December 19, 2004

Google and the Great Mousetrap Fallacy

What does Google Stand for. How has it Grown So Big So Fast. And Why Neither Yahoo! Nor Microsoft Pose a Threat   (UPDATE at the end.)


All of Google hardware from its Stanford days (1998) where it was initially called backrub. Google now operates over 250,000 Linux servers of its own design. Hold mouse over each box to learn more about it. View larger images.

Seth Godin recently proposed a human edited news service at Froogle to offer "news about sales and bargains and closeouts and new stores." It's not Seth's most original idea by his own admission but at first glance it does make sense as search for discounted products reaches its peak during the holiday season. There are other sites that offer this but one can always expect Google to do a much better job and attract much more traffic than anyone else. Yet, if you think through this, it’s a surprising suggestion from Seth because he usually has remarkably incisive insights to make.

A human edited product news feature is never going to happen at Google.


Google is a Design Company that Employs Technology to Solve Big Problems

Seth Godin's idea shows a complete lack of understanding of how Google works and what it stands for. Google isn't a content provider. They are not a media company, their strength is technology. As Sergey Brin once said, "Google tries to apply technology to media." Google Search employs mathematical algorithm to search, index and present the web in a way that’s most relevant to a user. Google News is a news aggregation service that collects news from all over the web. Gmail is a new, vastly more efficient way of accessing, managing and searching your emails over the web. Froogle is a product search tool that serves as an excellent comparison platform.

Two days ago, for example Froogle added a feature that automatically culls reviews from various web sources, identifies them to be positive or negative, weighs them over and correspondingly adjudicates a rating score for each product on a five point scale. All with the help of a mathematical algorithm.

What's common across all of Google's services and products is their focus on solving users’ problems employing technology in a lot of different areas. Google did not grow the “natural growth of a search engine” -- that of becoming a portal and diversifying into a provider of dozens of services, the kind Yahoo! does.

Google doesn’t work that way. It doesn't employ technology for the sake of it. It doesn’t enter a field, because it was ‘natural extension.’ It doesn’t even run after big money. Froogle, its shopping search is still in beta after two years and it doesn’t charge merchants to get listed nor does it take a cut from a sale. Google News, Google Image and Desktop search do not have a revenue stream yet.

Google is more interested in solving users’ problems. Search (web, desktop), advertising, email, shopping, news, localized search, wireless search, catalogs, print; all these areas posed big, critical user problems that Google identified and then solved in unique ways.

To me, this is Design. Very innovative design at that.


Better Mousetraps Do Not Necessarily Make Better Businesses

In 'What Should Google Do' ebook, Seth Godin describes what I believe is the greatest myth surrounding Google. I've heard it in numerous articles and reports over the past two years. The argument goes like this: Google search is just a technology. The company that comes up with a better search engine will surpass Google because there is no switching cost.

There's no truth in this assumption either. It's what Jeffery Timmons calls the great mousetrap fallacy made famous by Emerson's line, "If a man can make a better mousetrap (...) the world will make a beaten path to his door." Jeffery Timmons is an entrepreneurship expert and author of the best work on the subject, New Venture Creation. He calls it a fallacy because as every venture capitalist knows, success is never that easily assured.

Superior technology is never enough. For one thing, there has to be a need for it. All this talk about search industry heating up with competition from Yahoo and Microsoft, that Google will soon have to share ad revenues with them is nothing more than speculation. Why doesn't anyone stop to come up with a single reason as to why satisfied Google users will even consider making the switch?

A better mousetrap will not sell in a town where every house already has one that works pretty well.


The Forgotten Problem of Search User Experience

Did Google gain popularity because it built a superior technology? NO. That would be wrong way to look at it because it's not the complete picture. People switched to Google because it solved an important problem. A problem big enough, widespread enough and one that was understood [by users] to be a problem. The poor user experience of search. This is why users considered moving to Google when the alternative search engine came out.

There was largescale dissatisfaction with the performance of the mainstream search engines of the time. Remember this was when search was a decidedly unglamorous business with no money in it and when everyone was going the portal way. It was a time when search engine homepages were riddled with text and graphics; ads were disguised and placed within results; there was no certainty that the best result would be on top of the page or even in top five pages; and when search results gave little or no contextual information about relevancy of a particular link.

