Skip to content

Seven Reasons to Doubt Competition in the General Search Engine Market

New books on Google by Randall Stross, Alexander Halavais, and Jeff Jarvis have been getting a good deal of media attention lately. I highly recommend the Stross and Halavais volumes because they recognize that the unique power of Google is likely to be lasting. Stross notes that the company may be using a million computers to index and map the web at this point. If he’s even within an order of magnitude of the real number (a strictly protected trade secret), that ought to give pause to any technolibertarian who thinks a “Google-killer” can be cooked up in some Silicon Valley garage.

A critical mass of factors make it extremely difficult for any serious competitor to emerge in the search space:

1) Trade Secret Protections for the Search Engine Algorithm: Unlike patents, which the patent holder must disclose and which eventually expire, these trade secrets may never enter the public domain.

2) Network Effects in Improving Search Responsiveness: The more searches an engine gets, the better able it is to sharpen and perfect its algorithm. Incumbents with large numbers of users enjoy substantial advantages over smaller entrants. For example, if a search engine finds that everyone in a given area picks the third result instead of the first result in a given day, it can tailor results for that area to increase the salience of what was once merely the third result.

3) Content Licensing Costs: A key to competition in the search market is having a comprehensive database of searchable materials. The ability to obtain exclusive legal rights over searchable materials, however, may substantially increase the cost of obtaining and displaying this data and the metadata needed to organize it. Uncertainty of costs can also be a factor; if the Google Book Search lawsuit settles, Google will have overcome that uncertainty, gaining an advantage over all other potential entrants in that search space.

4) Consumer Habit: Many searchers are accustomed to using a certain number of providers, use them relatively habitually, and are reluctant to switch, despite the existence of alternatives.

5) Personalized Search: As a user queries a search engine, he or she can train it to custom-tailor the results to his or her interests. For example, if a user habitually searches for recipes, the search engine may weigh food sites more heavily than other sites when confronted with an ambiguous term (such as “cake,” which could refer, inter alia, to a confection or to the rock band Cake).

6) Two-sided market dynamics: Given Google’s market share of roughly 70% in search advertising, advertisers vastly prefer it to other, less trafficked alternatives. To see why, imagine you are looking for an internet dating service in a town with 100 people. Even if the service with 10 people signed up told you it had a wonderful new method of matching you with potential dates, wouldn’t you prefer the one with 70 members instead?

7) Non-portable AdSense Data: Harvard Business School Professor Ben Edelman has investigated another self-reinforcing aspect of Google’s market power: the non-portability of AdSense data, which makes it difficult for Google customers to apply what they have learned about their internet customers to ad campaigns designed for other search engines.

Nevertheless, the press is always looking for the next “Google Killer,” as this article shows:

[Google] has always fallen short of its most ambitious goal: getting a computer to answer a simple question, asked in natural language. Now, British scientist Stephen Wolfram claims that he may have the answer, and his new search engine could theoretically remake computing as dramatically as Google did a decade earlier.

However, the story quickly turns to a recharacterization of Wolfram Alpha as less a “killer” than a “complementer:”

[N]ova Spivack, a leader in semantic Web technology, hung out with Wolfram for two hours, playing with the new search engine. And he walked away awfully impressed. “Where Google is a system for FINDING things that we as a civilization collectively publish, Wolfram Alpha is for COMPUTING answers to questions about what we as a civilization collectively know.”

However, Spivack claims that Google and Wolfram Alpha will serve completely different functions in the online world, and users will ultimately use both in equal measure. “Wolfram Alpha, at its heart is quite different from a brute force statistical search engine like Google,” Spivack argues. “And it is not going to replace Google — it is not a general search engine: You would probably not use Wolfram Alpha to shop for a new car, find blog posts about a topic, or to choose a resort for your honeymoon. It is not a system that will understand the nuances of what you consider to be the perfect romantic getaway, for example — ”there is still no substitute for manual human-guided search for that. Where it appears to excel is when you want facts about something, or when you need to compute a factual answer to some set of questions about factual data.”

Expect many more web startups along those lines, particularly in “vertical search” (i.e., particular topical search spaces, such as gaming, B2B transactions, etc.). There’s good reason to think that specialty areas and functions will get their own search and computational organizers. But I find it hard to believe any prediction that a real Google rival will overcome all seven of the competition-impeding dynamics listed above.

PS: The seven reasons above have been developed in my work on search engine rankings and in work co-authored with Oren Bracha.

6 thoughts on “Seven Reasons to Doubt Competition in the General Search Engine Market”

  1. I do believe that Google will never face competition in the general search market; but not for the reasons stated. Every business in the world faces those same seven roadblocks, and more, as any potential Google competitor faces. Those reasons don’t matter, because all businesses, ultimately, fail, for one or more of those reasons. That is is the method of markets and market discipline.

    A market starts with several participants or competitors. Markets, functionally, do only one thing; which is to discipline competitors. By discipline, I mean drive the competitors out of business. Markets, left unchecked, drive all competitors out; save one — a monopoly, or a few — oligopoly. Monopoly and oligopoly are always and everywhere the eventual outcome, as well as the mark of a mature industry. Eventually, the industry and the remaining competitors disappear altogether because they or the industry become economically irrelevant Deductively, one can argue this point but it is empirically true. All businesses either die/go bankrupt, or are consumed/acquired by other businesses. Take a look at the Dow Index companies for each decade since 1900 and see what I mean.

