The Artima Developer Community
Sponsored Link

Computing Thoughts
Should Microsoft Buy Yahoo?
by Bruce Eckel
February 4, 2008
I read somewhere that over half of acquisitions fail. I think the percentage might be much higher than half.


I have certainly seen acquisitions that looked like obvious bad moves, and later turned out to be just that. But the corporate leaders had managed to talk themselves into thinking it was a good idea.

Part of the problem is that a publicly-held corporation must put its cash to work, or, I suppose, pay it out as dividends. Buying small, innovative companies that have developed something that the corporation hasn't been able to -- that makes sense. Google, in fact, has turned this into a business model for small companies who develop an idea precisely so that Google might buy them.

Buying big, established companies -- those who are themselves capable of buying other companies -- seems to be where most of the bad ideas come from. Here are three that I personally paid attention to.

Borland buys Ashton-Tate

Ashton-Tate invented the DBase database system and was very popular for awhile, then began sliding. Borland developed a much-improved clone of DBase and was doing very well with it, gaining customers left and right from Ashton-Tate. Ashton-Tate continued sliding, and everyone could see it was only a matter of time before they went on life support. All Borland had to do was continue on their current path and they'd win.

Instead, they bought Ashton-Tate for a billion dollars. They threw away a billion dollars on a failed company when they were going to get all their customers anyway. A few years later Borland really could have used that billion dollars. These days, Borland is in a different business altogether.

HP buys Compaq

Compaq had not done anything interesting or innovative for years. It was a failing company, even if it appeared to be only slowly declining. In the computer industry, it's "lead or fail."

What did HP think they were getting, and what did they think they could do with it? HP already made computers, at least somewhat better than Compaq did. Was the plan to manage Compaq back from the brink? I can never figure out what upper management is thinking when they make decisions like this, especially with all the history of all the other acquisitions that have failed. Perhaps they're thinking "this time it will be different."

One argument that always surprises me is "we're buying their market." Why not, I think, spend all that money on making your own product better. Then their market will just come over to you.

Perhaps buying a competitor is the act that shows that the "management Peter Principle" has been reached. All the tricks in the management bag have played out, and we can't manage ourselves into creating better products. So we fool ourselves into believing that we can "buy the market" since we can't create a better mousetrap.

Microsoft might buy Yahoo

I don't think Google is perfect, but I use their stuff. Yahoo is still making little innovations here and there -- typically under marketplace duress, rather than leading because they've come up with the next great thing. But the only time I go to Yahoo is for Yahoo Movies, to find out when and where movies are playing. Yahoo Finance is supposed to be pretty good, too.

Yahoo itself admits that we shouldn't expect a turnaround before 2009, and that sounds optimistic to me because one of the fundamental tenets of their turnaround strategy is to "open Yahoo's technology infrastructure to third-party programmers and publishers." Can this be done after the infrastructure of a company has been built? I doubt it. It's a bit like Microsoft trying to add "security" on top of Windows. Windows wasn't designed with networking and security in mind, and security in particular is not something that can be added as an afterthought. Security must be part of the fundamental architecture, not an optional add-on feature (see Unix for comparison).

Opening the technology infrastructure is only objective #3. Yahoo's first two objectives are to be a starting point for the most consumers on the Web and to make the company a top choice for marketers seeking to place ads on sites across the Web. Basically, to reverse the fortunes of Yahoo and Google. How likely is this to happen? For that matter, how could it happen? What is it that's going to make you go to Yahoo rather than Google when you want to search, or to place ads with Yahoo rather than with Google Adwords?

Here are two places where Yahoo could get the jump on Google, although I think Google is already working on both of these. First, charging for ads based on actual sales generated by the ads, rather than just click-throughs. Think about it. If you wanted to advertise something and you knew you'd only have to pay for the ad when the ad actually sold something, wouldn't you advertise a lot more? It would be completely risk-free. Yes, the cost of an ad-based sale would certainly be higher, but would that really make a difference if your choice was between selling something and not selling something? Personally, I think it would be a no-brainer.

The other thing that might bring me to Yahoo is the store system -- and I consider any changes in this area doubly unlikely because Yahoo already has an entrenched (and expensive) store system. Plus Google is apparently working towards what I'm about to describe.

