Search Engine History - How Google Came to Dominate

So how did Google get from a market share of lessIt was at this point that Sergey Brin was drawn into
than 1% in 1999 to almost 80% today? What is sothe project. At this time. Page estimated the size of
special about Google and how has it come tothe web to be approximately 10 million documents
dominate the scene? In this article, the third in myand the number of links to be perhaps 100 million.
search history series, I look at the role of luck, timingAltavista had impressed everyone by managing to
and the mistakes of competitors in the rise ofindex all those documents on a single large computer.
Google. I also evaluate the unique technology andHowever, indexing both the documents and the links
vision at the heart of the Google algorithm.was, to say the least, a big hairy problem; precisely
The Dot-com Shake-outthe sort of challenge that appealed to Brin.
You might be surprised to hear me moan about theIn fact, the problem was much tougher than either
dot-com mania of 1999-2000. After all, for many ofprobably realized at the time. At this point, the web
us it was an exciting time. However, for the earlywas growing at an exponential rate and the
search engines, it was essentially a death-knell! Thecomputing resources required for the task were in
new money entering the market was huge butfact well beyond the humble resources of a student
investors wanted to see returns. Search Enginesproject. This brings me to the part played by lady
were clearly part of the future, but where was theirluck in the formation of Google.
revenue stream? How could one make money fromIn late 1996, the BackRub crawler was consuming
free results - especially free results powered by verynearly half of all Stanford's network bandwidth and,
expensive computer infrastructures? This impatienceon occasions bringing down Stanford's internet
grew following the dot-com crash of March 2000.connection. Page and Brin had to beg and borrow
Yahoo! seemed to be the only company with anystorage capacity and were continually running out of
kind of sustainable "pure internet play" (from bannerspace and lines of credit. Whilst both were hardly
advertising which was at that time well funded bymucking about, they were not completely sure about
the dot-com dollar). At the same time, AOL werea career in business. As Page puts it now, he was
pursuing a moderately successful "walled garden"either going to be a professor or run a company that
approach to traffic (i.e. where 70% of searches onmade a difference. Nothing in between would do.
the site led to content within AOL properties). TheSo, believe it or not, in the period from 1996-7, all
argument at the time was that making it easy forthree of the then market leaders were approached
people to "exit" to other sites (the essential purposeby two students with an idea for licensing a new kind
of the modern search engine) meant losing theof search engine. In August 2004, at the Search
opportunity to sell them anything from your own siteMemories session at SES San Jose, the panel included
and content partners.Doug Cutting (formerly Senior Architect, Excite),
Altavista, acquired with DEC by Compaq in 1999, wasSteve Kirsch (formerly founder of Infoseek) and
relaunched at huge cost as a portal competitor toLouis Monier, (formerly CTO at AltaVista). Together
Yahoo! Similarly, Excite (now owned by broadbandthey reminisced about the early years of search
provider @home) tried to become an AOL clone.engines and all admitted, one by one, that they had
Lycos, already the largest "pure portal" in the Springsent the founders of Google packing. Steve Kirsh
of 1999, merged with the largest Spanish languagewas the most colorful; "Go pound sand I told them".
ISP, Terra Networks (and then ran itself into theDoug Cutting said "I wasn't impressed with their
ground).demo at all" and Louis Mornier shrugged "I didn't have
Disney/ABC and NBC had already invested heavily inthe authority to sign a check anyway".
Infoseek and Snap respectively but quickly began toWhat if Larry Page and Sergey Brin had been just a
realize that the revenue they could derive from thelittle bit better at PowerPoint? What if Louis had had
increasingly portal-like services was never going tohis check book with him? This book might have been
deliver the short term payback which shareholderscalled "How to get to the top of Altavista" and Digital
typically seek. Steve Bornstein, chairman of Waltmight still have been in business, continuing the make
Disney Internet Group, said in his press release "thethe machines that power the internet. Lady luck
Internet environment has continued to shift andcertainly played her part in the proceedings!
change, and therefore our strategies must alsoHowever, despite their efforts to dodge destiny, the
change". 400 employees at Go Network were layedStanford graduates did not find a suitable established
off and a buyer sought for the Infoseek engine.partner. Some are born great, others achieve
In December of 2000 a surprising report from foundgreatness and some have greatness thrust upon
that only 6.86% of all the referring traffic tothem. Perhaps Larry and Sergey fall into that rare
websites originated from search engines (comparedlatter category. By the end of 1998, the Backrub
to 47.01% from direct navigation or bookmarks andservice (renamed Google in September 1997) was
46.13% from internet links). The quality of thisgrowing fast and serving more than 10,000 daily
research has since been questioned, as it was basedqueries. The guys were introduced through a shared
not on server logs but on the global Hitwise statcontact to Andy Bechtolsheim, an active early-stage
counter data (the code for which was only placed byinvestor. After a short demo, Andy literally thrust a
many webmasters on the homepage of their site).check for $100,000 into their hands. Google inc. was
However, this data came at a time when there wasborn and Brin and Page famously celebrated with a
a collective loss of confidence in search and portaltrip to Burger King.
technologies. The dot-com crash of March 2000 hadThe guys literally lacked a bank account or a legal
left investors with a major hangover.entity called Google to deposit the check into. So,
Is it any surprise, then, that so many searchonce the burgers write digested, the two set up
properties disappeared altogether (or closed theirGoogle Inc and rented a garage from a friend. Five
internal crawl capability down) in 2001? With themonths later, the initial investment roll totaled just
closure of Northern Light (to the public at least) inunder $1 million and, by June 1999, a further round
January 2002, the Dot-Com shakeout had essentially(of venture capital funding) exceeded $25 million. If it
closed or fatally wounded WebCrawler, Magellan,were fiction, you would say it was a fairy tale... but
Excite, Infoseek, Go Network, Snap and Lycos. Itsometimes truth is stranger than fiction (particularly in
also critically wounded Altavista, who were tothe strange world of technology).