As a result, there was a lot of trial and error involved in finding useful information. With almost everyone on dial-up at that time and search engines being nowhere as fast as they are today, conducting a search was a taxing exercise for an ordinary user.

This is clearly not the case today. Google's the page rank algorithm, its speed, its clean and efficient user interface and continuing innovation to enhance its service has elevated search performance far above what it has been in past. While user expectations have risen as well and the perfect search seems a long way from realisation, but the search problem some would say, has been solved. Your favourite search engine does not frustrate you anymore. There's no large-scale dissatisfaction with its use.


What is the Incentive to Change?

We humans as a rule do not favour change when things seem to be going well. We go through it only when we are forced or when we are frustrated with our current environment. Even assuming that MSN or Yahoo! achieve the unlikely prospect of creating a somewhat superior search engine, there's no apparent incentive for a user to make the switch this time.

Yet tech writers continue to propagate the better mousetrap myth. In an indepth look at Google's growing stature in Technology Review, Charles Ferguson argues that Google must take control over computer architectures by making available proprietary APIs in order to compete with Microsoft. This is perhaps the most interesting look at Google but it is clearly written with a Microsoft bias. How Microsoft works, how do companies succeed or fail against it, what should Google do to tackle Microsoft and so on. The purpose of this post is just the opposite - to show how Google works, how it got to where it stands today and where it is going.

The footnote to the nine-page article adds that Ferguson "holds a substantial quantity of Microsoft stock." If that leaves one with any doubt over who he thinks will survive in the battle, it's removed in the next sentence. The note adds, "he has no other financial interest relevant to this article." Meaning he doesn't own Google stock. Meaning, "I'm betting on Microsoft and so should you."

On the very first page of the article Ferguson completely discounts Google's competitive advantage of a positive search user experience that it has built over the years and writes: "...nothing prevents the world from switching (painlessly, instantly) to Microsoft search services and software, particularly if they are integrated with the Microsoft products that people already use."

It's like saying that everyone in the town would discard their favorite mousetrap if someone supplied their home with a new unfamiliar one for free. But MSN search has already long been integrated with IE [since version 4?]. This hasn't led to users ignoring other search engines so far. Although it's a possibility that some users may move if Microsoft integrated its search seamlessly and kept it visible. In the past it has taken the form of a button that no one used. Microsoft would score some points if it makes it a text field in IE like that of the Firefox or if it placed it on the Desktop like the Google Deskbar. But even then it won't be easy.


Users Rule the Web and Google is a Darling

What prevents the world from abandoning Google - the switching cost, if you will - is the good experience users have had on Google in the past and the enormous trust it has thus garnered. Google is a darling of the masses. Call it brand loyalty if you will, this, in my view is one of the reasons why Google survived a possible IPO debacle on its debut. An IPO that the Suits at Wall Street effectively boycotted as they learnt they weren't invited to the party [of course, in the succeeding weeks they came back in hordes to buy the stock when it became apparent that Google was doing pretty well without them]. Google stock survived and is now thriving primarily because of its sound fundamentals but also because people have enormous faith in the company.

In the Tech Review article, Ferguson provides us with a doomsday scenario for Google, saying if Microsoft acted logically, it would "perform a 'cashectomy' on Google—just as it did in the browser wars when it gave away Internet Explorer. Even with nearly $2 billion in cash, Google is vulnerable to this tactic. For instance, Microsoft could offer free wholesale access to its search engine. Then it could attack Google’s advertising networks by offering free or subsidized advertising placement."

Thankfully, this is nothing more than his imagination running on steroids. Here's why. Microsoft doesn't control the web. To kill Google the way described here, it would have to first get users to abandon Google before expecting the world's advertisers to abandon it. Advertisers would go where the clicks are and right now they are with Google. If Microsoft followed such a policy, it would be inviting a new crop of cheap advertisers [read spammers] attracted only to the low priced/free service. You would have people linking to Viagra sites posing as ads for web hosting. Such a scenario inevitably reduces the credibility of advertising links, leading to further reduction in the limited click through rates.

There's no way to beat Google at its own game as long as users are with it.


