Related topics

IBM sniffs around Fortinet

But Big Blue is late to the converged infrastructure game

IBM last week polished off its acquisition of privately held Ethernet switch maker Blade Network Technologies for a rumored $400m. And just after finishing that networking meal, Big Blue appears still hungry for networking products as it is apparently sniffing around network security appliance maker Fortinet to do a possible takeover.

This once again shows that IBM is a big investor in "research and development", as the company's top brass is constantly reminding everyone. What IBM didn't add is that it was other people's research and development. . .

According to a report at Bloomberg, two people who are familiar with the deal say the two companies have been talking for two months and say that Fortinet, which went public last November, has hired Morgan Stanley to walk it through a possible acquisition by Big Blue.

IBM clearly wasn't thinking it had to directly compete against Cisco Systems and Hewlett-Packard in the networking racket as systems and networking gear are increasingly getting converged and virtualised. But the sleepy IT giant appears to be finally waking up to the fact that maybe it was really stupid to abdicate the networking market to Cisco more than a decade ago. And the pocket change that IBM has in its coffers because it spends way too much on share buybacks to financially engineer its earnings per share growth means it cannot acquire Juniper Networks, Brocade Communications, QLogic, or other big networking players to make a stand against HP and Cisco in the converged infrastructure future.

If IBM had been thinking ahead, it would have bought Fortinet ahead of its initial public offering a year ago, when it raked in $156m. And certainly before the network security appliance maker had posted its third quarter numbers. In the quarter ending in September, Fortinet's revenues rose 29 per cent to just under $90m and net income more than doubled to $14m. Fortinet floated its shares at $12.50 a pop a year ago, and this morning as El Reg goes to press, the company's stock is kissing $38 a pop, giving it a market capitalisation of $2.14bn. Ahead of its Q3 numbers, Fortinet's stock was trading in the range of $25 a share and has been rising steadily from the high teens since summer, based on improving financials and takeover speculation.

The acquisition of Fortinet is pricey compared to the benefits that IBM might gain, but if it wants to be a network player in its own right, it has to start somewhere. Fortinet had $292.3m in cash and short-term investments as the quarter ended, and is on track to do around $320m in sales, maybe $40m in net income for all of 2010 – according to El Reg's back-of-the-cocktail-napkin methodology. Paying something on the order of $2.5bn for such a company seems insane, but so does spending $50bn in share buybacks over the next five years but only $20bn on acquisitions, as is the IBM plan. (IBM could buy Fortinet with shares, of course, and with Big Blue's stock at an all time high, that may be the smart thing to do.)

IBM's managers have it exactly backwards, and to put it bluntly, they are taking the easy EPS engineering way out. What IBM should have done is gone on an acquisition spree two years ago during the downturn to buy companies that would give it real earnings in the future, not fake up some EPS growth by burning so many bales of cash.

And now, IBM is going to pay a hefty premium for each and every networking acquisition it makes and lose a lot of leverage with its partnerships. IBM needs networking partners more than they need it.

While I am thinking about it, Big Blue needs its own virtualisation and hypervisor stack for x64 platforms in this converged and heavily virtualised future, and that means buying either Citrix Systems or Red Hat - neither of which IBM can afford, either, because of its EPS crack habit. ®

Sponsored: 10 ways wire data helps conquer IT complexity