struggle on without any real investment in theirFrom backroom to boardroom in just 5 years
search capability until their acquisition by Yahoo! inSo what does Google mean anyway? Brin and Page
2003.close the name, from a brainstorm with fellow
The Emergence of Paid Searchstudents, as a play on the word Googol, meaning the
If the rush to convert all search engines to portalsnumber one followed by 100 zeros. Google's use of
was (at the time at least) flawed strategic thinking,the term reflects the company's mission; "to organize
bad timing also played a part in the demise of sothe immense, seemingly infinite amount of information
many early contenders.available on the web". As an interesting footnote, the
Quietly and almost unnoticed during the mania was aword Googol itself is traceable to the work of
relatively minor search player called go-to.com,American mathematician Edward Kasner in the 1940s.
renamed Overture in 2001. Whilst paid listings hadHis descendants briefly considered suing Google in
been tried before, Overture (with a 2.76% market2004 for using the name without royalty. Only in
share in January 1999) were the first company wasAmerica!
make a success of it, with some advertisers payingUnlike their competitors, Google had a clear vision,
up to a dollar a click by 1998/9. Go-to.com, in fact,perfect timing and all the luck going. Whilst the big
only carried paid results! Personally, I never really likedboys fiddled with their content-rich portals, Google
or used the site myself for that exact reason.ruthlessly focused on building the biggest index in the
However, they persevered with their strategy,world, ranking sites organically by their authority and
determined to find a sustainable revenue model andserving paid listings alongside, in emulation of
confident that the research-oriented search enginesOverture's successful model. As the web grew into a
of the time had little to offer a growing army ofgiant haystack, Google helped you find the needle.
web users looking to shop 'til they dropped!These factors, above all, laid the foundation for
History has proved that Overture were right (toGoogle's later dominance.
persevere with paid ads) although initially there wasThe company was also pretty thrifty and managed
limited evidence of this. Perhaps if the dot-comto make it's initial venture funding go a very long way
investors had only waited a little longer (or started a(at a time when other dotcom millionaires were
little later) the value of sponsored results would havespending money like it was going out of fashion). In
become clear. Either way, the big boys hared off intothose early days, licensing the search results to other
portal oblivion, whilst the Overture tortoise (and theinternet sites and portals was the key source of
model it spawned) ultimately won the race.revenue. Key moments included the use of Google
All the best businesses start in Garagesby Yahoo! in 2000 to power their search results and
It really is true. The 12x18 foot garage, in whichthe later decision by AOL to do the same. Suddenly,
David Packard and William Hewlett launchedthere was only one serious player in town!
Hewlett-Packard in 1939, is considered by many toFollowing the important business wins of 2000 (and
be the birthplace of Silicon Valley. A short drive awaythe launch in that same year of Google Adwords and
is the Los Altos garage of Steve Jobs' parents,the Google Toolbar) the search engine was handling
where Apple Computer Inc. was born. Google Inc,more than 100 million searches a year. By the fourth
similarly traced its early days to a garage - this one inquarter of 2001, the company was turning a profit -
Menlo Park, California - where the newly formedan incredible achievement when one considers that
Google Inc spent its first five months as a properAmazon (founded five years before Google in 1994)
business in late 1998.only turned its first quarterly profit at the same time.
However, the real story of Google began more thanBetween July 2002 and June 2003, global referrals
two years earlier, as the research project (nicknamedfrom search engines almost doubled (from 7.18% to
"BackRub") of two Stanford Ph.D. students:13.46%) and have continued to increase quickly since.
Michigan-born Larry Page and Russian Sergey Brin.Most small business sites now obtain more than 85%
The simple obsession which preoccupied both wasof their new visitors from search engines, with
the links between sites on the exponentially growingGoogle typically sending about 70% of them. In the
web. In his original thesis research, Page hypothesisedUK, search engines have overtaken adult sites (in
that any link from one site to another was akin to anOctober 2006) as the most heavily trafficked
academic citation and the description (or anchor text)category on the web.
attached to that link like an annotation. If one couldIt was perhaps inevitable that Google would
identify the sites most cited for any given annotatedeventually come to market and it's initial public
subject, one could establish a pecking order ofoffering took place in August 2004, creating a
significance for any given subject area.company worth $23 billion. More importantly, the
What particularly fascinated Page was what he sawflotation raised some $1.67 billion, which Google could
as a fundamental weakness in Tim Berners-Lee'suse to develop new products and make acquisitions.
original design for the web; namely that it wasIt also made instant paper millionaires of many of
possible to see easily what any site was linking to butGoogle's employees.
very hard (if not impossible) to see the sites linkingPerhaps the biggest irony of search history is that
back. Backrub sought to catalogue all the links on theAltaVista and the rest were not necessarily wrong in
web and, through doing so, establish which sites hadtheir obsession with portals. They were simply right
the most "authority" overall and in any given topic.at the wrong time! The preoccupation of today is
Page insists that it was never his intention at thisnot dissimilar to that of 1999, with Google and others
point to create a search engine. However, intellectualall seeking to built "stickability" into their proposition.
curiosity drove him forward into the creation of aSuffice to say for now that those IPO billions in the
new kind of crawler, based on a link ranking systembank were helpful in the acquisitions of YouTube and
(now called PageRank after it's inventor) which wasDoubleClick in 2006-7. This concludes my three part
the first of it's kind.series.