    The reason that Google will not face competition in the general search market is because market discipline has driven all of the competitors out of the business of general search. General Search is now a mature industry. After a mere decade, you can put a fork in it, it’s completely done. No mature market has ever seen new entrants successfully compete without government intervention(Take Airbus as an example in the civil aircraft industry).

    Going forward Google will become increasingly irrelevant, unless Google can effectively provide products for a search market that becomes more segmented and more diversified. I don’t know how long it will take, but it will happen. The diversification of search is already happening.

    Presently, Google is hamstrung by a mentality and a desire to dominate general search. There is nothing left to dominate. Personally, I believe, Google is too big and too institutionally calcified to shift gears and do anything else efficiently or effectively.

    On a side note, I find the assertion that trade secrets are an effective way to stifle your competitors silly. It is both inductively and deductively untrue. No company, anywhere, would prefer keeping a trade secret to a patent. Everyone, everywhere files for a patent if they can. I would be very interested in knowing the name of one company that prefers a trade secret to a patent. Furthermore, trade secrets, make no sense since they are impossible to keep. Trade secrets can be lost or stolen and since they lack the protection of a patent, you have no legal recourse to get them back.

  2. Jardinero, I almost stopped reading after you said “Every business in the world faces those same seven roadblocks.” Every business in the world is operating a platform for a 2-sided market?

    Also, you say “Everyone, everywhere files for a patent if they can.” This flies in the face of virtually everything on trade secrecy I’ve read. There are many industries where a TS is preferred to a patent because of its potentially indefinite duration.

  3. I don’t know of any businesses that prefer a trade secret to a patent. That’s not the same as saying that there aren’t businesses that keep trade secrets in lieu of a patent.

    If Google keeps trade secrets in lieu of patents; that’s a weakness. As employees leave to start their own companies or work for others, those secrets leave with them. Businesses that keep a process secret do so, mostly, to hide the fact that they stole the secret in the first place; oftentimes from a competitor’s former employees.

    Again, I would be very interested in knowing the name of any business that openly prefers trade secrets to patent protected processes or products.

    I didn’t address two sided markets. I said that all market participants in any market face hurdles and all will ultimately fail. This is historically/empirically true. It is also empirically true that all industries and markets tend toward monopoly over time and then stay that way. My statement gave merely the what, not the why.

    I think two sided market theory explains why, in certain markets, monopolists or oligopolists stay monopolists or oligopolists. But, it doesn’t really say how they got to be monopolies or oligopolies in the first place. How do you get to the critical mass where the network effects work to your benefit?

    Very few markets start off two sided unless the government sets them up that way, like telephone or broadcasting. The network effects don’t happen until after hegemony over a market has been achieved. What happens between the birth of a market and that hegemony has never been explained to my satisfaction. My own theory is that luck and pluck plays the largest role. Of course, in academia, chalking causality up to chance is never an adequate explanation unless you are an evolutionary biologist or cosmologist.

  4. I thought about the two sided market thing some more. The dynamic can work against a business as well as with it. Here’s what I came up with in the particular case of Google.

    In the case of Google, it provides a platform for general search. The platform is used by two groups, eyeballs and advertisers. Currently, Google absorbs all the costs for the eyeballs and charges the advertisers to use the platform. That works great for Google as long as general search is the only game in town. That is the case today.

    But what about market fragmentation and substitution effects? Is general search the be all and end all for the rest of human history? I doubt it. Is Google going to fragment the market itself, personalized search not withstanding. I doubt it. People will seek out and the market will provide other types of search. The market for general search won’t disappear but it will shrink. What happens to Google if they control one hundred percent of a shrinking two sided market. The cost per eyeball increases. At the same time, the value of Google to advertisers decreases. So Google will find itself in a position where its cost per unit(eyeballs) increase as its revenue(ads) decreases; a very bad position to be in.

    This situation does not require a killer app to happen. It just requires the market to become more fragmented and for eyeballs to spend more time in the fragments and less time in general search.

    This is not unprecedented. ABC, NBC, and CBS owned all the television eyeballs once and they sufferred horribly with the market fragmentation which Cable TV created.

    I still concur that Google will never be bested in the general search market. I think it will matter less and less as the market for general search matters less and less.

  5. I’m still unconvinced. Google’s various products are popular almost in direct proportion to how good they are. Google Search is outstanding; Gmail is excellent, Maps and Docs are very good; Google Video was lackluster; Knol is teh suck. And guess what? Google Search is dominant in its category; Gmail is first among equals; Maps and Docs are doing very well in a crowded pack; Google had to buy out YouTube; Wikipedia is kicking Knol in the articles.

    You make lists of factors that might explain why Google’s search success is illegitimate, phrase them in unfalsifiable ways, and conclude that it’s “extremely difficult” for any competitor to dislodge Google. So it is, as the last decade has shown, but you haven’t given a good reason to budge from the default hypothesis that Google is beating its competitors because it offers better search.

  6. James, Isn’t it possible that Google’s success is overdetermined–that it is both “better at search” and has other non-quality advantages? Or, might one argue that the various examples of network effects, network power, and legal advantages I describe above are important causes of why it is better?

    You say it “had to buy out YouTube,” which makes me think I need to add an eighth reason: the ability and propensity of the dominant search engine to buy up competitors. I think the AltSearchEngine blog has documented that process.

Comments are closed.