Hacker News new | past | comments | ask | show | jobs | submit login
What were the key decisions that Page and Brin made in the early days of Google (quora.com)
157 points by helwr on July 10, 2010 | hide | past | web | favorite | 50 comments

The highest ranked answer on that thread, though it contains some interesting details, is way off. Managing VCs well was not what made Google successful-- or what makes any startup successful. The real key decisions were things like realizing search itself was important, at a time when all the other search engines thought it was unsexy, and were trying to get people to start calling them "portals" instead; designing the architecture to work on large numbers of unreliable, cheap computers; understanding how important speed was; making the site uncluttered; deciding to hire only very smart people; etc. That's what made the company valuable, and if it hadn't been valuable it wouldn't have mattered how well they'd avoided dilution.

It's also full of inaccuracies. The "interesting details" are largely wrong. Unfortunately it's written in an authoratative style, so people assume it's true and upvote, demonstrating one of the major flaws in vote based systems.

Plus, quora is not really living up to it's promise of a high quality question/answer social site. Most questions and answers are as inane as on yahoo answers and the only difference is the more pleasing UI but that wears off pretty quickly.

I don't know it seems to have a reasonable summery:

The most common decisions cited are:

    * Realizing that search was important.
    * Focusing on speed.
    * Maintaining corporate control.
    * Focusing on the user experience and deferring revenue.
    * Scaling out computation over a large number of cheap computers.
    * Hiring very smart people.

Agreed re voting systems.

I believe that a good Q&A site should use "voting" to recursively find smart people who are similar to you and can provide you with good answers....as opposed to showing the most popular/well-written/pithiest answer. "PeopleRank", maybe.

That's an interesting idea.

The problem is preventing this leading a site dominated by a group of people who share a set of extreme opinions of one sort or another. There's no now site where you can't get karma for reinforcing some commonly held group opinion and if Karma became self-reinforcing, the problem would get harder.

It might not be impossible - maybe you could have secret algorithm for rewarding people only for making well-respected non-polarizing posts. But it would be quite tricky.

Ah, yup, there's ways around that.

The Q&A PeopleRank idea will probably be my next start-up, if Vark doesn't beat me to it. Reading http://blog.vark.com/?p=352, it seems that they're on the right track. (Although their UX is a little clunky).

However, even if Vark builds a kick-ass product, you can still apply the PeopleRank idea to news-sites like HN.

Anyone interested in building a Q&A/news site feel free to message me.

I agree with this and have been saying it for a while. Quora could avoid inaccuracies like this one by weighting a person's answer on a topic based on their endorsements in PageRank fashion. e.g. I endorse you on Google, which has more weight if I am myself endorsed on Google with the weight increasing based on the number of times I am endorsed and how many people have endorsed the people endorsing me and so on. I'm relatively sure Quora will do this eventually.

I started building a site that works kind of like that. It is essentially a reddit clone that will rank articles by how likely you are to vote them up. I put it on hold and decided to first build a product that users would pay for. But I would still like to build this site someday, or use it if someone else builds it first.

What do you consider the best available information on google's history?

hey paul! why don't you write an answer there, instead of commenting here? the quoran community could benefit too :)

I suppose, before I upvote you, that I should ask you for sources then.

Paul was Google employee #23.

I would also like to point out that he has been a member of Hacker News today for exactly: 1234 days =)

There are really only TWO decisions that google made that made them successful. When something works, people like to go back and look at complicated things and say that those are the reasons it worked. The reason that google beat the others were really very simple and they had to do with the front-end only:

1. They were the simplest to use at the time. All the other search engine things had a lot of frills and pages and ads and so on. Google was just very plain, and as a result, simple

2. By default, they included ALL the words you typed in. Competitors at the time were trying to maximise search results by searching for each word individually, which led to a lot of irrelevant results. Just by including all the words in the query, google had a much better hit rate

All the rest was unimportant. Those two simply front-end decisions made the google search engine loved by the geeky anti-corporate people who ruled the web at the time (slashdot, linux advocates), and they recommended it to everyone who was joining the net. By appealing to the geeks, who then acted as the authority figures to the normals, google was able to take over the search market back then.

It was really not that complicated.

I beg to differ.

First, it is simply not true that Google "included ALL the words you typed in". Google ignored "common" words like conjunctions and certain pronouns for a very long time.

Second, I think it's fair to say that Google's speed and quality of results were significant contributors as well.

Third, Larry's decision to make AdWords ads live immediately was, I think, one of the key factors in drawing users to AdWords.

Fourth, ad syndication through AdSense was a significant contributor to Google's revenues.

I could go on. I think it's nowhere near as simple as you say.

