Google's Marissa Mayer on search evolution

Google executive on search iterations, such as Twitter, Cuil, human-powered search

Entrepreneur interview by Bambi Francisco Roizen
November 26, 2008 | Comments (3)
Short URL: http://vator.tv/n/57a

In the future, the way we search today - typing into a keyboard, retrieving a list of results - will seem pretty primitive. That's why new search engines keep sprouting up each day. Even though Google continues to squash them all, and grab market share, there is still room for significant innovation. Search is at its infancy, said Marissa Mayer, VP of search products and user experience at Google, in the first of a five-part interview with me at the Vator studios. In this segment, we talk about how search is evolving, and how far we've come. 

Mayer's role at Google has included developing Google's search interface, and launching more than 100 features and products on Google.com. To this end, Marissa's view on how search is going to evolve is pretty valuable, if we want to know what's ahead for Google.  

Mayer looks forward to the day that we can start thinking of search as "answers" rather than a list of things. She also thinks search will become far more personalized in order to become more relevant. (Watch for a future interview in which Marissa talks about how Google is become more social and personalized).

Having started at Google in 1999, Mayer certainly recalls the days when Google was counting the pages they indexed. I asked Mayer how search has improved and how the algorithms have changed in the past decade.

Mayer said that what started off as a simple yet incredibly useful Page rank (the algorithm named after Google co-founder Larry Page) equation has now grown to be "hundreds of signals that go into the ranking function." Additionally, Google has the benefit of a decade to understand user behavior when it comes to how they search and the obstacles they face when searching.

We also talked about search engine optimization, and how this mansion industry is affecting Google's pursuit to create relevant organic search listings. Marissa says people should optimize their pages in order to be indexed appropriately. But there are nefarious attempts to manipulate a page, like "cloaking," that Google is challenged to detect. Cloaking is an SEO technique in which content presented to the search engine is different than what is presented in the browser. The purpose of cloaking is to trick search engines so they display the page, when it would otherwise not.

What about InLinks? I asked, referring to a service that allows you to buy keywords in the text of content pages, and replace those static keywords with hyperlinks to your Web page, resulting in higher page ranking for your Web site. Mayer said that as long as the URL is relevant to the keyword, it's probably OK. “If the link itself accurately reflects the content it’s linking to. It's in the spirit of the Web, and spirt of page rank and our overall relevance algorithm," she said.

We also discussed semantic search, Powerset and what, if any, were Google's plans to incorporate natural language.

"We’re not focused on semantic search," she said. "[But] when you get to extremely large amounts of data you see a precision that mirrors semantic search."

We then moved into the discussion of Twitter, and whether Google had any plans to incorporate a search engine for real-time status information.

Twitter may be great at providing "clues" to news or events, she said. Knowledge of this accumulating sentiment could be used to drive crawlers to be more aggressive in gathering relevant results based on such clues, she said. "Twitter does have some interesting implications," she added.

We also touched on the search experience for Google's Android, and her view of human-powered search engines, such as Aardvark and Mahalo. As for Cuil, which was started by ex-Googlers and whose interface looks like the anti-Google, Mayer said it needs more indexing to better understand the value of its algorithm. 

The final question was what Marissa thought search would be like in five years. You'll have to watch to hear her answer. 

(Note: Marissa was our guest host on Vator Box evaluating SearchMe, WooMe, Viewdle, and Appssavvy. Her other interviews with me, along with our Vator Box segments will be released throughout the coming weeks.)

(This piece was originally posted on the 25th. It was re-published to highlight on the VatorNews homepage)

Related companies, investors and entrepreneurs

Description: Viewdle is about capture in the moment.  We’re focused on enabling consumers to manage, tag, and share their media a...
Description: What is Twitter? Twitter is an online information network that allows anyone with an account to post 140 character messages, called tweet...
Description: Searchme lets you see what you’re searching for. As you start typing, categories appear that relate to your query. Choose a category, a...
Description: WooMe is an online speed dating platorm that leverages cutting edge in-browser voip and video technology to enable users to meet new peop...
Description: appssavvy is a mobile advertising technology company with the mission of rethinking the delivery and reception of advertising. Focused on...
Marissa Mayer
Bio: Marissa leads the company's product management efforts on search products – web search, images, news, books, products, maps, Google...


S A, on November 25, 2008

"In the future, the way we search today - typing into a keyboard, retrieving a list of results - will seem pretty primitive" - check out www.cloudtuner.com

Danny McGowan
Danny McGowan, on November 25, 2008

"Natural Language Location" is the future of internet discovery. Watch my video.

Jean Ferré, on December 5, 2008

Without employing semantic search technology it is difficult for enterprise Search vendors to provide a comprehensive and user friendly service for their customers, so I find it surprising that Google are not interested in it, unless they only aim at searching the web.

For search to be successful in the corporate environment it must be intelligent and intuitive, as content saved on company networks is not designed to be found. Documents are saved in silos, saved in various formats and are rarely where you expect them to be. This makes finding information extremely difficult. Moreover, search must be deterministic, which means there are no corpus to learn from to better understand a given document!

By understanding the meaning of words, by using all the available knowledge provided by language models and rules, semantic and linguistic based search can provide users with more relevant results, therefore providing a better user experience. But semantic search alone is not enough. Extracting metadata (entities) such as names of people makes a huge difference. Enterprise search vendors must provide a fully comprehensive solution in order for users to receive the results they need, including documents or people. This can make it possible for staff to connect with their colleagues and make better decisions faster, which is what businesses demand from enterprise search. In addition, security and connectivity are mission critical factors to take into account.

Jean Ferré, CEO, Sinequa.

blog comments powered by Disqus