What we need is a very easy way to know that a transaction has taken place, so it becomes trivial to program against the transaction-processing system. Right now it's far, far too hard. I tried it with Paypal, and that turned me off. The coupling is much too high. What I really want to do is know when a transaction has happened, and what that transaction is, and that's all. My program can then do whatever it needs to do without becoming entangled in the transaction-processing API. If this process becomes easy then it will cause a big increase in online transactions. If Yahoo did it before Google then I might try it. The downside is that it's going to cost me more because of Yahoo transaction fees, and with Google Checkout all I have to do is buy some adwords each month and I don't pay transaction fees. So I might change to Yahoo for awhile, but as soon as Google solves the problem and becomes the cheap solution then I'll probably just switch to Google.

And that's really the point. Maybe Yahoo has some fraction of the market now, and maybe they will recover in 2009, but right now they're kind of a failing company. What is a hostile takeover by Microsoft going to do for that?

Is Microsoft going to manage them back to health? It seems unlikely that the "maybe 2009" recovery will happen sooner after a hostile takeover.

And the really big question: Microsoft already has an Internet division: MSN. But they haven't been able to make that go anywhere. Yahoo made a good start at the beginning of the Web revolution, but then got stuck and faltered. Why on earth would combining the talents of Microsoft MSN and Yahoo produce anything other than more stumbling around in the dark? Which company would lead this recovery? At least Yahoo managed to make a strong start on the Web before driving into the weeds. But no, it would be the company that takes over, Microsoft, who brought us both MSN and Windows Vista.

How does combining a failing company (Yahoo) and a failed division of Microsoft (MSN) produce a successful division of Microsoft?

The one thing that the two companies have in common is the very thing that will make them both fail at this Web thing. They both believe that it's about centralization. The companies are built around that philosophy. Yahoo tried to create a central place where people would go on the Internet, which worked in the early wild-west days of the web -- Yahoo was the best thing out there for awhile -- and it's what caused Yahoo to slide as the Web became more decentralized. Yahoo's centralized services are no longer as good as their decentralized competitors. Now we have BBC for news, YouTube for video, etc. And Microsoft, of course, is the "Cathedral" in The Cathedral and the Bazaar.

One Trick Per Company

I think each company rises to prominence, and then becomes ossified, based on a single Trick that they can do that other companies can't. For awhile, this Trick keeps them out in front, nimbly anticipating anything the competition tries to do. Everyone in the company starts feeling like they've got it all figured out. Managers and vice-presidents are chosen based on their dedication to and ability to mimic the Trick.

By this time, the Trick has become the only acceptable way of thinking within the company, so business innovation has become impossible (but if the Trick is a good one, it can serve the company in its "stable" mode for a long time).

At this point, throughout management, people have set up fiefdoms based on their perception of the Trick (which is usually distorted) and will fight against any changes in the structure of the company, because it threatens the security of their fiefdom. The company plays into this -- the company needs a way to keep managers in line so it usually measures them by how well they believe in the Trick. At the same time, it can't quite let them think they are good enough at the Trick that they could leave the company and go off and do it themselves (ironically, those are the very people that would be the best managers for the company).

At this point the company has become ossified. They still believe in the mythology of their youth, but the world has changed around them, either implicitly or explicitly adapting to their Trick, so the Trick no longer works. The company can't believe that it doesn't, so it just repeats the trick with more force. One thing I haven't seen is a company that has been able to realize that it's old trick no longer works, and come up with a new one. The entire company is built upon belief in this single trick, and it resists mightily any change in that belief system.

Before the computer age, companies went through these phases so slowly that very few Tricks worked, so most companies of a certain type tended to look like each other, and it became easy to say that "this is the way business works." Computers are about speed, and Tricks that never had a chance in the old world could work spectacularly well. Thus we've seen, in a short time, many companies gain success based on their Trick, and just as quickly go through ossification and eventually fail. As the examples pile up it becomes harder to pretend they are special cases.

Here are the tricks of the companies we're interested in:

Microsoft: The Quick Move. Everyone knows the story. Microsoft bought MS-DOS from a little company and then turned around and sold it to IBM, cleverly keeping the rights. As time passes, they improve the product. The ideas always come from somewhere else (Office and .NET are two examples). First they get into the market, then over time they improve the product.

Yahoo: The Only Playground You'll Ever Need. When they started, there was no place to "go" on the Internet. Yahoo provided the best place (at the time) to "go", which was a great service, but then, instead of continuing to think in terms of serving the visitor, there was a subtle change: Yahoo wanted to keep the visitor within Yahoo. So the goal went from "service" to "control." This looked better to shareholders and to sellers than it did to visitors. And since there was no real fence to force visitors to stay at Yahoo, they began leaking away. But within Yahoo, the shift was so subtle that they don't realize what's happened.

