, and lead the migration of web search model on the basis of the vision of web evolution, Google is going to decline.

Would Microsoft be like to discuss more on this new vision? Or would anybody else be like to discuss more about this new search model with us? I would like to have email discussions with you.

Yihong