Jonathan pointed out that “ 100% of the products have been revamped”
Its a fair point – Sun really has gone through a serious and far-reaching infrastructure refresh. Three years ago its server portfolio probably deserved to be called “ageing”. IBM and Dell mopped up accounts accordingly.
Not any more. On the hardware side Sun is a hot box company again, or at least that’s what customers, developers, regional SIs are telling me. Sun is always going to be at its best when it has good kit to sell. That is the company’s DNA. When I first heard about the Niagara chip architecture back then it sounded a bit like wishful thinking on Sun’s part. But Sun has delivered.
Participation drives conversations with customers and prospects, but it is performant, slickly built hardware and software infrastructure that will turn those conversations into deployment and purchasing decisions.
Jonathan’s priorities – how to grow value. That is one thing I do like about Sun’s managers – they know firing thousands of people is not the key to becoming a more effective. headcount reduction is not a panacea.Value is what counts. Headcount should be a function of value, not the other way around.
So what about the new “HP and Sun to partner ” mischief theme? Jonathan claimed that Solaris 10 has seen 4m downloads and “the majority of our downloads are onto HP.”
Unlike JBoss head honcho Marc Fleury, who says paid contracts with customers are more important than broad distribution of the platform, Jonathan says he is happy to see his platform downloaded on any platform for free, because these downloads might turn into revenue in the future.
So what about customer evangelists? I liked Jonathan’s high risk strategy when discussing Wells Fargo – which was to talk about the extremely negative perceptions facing Sun in many accounts, and how those perceptions can be turned around.
“We visited Wells Fargo and the CIO said: you’re gone. You’re sunsetting. Your OS is dead. We think open source is the way to go.” Apparently the Wells Fargo guy even asked the Sun sales rep- why did you join Sun?
It doesn’t get more ouch than that. But…
“A few months ago things had changed”. Comparing legacy with the future. Systems that just ran. Jonathan claimed that Sun’s acquisition of STK was also significant in changing the perception. “ we were at the center of the business.”
Jonathan also talked about how declarative living was the basis for how Sun charges its Java Enterprise System customers. That is, public companies make statements about the number of employees they have. Sun uses those statements to drive licensing discussions, rather than traditional fear-based audit. This is a transparent, non-confrontational approach, that allows for organisations to scale software purchasing up or down pretty easily, with no per processor actuary rocket science required.
“I don’t think we should be in the business of auditing our customers.” Amen brother.
What about other customer stories? Marc Andreessen and Ning. Ning is a situated software company.
It was funny to see Andreessen – the Net leviathan – fumbling around with a bunch of calculations on the back of envelope… I am hoping the numbers will get blogged or posted so i can point to them. But basically Ning has chosen Solaris on Sun hardware because it calculates Sun hammers Dell/Linux on TCO over a three year period…
Fully loaded server cost over 36 months:
- For white box intel + linux $10,200
- Solaris 10 $5700
- Solaris 10 and AMD $4760
There were also, according to Andreessen, signfificant advantages on power consumption and rack space utilisation. Commercial grade linux support, he said – “you pay through the nose for that kind of support”
So is Sun’s strategy working? One very interesting stat – US bookings are apparently 10% year on year.
For enterprise customers, even those of a Linux disposition, I would say try talkig to Sun. Its not as if they make it hard to get hold of their technology.
Participation isn’t a business strategy in itself though, and this year is completely critical for Sun. If it can’t turn technical leadership into sales this year, then we’re almost certainly going to see some shareholder-driven carnage. The investment bankers in the audience have been selling Sun short for five years now. They want to see sales, not narratives, no matter how compelling the stories or associated technology might be.
/pd says:
February 2, 2006 at 12:44 am
I think that SUNW will soon move upward into the ‘outperform’ status. given the Vertical segment, they are head on head with the others. Thier EPS is in red, so off course its an edging markert
Kevin McIsaac says:
February 2, 2006 at 3:15 am
Not only do Sun need to turn technical leadership (which for the Opteron systems is probably another year at most) into sales, they need to turn sales into profits.
Even if Sun can get volume on the Opteron can they generate sufficient profits from the thin margins to offset the decline in revenue of their SPARC systems that have huge margins?
Even if Sun can do all this, they then need to a better job with manufacturing and supply chain management than DELL? If not DELL will just execute its old strategy of being a technology follower with a lower cost base and slowly undercut SUN.
James Governor says:
February 2, 2006 at 8:21 am
hello kevin – how are you?
I don’t believe you can sell servers without offering decent customer service, and sustain a business in the enterprise for the long term. Dell has its own issues to address. I am talking to more pissed off Dell customers lately than those annoyed by Sun.
I also tend to think a lot of the initial Sun carnage is done. Check out the T-1 boxes too – opteron is not the only game in town.
Kevin McIsaac says:
February 5, 2006 at 4:57 am
Hello James – I’m well
Yes, Dell have their problems but not with revenue, revenue growth or profit. Using full year, 05 figure.
Revenue Profit Revenue Growth
Dell $49B $3B 18%
SUN $11B ($0.11B) 0%
As for you comment I don’t believe you can sell servers without offering decent customer service, M/F vendors said that in response to UNIX/RISC in the 80’s and look what happened to them.
I’m sure you would remember Alpha, the hot CPU from DEC in the early 90’s. Not only did it run VAX VMS (in many ways still a better OS than Solaris, e.g. real clustering) but also UNIX and NT. Even though it was billed as a very, very fast chip that would crush Intel, it did not nor did it stop DEC from flaming out.
The T-1 boxes are interesting, and Solaris is a great O/S, however so too was the M/F, which was largely surpassed by commodity, volume servers in the ’90s. I suggest this is another case where history repeats itself and “good enough” (i.e. x86) wins out and the T-1 will be Suns last gasp.
Last couple of times I spoke to Scott he sounded just like IBM in 80’s, listing all the reasons why the M/F would win and UNIX/RISC would not. I find it rather ironic.