Adwords came MUCH MUCH later, when Google dominated the landscape. Of course, by all words, I don't mean useless words like conjunctions.

When people think back, they blur factors (like you are blurring adwords with googles initial success). I was there when Google took over, and it WAS as simple as I said.

They first had a popular search engine, and they built a good business on top of that. The business building part was complex, probably, but getting the search engine to be popular was just a front-end thing.

Googles speed was not relevant, Altavista and many other search engines prominently displayed how fast they were. There was a sort of speed competition at the time, so Google was not significantly faster than the others. Those other engines (like altavista) did not even have websites with lots of images or so, so it loaded just as fast as google.

Picking the right user-interface has been the reason for success for many companies. People will come and overcomplicate it later, but in the end, it was just the user-interface that made them win.

> Adwards came MUCH MUCH later

It's AdWords, not Adwards. And it launched in September of 2000, when Google absolutely did not "dominate the landscape."

Do you even bother to check your facts, or do you just make shit up?

In September 2000, Google was already the most important search engine out there. It was already successful, and adwords had nothing to do with its success. It had 20million in the bank already.

Please avoid the personal attacks, it's not good form.

In September 2000 Google had money in the bank only because it had closed a $25M VC round just over a year before. Its revenues were tiny. Its market share was tiny. It was not profitable. Its success was far from assured. By the time Google turned its first profit, the vast majority of its revenues came from AdWords. To say that AdWords had nothing to do with Google's success is akin to saying that Windows had nothing to do with Microsoft's success.

And personal attacks are not nearly as bad a form as promulgating demonstrable falsehoods as if they were facts.

It's market share was not tiny. It was growing rapidly, and was already dominating.

I think I've had enough of this argument. People like you are never wrong - you are arguing absolutely orthogonal to me, and punctuating it with personal attacks.

Actually it is you who are in the wrong here, though the voters on this thread don't seem to realize it. The guy you're arguing with was an early Google employee. He knows what he's talking about.

Thanks for backing me up PG.

BTW, just noticed this too:

> by all words, I don't mean useless words like conjunctions

Conjunctions and other common words are far from useless. If you ignore them you will go badly wrong on queries like "To be or not to be".

That's not a particularly strong argument. It's just a call to authority argument without actually adding any real information one way or the other.

You do not understand the argument from authority. Argument from authority would be if PG had said that I'm right because, say, I had a Ph.D. But being an early Google employee when the topic under discussion is the early days of Google doesn't make on an authority, it makes one a primary source.

But you may have noticed that I did not bring this up. That's because your factual claims are easily debunked with information from the public record.

How can such a long argument go on and nobody is simply making a post with a series of numbers showing search engine marketshare from 2001?

pg, the early Google employee has not made any statements to contrary.

He has not asserted anything but the existence of unspecified inaccuracies. Paul was neither on the board of directors and was not involved with this decisions. He is not best source.

Growing rapidly is not mutually exclusive with being small. In fact, it's much easier to grow rapidly when you're small than when you're big.

Here's a reference in support of my position:


This article is from October of 2002, two years later than the time frame in question. At that time, after two years of exponential growth, Google's market share was 30%. In September 2000, Google was still an upstart. Three months earlier the big news was that Google had beat out Inktomi for a contract with Yahoo:


Google's overall market share at the time was so small that the press reports don't even mention it. Google didn't even register on anyone's radar as a major player.

Can you cite any references to support your claim that Google was "already dominating"? No, you can't. Because they weren't.

> I think I've had enough of this argument

A common tactic from someone trying to defend an untenable position.

> People like you

Excuse me? Weren't you saying something earlier about personal attacks?

Here: http://www.google.com/press/pressrel/pressrelease25.html

Google was supplying the search results for Yahoo in the year 2000, long before Adwords. Yahoo was one of the biggest portals at the time. I really don't understand how you can say Google was not relevant before Adwords.

Google was the king of the search business back then, and that was long before Adwords came along.

Adwords was not critical to Googles share when it came to searches. Google was growing massively, everyone was paying attention to Google long before Adwords came on the scene.

Adwords allowed Google make money, but it did not contribute to growth in search.

Who was the major player in search at the time Adwords was introduced? Ask.com? Lycos? Altavista? Or perhaps, Yahoo, powered by Google?

Your position has gotten a bit blurred the more you argue. To clarify again:

- My statement is that Google was a rapidly growing search engine because they made the right UI decisions

- Your statement seems to be that without adwords, Google would have stopped growing and ultimately failed

Am I understanding your argument wrongly? Because it does not make much sense.

