3 ms·
The logic is that Google, whose Internet cash cow probably has more milk left in it than Microsoft's Office/OS cash cow, is already more profitable per employee
by sutro 18y ago
The logic is that Google, whose Internet cash cow probably has more milk left in it than Microsoft's Office/OS cash cow, is already more profitable per employee than Microsoft, so its incursion into the Office space is not a serious quest for higher margins but rather a tactical move made primarily to put Microsoft on the defensive. It makes perfect sense.
Not knowing anything about you or your startup, I will nevertheless go out on a limb and guess that Sridhar Vembu's English is better than your Hindi, and that his fully bootstrapped startup's $40 million in revenues and $12 million in profits [1] puts it in better financial condition than your business.
Try to open your mind and learn something from the man behind one of the most successful fully bootstrapped tech businesses in the game today. Most entrepreneurs wouldn't consider going up against both Microsoft and Google with no investment backing, but Vembu saw an opportunity, took a big risk, and is now in a position to reap sizable and well-deserved rewards.
[1] http://www.forbes.com/technology/2008/02/22/mitra-zoho-india-tech-inter-cx_sm_0222mitra.html http://www.forbes.com/technology/2008/02/22/mitra-zoho-india...
- mattmaroon 18y agoIt doesn't make sense. Office is a very high margin business. Surely Google won't achieve the profit margins MS has even if they take it over, but it will still be substantial. Revenue per employee is largely an arbitrary number as well, hence I object to its being used as evidence for anything. You could just as well use revenue per unit of cheese consumed in the cafeteria on Tuesdays to prove your point. I didn't write a babbling incoherent article in Hindi, but if I do than that observation would make sense. He, however, wrote one in English. Also, his startup success (or mine) doesn't have anything to do with whether or not that article is logically coherent.
- netcan 18y agoI don't think the argument is about (or entirely about) attractiveness (or lack thereof) of high margin sectors. It's about being able to thrive in that kind of environment. For example, Apple cannot thrive in the same kind of environment as Dell. Apple's corporate culture, habits & history would put it at a disadvantage in a world where margins are thin because it has not evolved or been optimised for that environment. In a world where competitively driving down costs via stock turnaround times & cheap production determine success Dell has the upper hand. Similarly, in a world where revenue per employee is low, a company such as Google might be ill equipped. I am not agreeing or disagreeing with this position (I don't really know enough to). I'm sure you could argue other forces at play might be more significant or that his assumptions about Google's tactics are incorrect. But it's not nonsensical. What I did find unclear was which category he places himself in: SMB software or enterprise. But that's a limitation of the medium (personal blogs). Some background is assumed. I'm willing to take that as it means writers don't have to start from scratch.
- pramodsub 18y ago>> Revenue per employee is largely an arbitrary number as well, hence I object to its being used as evidence for anything. I disagree. It is a pretty good quantitative measure of how your business is setup. How much money a company has obviously affects its decisions in a big way, and it is only logical that how much money a company has to spend per empolyee affects the business decisions it makes. I am not taking a position for/against the rest of the article, but you argument that revenue per employee shouldn't be used as evidence for anything does not pass muster.
- startingup 18y agoAll business is about adding value, so that your output is bigger than your input. The most important input in a tech company is human labor. Google is the most efficient at this game today, as measured by how they transform their human labor into value. Revenue per employee is the best proxy for value-added per employee.