Google's Really Good and Getting Better Everyday

As of today, any new search technology would have to be vastly, tremendously superior to expect a significant percentage of users to consider a switch. It would have to be better than Google by a magnitude much greater than the Yahoo vs. Google divide of 1999/2000 because people now have little reason for change.

This appears impossible in the near future because Google already has a pretty good lead. Yahoo! and Microsoft are still playing catch up to the technology Google mastered four years ago. There have been a number of improvements to Google over the years and it adds newer algorithm subsets every month to enhance its core search technology.


The Perfect Search Engine Requires a Breakthrough in Personalisation

Google is very very good at what it does even at the subtlest level. I don't see a "quantum leap" in search quality coming anytime soon. If it does, expect Google to be at the forefront. There are a few reasons why. One, as I already mentioned, Google is way ahead of others and is ever improving. Two, Google is an innovation engine. No company is spewing out innovative technologies at the rate at which Google does. If any breakthrough will come, it's most likely to happen at Google Labs.

Third is what I'd call the Java search paradox. It needs a bit of explaining. Web searches as they are done today, can be divided into three main components. Conducting search - translating your need into a query and entering it in the search interface. Result presentation - the results UI, the way they are presented for interpretation. And result accuracy - relevance of the results or search precision.

Any quantum leap in search user experience in future would have to occur in search relevancy as it has in the past. Improvements in conducting search and result UI, Although important, can only enhance the user experience incrementally. They are irrelevant if results aren't accurate. What's most important is to lead the user to what he's looking for. If a search engine can predict that even with failing accuracy, it will be the beginning of the perfect search.

Such precision in result accuracy however, is held back by what can be called the Java paradox. As Yahoo's Jerry Yang recently noted in the Yahoo! Search blog, "people expect to find precisely what they're looking for exactly as it relates to them. It's the old example of the "Java" search query. Are you looking for coffee or for the programming language?"

When Sergey talks about the perfect search engine being like "the mind of God," he's perhaps talking about this very mind-reading quality of a search engine.

If such a leap is possible, it's difficult to imagine it coming from anyplace other than personalised search though it isn't as simple as it sounds. Personalisation is extremely complex. Google has made some headway in it (with the beta launch of their Personalized Web Search) though I don't believe it's the right direction. Efforts by Yahoo! in this area have been commendable. Yahoo! also has the advantage of having easy access to personally identifiable information of its users.


Flexible and With Eye on Emerging Technologies

Lastly, even if a brand new start-up does create a vastly superior search engine sometime in the future, there's no guarantee that it will be able to make it into a successful business as well. One may speculate Microsoft or Yahoo buying out the venture but that's no more likely than Google buying it first. Also, big companies are traditionally slow at recognising emerging disruptive technologies. Remember just a little over four years ago, Google approached Yahoo with their algorithm and was turned away.

Recent acquisitions by both firms throw some light over what to expect in the future. All of Yahoo's recent acquisitions have been reactionary, made in a strong bid to survive the search wars. Google, on the other hand has proactively sought and acquired companies wherever it has seen an innovative use of a technology to solve an unsolved problem. The other area in which Google has invested is where it expects rapid growth. Google has acquired companies even when it may have no plans to integrate that technology in current products.

So in the event of a startup gaining an advantage, we can expect Google to identify it before others and moving swiftly to make an acquisition.


UPDATE 22-Dec: In the past two days I discovered more threads from across the web that confirm some of the arguments I made here.
  • Ten days before my post, Danny Sullivan of Search Engine Watch made exactly the same argument about how Google evolved and why a better search won’t take users away:
    We've long had reports that consumers use multiple search services. But the classic AltaVista to Google defection example shows that defection tends to happen not just because another tool is better but also because the existing tool is bad.

    In other words, my view is that it's not that MSN needs to be "good enough" to get people from Google or even "better than Google" to gain defectors. People will only kick the Google habit if they feel Google is getting worse.

    If Google itself continues to be "good enough," helping consumers find what they are looking for most of the time, I don't think you'll see big defections happen. In fact, if AltaVista had continued to be "good enough," Google might never have been able to emerge as the powerhouse it is today. But AltaVista wasn't good enough -- it had gotten bad.

