When It Wasn't Stuffing Cars, EMC Was Doing Real Business
When it wasn’t stuffing a Mini Cooper full of dancers storage concern EMC actually did launch a huge batch of new products yesterday.
The headliner was VNXe, its first low-end offering, priced at less than $10,000 and aimed at small and medium businesses, a segment where Dell used to resell EMC equipment. In another bit of product-launch theater, EMC had a fourth-grade boy onstage to demonstrate that the box–which in this case was mounted on the back of another Mini Cooper–could be managed and configured from an iPad.
I caught up with EMC Chief Marketing Officer Jeremy Burton to talk about it and the 40 other products EMC launched yesterday.
NewEnterprise: So 41 products all at once?
Burton: I’ve never been in a situation where the release dates of so many products aligned. We realized we might as well do them all at the beginning of the year. Internally we called it the “mega-launch.”
What kind of opportunities do you see in that lower-end market. This was your first entry into that market.
We estimate maybe a $4 billion opportunity there. We don’t have much of it now, call it zero. We’ve never really built a product that’s tailor made for that market. And for a product like that, you can’t just build it–you have to build it in a way that the channel can make money on and create customer satisfaction. We’ve got several partners who will take this product to market. We’ve committed $20 million there to generate demand and bootstrap the whole thing.
Tell me about this go-to-market effort.
Traditionally EMC has been led by direct sales. We have a sales force and they call on the customer directly. With a products that sells for $9,000 or $10,000 you can’t afford to sell that in the same way. We have to create pull for the product with our partners. You have to get the customers calling to ask for the product. It’s a little bit of everything. There’s advertising, there’s direct campaigns. Anything to get the phones to ring. To get the reps at the events jazzed up we’ve leased a fleet of 21 Mini Coopers. We’ll be doing 108 partner events around the world.
So who do you see as a typical customer for this?
Before I worked for EMC I ran a software company that had about 700 or 800 people. We had about 20 guys in the IT department. We didn’t have a lot of specialists, we had a lot of generalists there. So I’d say any company that’s at less than $25 million in annual sales is a perfect candidate. They’re not going to have the high-end skills to deal with the complexity of the high-end arrays. But they’ll have VMWare, they’ll have exchange environments, they have file shares, and they’ll want to get going quickly.
It’s funny I should be talking to you today. I just published a Q&A with Ping Li of Accel Partners. We got to talking about the storage needs of companies moving to the cloud, particularly around their database environments, and he said the trend is toward running open-source things like Hadoop on commodity hardware. He said he’s not seeing a lot of EMC gear at Google or Facebook or many of the other Web companies. There’s a lot of people who are seeing both a trend and an opportunity around that. What do you see?
If you’re Google and you’ve got your own team of rocket scientists who can build your own file system and kernel and download modules from the Internet every day, you don’t need it. But if you’re Pfizer, you probably have a lot of rocket scientists, but you probably don’t want them working on reconfiguring kernels, you probably want them working on discovering new drugs. And so, picking the techiest of the tech companies and saying they don’t use our stuff, yeah those are companies with the smartest tech guys on the planet. The problem is they’re not in all the Fortune 500 companies in the world, and in fact I’d argue they’re in almost none of them.
So if you want to have that scaled-out commodity storage and you want to manage big data, and you don’t want to hire 1,000 rocket scientists to do it, we can sell it to you. It won’t be true commodity hardware, but then you won’t have to hire so many people to manage it. That to me is kind of the rub. EBay is a big name on the Web, and it uses our Object Storage infrastructure. Could they have built it themselves? Probably. But there’s a little intellectual snobbery inside these companies. They say, “We don’t want to buy your stuff because we’re smarter than you.” Those are the edge cases. If we just get the rest we’re happy.
Let’s talk about the broader picture in IT spending. What are you hearing from your largest customers about their intent to spend this year?
2010 was a decent year. Going into 2010 folks said they thought their spending would increase two to three percent. They probably ended up with three to four percent. Looking out into this year, people seem a little more optimistic. But even still I think it’s in the three to five percent range. One thing we saw in 2009 is that folks didn’t buy much storage capacity last year and instead tried to use what they had. Going into 2010 there were signs of recovery and people started to spend again, and we see that continuing into 2011. One reason for the Isilon acquisition is that we do see a trend toward spending into different areas of the business.
At another level I think I agree with you and with Ping that certain companies will move to Hadoop for a certain class of application and we’ve got a pretty strong relationship between our Greenplum division and Hadoop. What a lot of people want to do is analyze traditional enterprise data in conjunction with something else. What Greenplum has tried to do is bridge the gap between Hadoop and the more traditional storage infrastructure. Hadoop is not going away, and its something that we fully intend to work with.