> Adwords was not critical to Googles share when it came to searches

Even that is arguable, but that is not the topic under discussion. The topic under discussion is this statement you made:

"There are really only TWO decisions that google made that made them successful"

> Adwords allowed Google make money

What in the world do you think "success" means? Despite the fact that this forum is called Hacker News, it isn't really about hacking per se. It's about starting companies. "Success" in such endeavors is not measured by how many people use your product, it is measured by whether or not you make money. To be sure, all else being equal it is easier to make money with a large market share than a small one. But you can make money without a large market share (e.g. the Apple Macintosh), and you can have a large market share without making any money (e.g. the Apache web server).

So maybe you meant to say, "There are really only two decisions that Google made that allowed them to achieve a large market share" or something like that (though even that is arguable). But "large market share" and "success" are not the same thing.

But in September of 2000 Google had neither success nor a large market share. All they had then was a kick-ass product and a really good team. Again, all else being equal, having a kick-ass product is helpful. But it is neither necessary (c.f. Microsoft) nor sufficient (c.f. Napster) for success.

This is true. Look at Google S-1 filing. It validates everything you said.


There are many people here who want to argue back and forth about their opinions and are not concerned about having data.

I cannot believe you are being personally attacked over this when the facts of matter are so easy to research.

My takeaway was that they managed dilution so that they could guide the product and the company the way they felt was important. Managing VCs well allowed Google to manage the product well.

Interesting thought experiment: Would Zappos have sold to Amazon if they had managed dilution better?

According to Tony Hsieh's book, no.

The two of you are looking at their success from opposite directions. You're looking Google from day zero and asking, "Why were they successful at all?" The first answer looks at their mammoth success and asks, "How did they avoid becoming a smaller company as they grew?" The answer isn't as simple as the initial design of Google, or how they managed investors. They took hundreds of great small steps in the beginning that made them number one. Cultural ones, such as free food, "don't be evil," and hiring the best people. Technical ones, like BigTable, MapReduce, the Google File System, and their obsessive focus on operations. Business ones, like the terms of their IPO and their managing of VCs. And entrepreneurial ones, like 20% time, acquiring lots of startups, and awarding employee bonuses by product usage. "In the early days" is a fuzzy measure, but these are all decisions that Google has made to continually grow their company in value

Yup. At a time when everyone else had given up on search, mainly because they thought that was as good as it could possibly get and thus the future was in other "value add" areas such as portals and directories, Google's founders saw a way to do it better. They married multiple revolutionary innovations (page rank, commodity PCs as servers, sharded computation / "datacenter as a system") and combined with a little bit of usability sense they provided higher quality search results which rendered faster and at lower per result costs.

What they did was enter an existing market with a lot of competition and do something that the competition was absolutely incapable of matching without going back to the drawing board (there are a few other companies who do this too, the list is left as an exercise for the reader). They followed it up with new ways of advertising online that made search a money maker.

What they did was enter an existing market with a lot of competition and do something that the competition was absolutely incapable of matching without going back to the drawing board (there are a few other companies who do this too, the list is left as an exercise for the reader).

But that's teasing! Your observation is a shrewd one and now I want to know who you think the others are.

I think the key point re managing VC's is that they avoided being forced into an early exit and retained control over the product by keeping control of the company.

Most mature investors want two things: The value of their stock to increase and liquidity opportunities (opportunities to realize that value). They want both to happen rather sooner than later due to the time value of money. They also believe that if they have more control they can achieve these first two objectives more effectively.

These investor goals are not always aligned with what's in the best interests of the product or company which is why it is sometimes a good thing for founders to retain control.

VCs don't want early exits. They want companies to go public, which Google did pretty much as soon as the IPO market returned. So whatever friction there may have been between Google and their investors, it would not have been about that.

That, however, conflicts with the alleged desire of VCs and CEO to focus on enterprise search and abandon consumer search. I don't know how true that allegation is, but if it is true, and VCs had a larger slice of the pie, it's possible that that desire would have been followed, hugely reducing Google's value.

Paul, I wrote that answer to be concise. I am not trying to write another book on Google. When you build a Google, you can tell me I am wrong.

Everyone I have spoken to that has built a successful company has told me that their ability to manage the long term interests of their company was impaired by losing control of their board of directors.

If Mark Zuckerberg did not have control of his board, the company would have been sold to Yahoo for just under a billion dollars, where Facebook would have died. The Yahoo executives wanted to "Nascar up the site". It would not be a 30 billion dollar dollar company today if the board had been investor controlled.