Google: Free Services to Sell Ads. The reason this is deeply clever is that the money stream, which is where managers set up their fiefdoms, is completely decoupled from the products that attract and serve the consumers. Which means that the product-creation part of the company can remain un-ossified for a much longer time. Even as Google has become big, the products continue to be creative and customer-oriented, which suggests the lack of involvement by managers and marketing people.

On top of that, the business model -- the advertiser only pays for a click-through -- is so simple that there is less leverage in setting up a fiefdom, because there's much less possibility to abstract or obscure information, and thus less potential for power. The formulas behind the business model are controlled by technologists.

The one thing that will probably keep Google out in front no matter what Microsoft and Yahoo do is Google's freedom to create end-user services. Because Google's business model is in selling ads, everything else they do can be done in the name of the consumer. This minimizes the need for and influence of marketing people. Which is not to say that technologists always create great things, but it seems that whenever marketing people get involved, their motivations and perspectives tend to make things that much worse. Look at Microsoft Office. Even with Microsoft's vaunted "Usability Lab" they managed to come out with new versions of Office where they moved features around and basically hid them. Now, instead of being afraid that the next version of Office will be more broken and buggy than the one you're using, you're afraid that you will have to go up an entirely new learning curve.

Microsoft's goal is to own the desktop, and the desktop is slipping away from them. Yahoo's goal is to keep people at Yahoo, and people leave the moment they find a better site. Both of these goals are primarily designed to serve the shareholder.

Google has shareholders too, but those shareholders have been specifically told that revenue all comes from ads. "Pay no attention to all those products and services, they are only indirectly associated with revenue. Eyeballs, you know."

I certainly don't believe Google's approach is flawless. For one thing, I think Google has gotten far too mired in a single technology -- Javascript on the Browser -- even as Google themselves have demonstrated the implacable problems with that technology.

However, Google's Trick looks like it's going to work for awhile, far better than Microsoft's or Yahoo's. And the latter two companies seem to be fully in the throes of denial, as they repeatedly throw themselves against the wall, declaring "Trick ... works ... must ... try ... harder!"

Microsoft would be better off buying a rocket and boosting their 44 billion dollars into space. They would lose less money that way, because the 44 billion will just be the start of the costs. All the restructuring and busy work and vision statements (hey, for 3.75 million dollars I'll come up with your new vision statement!) and opportunity costs while everyone sticks to their guns and nothing changes will add a lot more to the cost. Ultimately they will discover that combining two failures will not produce a success.

As long as you're mindlessly flinging money around, why not do something truly bold? Take a tenth of what it would cost to buy Yahoo, or less: "only" a billion, and start a skunkworks project. Keep the management structure away from it, let it actually innovate something (note that declaring that you are "innovative" doesn't make you so; in the words of the Bard, "Methinks thou dost protest too much"). Keep the management structure very shallow and fiercely dedicated to supporting the project instead of building status. You probably need to get managers who've never worked at Microsoft.

Heck, it's only a fraction of what you're talking about throwing away on Yahoo -- why not create a few of these projects. Who knows what you might come up with? But if Microsoft buys Yahoo, it will only produce yet another case study for business-management textbooks on why takeovers usually fail.

Here's another take on what Microsoft should do instead with its 44 billion.

Talk Back!

Have an opinion? Readers have already posted 19 comments about this weblog entry. Why not add yours?

RSS Feed

If you'd like to be notified whenever Bruce Eckel adds a new entry to his weblog, subscribe to his RSS feed.

About the Blogger

Bruce Eckel ( provides development assistance in Python with user interfaces in Flex. He is the author of Thinking in Java (Prentice-Hall, 1998, 2nd Edition, 2000, 3rd Edition, 2003, 4th Edition, 2005), the Hands-On Java Seminar CD ROM (available on the Web site), Thinking in C++ (PH 1995; 2nd edition 2000, Volume 2 with Chuck Allison, 2003), C++ Inside & Out (Osborne/McGraw-Hill 1993), among others. He's given hundreds of presentations throughout the world, published over 150 articles in numerous magazines, was a founding member of the ANSI/ISO C++ committee and speaks regularly at conferences.

This weblog entry is Copyright © 2008 Bruce Eckel. All rights reserved.

Sponsored Links


Copyright © 1996-2019 Artima, Inc. All Rights Reserved. - Privacy Policy - Terms of Use