  • Tom Foremski, a senior writer in the Silicon Valley bureau of the Financial Times wrote a personal account of Google culture in his blog. He talked about why Google is a technology company in the heart of a Media business; something that I discussed in the beginning of this post. (also read Yahoo’s view)
    I asked Mr Rosing [VP of Engineering] why Google News doesn’t use a few full-time editors to clear out duplicated stories and clean up sometimes poorly aggregated pages. He looked at me as if I was crazy. Why would Google want to use people? Google News is an engineering solution, he said.
    [...]
    The conversation with Mr Rosing was striking because it revealed how Google sees itself. It sees itself as a hard-core engineer culture that has virtually no managerial controls, and yet can toss out beta projects such as Google News for many years to come. That’s fine, but, Google is a media company isn’t it? But no one seems to have told them that.

    Google publishes pages of content and sells ads on those pages. That’s a very traditional newspaper or magazine business model, except that Google uses machines to harvest and publish the content instead of using editors, reporters, copy editors, etc. Advertising is also harvested by machines, running a simple auction system between advertisers. And the distribution channel is the global computer network of the internet. It’s a very efficient business process, but, it is a media business process.

  • I also received a note from Herb Brody, an editor at MIT’s Technology Review magazine, which published the Charles Ferguson article I mentioned in my post. He was nice enough to stop by to read the post and wrote to ask if I’d write a letter about this article for their magazine’s next issue. Just sending him a concise version of the argument.

Notes and Links

Seth Godin’s New Google Idea. Read the post. In the past his some other Google ideas have been equally far out. For example, from Google's big opportunity [Apr 2004]: Turn Gmail into a pay service at a nominal cost and require users to pass through some kind of validation to get a Gmail address thereby making it a 'gold standard'.

More ideas from What Should Google Do pdf (253 kb) Aug 2003 : Get email addresses of users and mail them coupons for the stuff they're searching. Record personally identifiable information with their permission and track what they are searching and where they are going. Use this knowledge to organise the web better by offering webmasters predefined matadata that they could apply to links called plex that ‘Google controls.’

Further: "Click on the plex and Google automatically takes you to the next page most likely to satisfy someone like you doing a search like that. Pages without plexes, naturally, won’t be as appealing to surfers, so quickly every webmaster will install a plex, which furthers Google’s functionality and power."

Sergey Brin on Google as media or technology company. In a News.com report from Aug 2003, Brin said that despite the changes and questions of what exactly the company is--a technology or media company--it remains a technology company that "tries to apply technology to media."

Froogle’s Reviews and Ratings News.com: Froogle turns to Web for product reviews

The mousetrap fallacy. Ralph Waldo Emerson’s complete statement: "If a man can make a better mousetrap than his neighbour, though he builds his house in the woods the world will make a beaten path to his door."

Jeffery Timmons called it the great mousetrap fallacy adding that Emerson did the greatest disservice to future entrepreneurs with that line because it suggests that success is certain with a new technology idea. Look up Timmon’s tome on Entrepreneuship- New Venture Creation

Famous management guru, Peter Drucker flayed this quote too in his book Innovation and Entrepreneurship: Practices and Principles.

In a 1997 HBR article, Harvard Business school professor William Sahlman says, "whatever the reason, better-mousetrap businesses have an uncanny way of malfunctioning."

Charles Ferguson on What’s Next for Google in Jan 2005 cover story of MIT's Technology Review magazine.

Yahoo’s Jerry Yang talks about What’s Next in Search in Yahoo Search blog.

Personalized Web Search by Google. View Frequently Asked Questions.

Yahoo! on Personalisation Interview of Yahoo! executives about personalisation and competition from Google. Part one and part two.

December 06, 2004

In Bangalore

I'm in Bangalore for the two day IHCI conference. Today was the first day. I will perhaps write more about the conference later but just stopped by to say that I love this city. I lived here for about three years and am visiting it after a gap of two. Just wrote this to a friend:
Never thought meeting old friends and travelling down the once familiar but forgotten roads... that become familiar once again would give me so much pleasure.

Bangalore is like a distant dream you dreamt the other night that suddenly comes back to you one day and then you watch it being laid out in front of you.

As someone said, it's deja vu all over again. :)

I leave Wednesday evening.

Notes and links

Learn more about the conference and the organisers, Indo Europeon Systems Usability Partnership.
Nice website of conference venue, Taj Gateway Hotel