The reason Friedster failed was that the company was managed from the board of directors by investors which did not use their product. The founders of Friendster were screaming at the board of directors that the website took 40 seconds to load, but the board was more concerned with expanding into new markets and adding features, rather than solving the scalability issues the company was facing. Myspace copied Friendster, solved the scalability issues and dominated until Facebook introduced the feed and notification mechanism and took the college demographic.

If you want to build a long term success like Google, you have to be in a position to ignore your investor's short term desire to improve the company's financials by sacrificing long term growth. The quickest way to increase your margins is to reduce investment in the future.

Most of Google's costs were from their consuming facing search product. Almost all of Google's revenues were from enterprise search. Dropping consumer facing search would have improved Google's financials for sale or IPO. Everyone at the time believed that search engines were loss leaders for portal sites. If the management team did not have board control, the decisions that would have been forced on the company would had led Google to a different, much grimmer corporate destiny.

The management team at Google believed strongly in the future of internet search and the growth of online advertising. They saw the potential of using search queries to target advertisements and that search engines were uniquely positioned to monetize lead generation; at a time when search engines were loss leaders for portal sites. If the company had cut its loss centers and focused on its profit centers, the company would not have been as successful as it is today. The operating autonomy of Google's management team was critical to Google's long term success.

I do not care what Paul Buchheit says. Eric Schmidt has on several occasions discussed the advantages of Google's ownership structure and how it enabled Google to ignore outside influences and focus on the long term success of the company. If Paul Buchheit disagrees with the CEO of Google about the importance of Google's ownership structure and the operational autonomy of the management team, then its unfortunate.


LOTS of useful stuff here. In addition to a lot of the usual stuff, most especially keeping control, there are these not unrelated gems:

"Retaining control over the company proved to be crucial to Google's success. Most of the early company revenues came from enterprise search, and the investors and Eric Schmidt pressured the company to drop consumer facing search and to focus on the enterprise market. However Page and Brin disregarded this advice as they anticipated the growth of a market for online advertisement. At the time this decision was being made, the New York Times was quoting experts as saying "No one will ever make $250 million dollars a year from online advertising." It has also been said that Page and Brin disregarded pressure from investors to copy Yahoo! and diversify Google into a portal site, deciding instead to focus on the core search and advertising market."

We also now understand their particular style of secrecy: avoiding getting crushed by Microsoft back when that was the fear of every high tech company it could possibly compete with.

The comments as of now are also of high quality and some are not to be missed.

It was surprising to learn that Larry Page's brother already had successful startup experience.

As a customer I know these key decisions: i) Self managed ads system. Before AdWords it was impossible to buy $ 5 of ads. Even years after AdWords Overture (Yahoo) and AdCenter (Microsoft) couldn't even copy the smooth process (I remember Overture charging me a minimum of $ 15 and their staff changing my keywords) ii) Text Ads instead of banners as a more respected way of advertising.

Lots of entrepreneurial navel gazing going on both in the linked site and in these comments. Everybody has their own theory about the reasons x y z why company a b c is successful, be it Google, Twitter, Facebook, etc. The problem is every opinion is partially correct, no one can disprove anything, and there is no right answer.

The stories of successful companies are inspirational to some degree, but as a way to vicariously learn how to be successful they fall woefully short of the mark.

The best thing for entrepreneurs to do, is simply to make something happen. Build something people want, and if you fail, learn what you can and try again.

You never know and can't know which decisions are the ones that matter, so forge ahead.

God bless you.

This is the most reasonable thing I have seen said here.

I think none of the answers in that summary really say it. Google won (in the beginning) because:

  1. It *just* searches. Really well.
  2. Ad auctions.
That's basically it.

It's the same way Starbucks won in the beginning:

  1. *Just* coffee. Tasty coffee.  
  2. Charge for it.
Or Apple:

  1. *Just* the user experience.  A great one.
  2. Charge for it.
(Google, Apple, and Starbucks have now all diversified widely, with mixed results).

Excellent set of answers, this was my first foray into Quora, btw; if other answers are half as good, it's a great resource.

Among other key decisions they got right (and some lucking out), the one thing they got absolutely right is the culture. Not just perks, food, etc. because during the boom almost all cool startups had these (or better), it's much more than that. The culture components ones that I think were most influential were the 20% time and committee decision making (with actual engineers, not suits).

Having the right culture is not sufficient for success, but without it, I think, all most certainly will be lost.

There are many factors that lead to Google's success, but I know what made me switch to Google back in its early days. It was faster (probably because it was sparse), it got good results, and I felt cool for using it (maybe because it was the underdog, or maybe because it had a "for geeks, by geeks" feel to it).

Provides a good reason to downplay having a fancy interface - avoid tipping off the competition.

Nobody was really serious about web search prior to Page & Brin.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact