An EMC Critic Enters the Fold

At Hewlett-Packard, Mark Lewis came down hard on the storage giant. Now that he has joined the outfit, he explains the change of heart

The storage market has grown increasingly competitive over the past two years. Even sector giant EMC has struggled. To stay on top, it has brought in several new executives to help with the turnaround. Mark Lewis, 40, who joined as executive vice-president for new ventures and chief technology officer in July, is one of the more controversial hires. Not only did Lewis come from EMC's toughest competitor -- Hewlett-Packard/Compaq, where he had been head of Compaq's storage group -- but he has also been one of EMC's most vocal critics.

Lewis roasted EMC in the past for providing proprietary software solutions that didn't work well with other companies' hardware. He also critiqued it for pushing big, expensive storage-system installations when most customers wanted a modular approach -- which he had pioneered at Compaq.

Now, Lewis says, over the past 18 months EMC has begun to change its stripes. Which is why CEO Joe Tucci was able to persuade Lewis to come help EMC develop partnerships with other vendors, improve its midrange modular products, and bulk up its software offerings. It's common for storage executives to move around, but "it's a real coup for EMC to get Lewis," says Alex Mendez, a general partner at Palo Alto, Calif., venture-capital firm Storm Ventures.

Lewis spoke with BusinessWeek Online Associate Editor Amey Stone by phone recently about his reasons for joining EMC and his vision of the future of storage networking. Following is an edited transcript of their conversation:

Q: Why did you decide to leave Hewlett-Packard (HPQ ) in favor of EMC?


A year and a half ago, I would not have joined EMC. I was running the storage division at Compaq and thought it had a much better vision of storage than EMC. If you look at the past 18 months or so, a lot of change has been going on in the storage industry -- and tech players know they can't become stagnant. I spent a lot of time talking to Joe Tucci about EMC, and I came to realize that the company has a new understanding of the changes it needed to make to remain very competitive in market.

Q: What were some of the business decisions EMC made that convinced you the company was changing?


One clear sign was EMC forming a relationship with Dell Computer last November. Dell lacked networked storage technology, and EMC needed a way to reach into smaller markets. The deal allowed EMC technology to be brought in and supplied to a new segment of the market at a competitive price. EMC has also refocused on its Clariion lines, which are midrange in price. We just introduced a new high-performance, modular product co-branded with Dell that's a lot more scalable. That's something the market is demanding now.

Q: What are your goals as chief technology officer?


Our goal is to be first and foremost a technology company that continues to innovate in storage. So we will continue to spend on R&D. We are increasing our storage-software content in the overall solution. We feel it's very important for storage networks to become increasingly automated, and we're making an important push there.

Q: Does that mean EMC is becoming a software company?


Becoming a software company for us doesn't mean getting out of the hardware business. It means that we plan to drive more of our sales through our software. By the end of 2004, we want a sales ratio of 50-30-20. That means 50% of sales will come from hardware, 30% from software, and 20% from services. Today, software is 22% to 23% of sales.

To move more to software, we want to have an independent value proposition. We believe a customer should be able to purchase storage-management solutions without buying EMC hardware. Our announcement in mid-September of new open-software initiatives...was all about this. Now we can deliver most storage-management software in a heterogeneous environment [where the system includes hardware from other vendors].

Q: What will the storage area network look like five years from now? Will fiber channel (a networking protocol now used for storage area networks) remain dominant?


I believe fiber channel is here to stay, since it has inherent benefits over other networking protocols. We will do what our customers ask -- and the market could change.

I think we will continue to see a decoupling of servers and storage. Now, 70% of storage is still sold directly attached to servers. I expect that will drop to under 30% in next few years. By 2005, you simply won't be able to buy a server that has dedicated storage. Eventually, customers won't be buying servers with dedicated storage anymore. The storage will be a central resource on the network that they'll access.

Storage will even move in this direction for home users. A computer in the home will still have a disk drive, but storage there will be almost like cash in your wallet -- a staging place for your information, not the permanent repository for your data. Over time, I believe we'll see the development of more central-storage repositories where your data is secure and protected, like at a bank.

Q: EMC just announced it's buying Prisa Networks for $20 million. Can you tell me more about your acquisition strategy?


One thing I've learned is that you don't discuss the details of an acquisition strategy. That's like telling real estate brokers where you're going to buy your next house. I can say that, as a corporation, our emphasis is on software and on increasing our software sales, and that we will continue to do acquisitions.

Q: EMC has quite a bit of cash, doesn't it?


Our liquid assets total $5.5 billion, and we've been generating cash, even through the last three or four quarters. You have to have a healthy balance sheet during times like these. That's a big part of being successful and surviving.

Q: Do you see any signs that demand for storage equipment is improving?


We would continue to characterize the market as tough, and we don't have any insights that it will get any less tough in the near future. It has turned into a market where customers really want to see their basic needs met and will do new projects as long as they have very, very rapid ROI [return on investment]. That needs to happen inside of a year these days. It used to be that a return on a sale could take two to three years.

Our focus now is on achieving growth by taking share in the storage market. IDC's latest numbers prove that we were able to do it in the second quarter.

Q: Is EMC increasingly competing on price?


Not at all. Obviously there were a lot of pricing pressures last year in the industry. Nowadays pricing has stabilized, and we're not fighting it out on every last dime. The sale is much more about spending a lot of time with customers and showing how certain pieces of software will save them money.

Throwing technology at a problem is not always the answer. It really is about being very wise in the use of technology and understanding when it's good to move to a new technology and when, maybe, it isn't. Today, buyers are more skeptical of visions and wild claims than they used to be. They're looking for proof.

Q: Does it seem like there's more confusion among buyers or less than three years ago?


You do tend to see in tough times folks looking back to partners that have financial stability, that are market leaders. In periods of high growth, there's a lot of new technology hitting the market very rapidly. Customers are willing to take more chances in trying out new things. But in times like these, obviously what we want to do at EMC is be a very trusted partner.

    Before it's here, it's on the Bloomberg Terminal.