Response to Radicati's Messaging Market Analysis 2005-2009, Aug 4

Microsoft must be pretty pleased with Radicati's latest Exchange Server Market Share Statistics 2005 report. Based on Radicati's methodology and market place analysis, it said that Microsoft's take from Exchange Server licensing in 2004 was $1.45 billion, about $0.45 billion more than Microsoft itself said that it earned from the product. Further, Microsoft said that the first year it earned more than $1 billion from Exchange was July 2003 to June 2004, whereas Radicati paints Microsoft as a liar, saying it has earned more than a billion each year from Exchange since 2001 (see Table 5). Strange ... if Radicati can't get the dollar figures for historical years to match reality, then its forward-looking figures are highly suspect.

I was less than impressed with Radicati's analysis of the Exchange vs. Domino question in 2004, so I approach this latest report with a medium-high degree of doubt. Here's my response:

• Table 2 looks at Worldwide Corporate Mailboxes by Business Size. Radicati says that we should expect a 43% growth rate in mailbox numbers across all segments between now and 2009; that's 35% in small businesses, 26% in medium-sized businesses, 60% in large businesses, and 63% in very large businesses. Per the Scope statement at the beginning of the report, a "mailbox" is one that is used at least once within the preceding three month period. I didn't think there was this much growth left in the market, particularly in the large and very large business segments, so perhaps Radicati is trying to include mail-enabled applications that accept inbound email, eg, a collaborative workspace.


• Radicati has consistently said that customers don't understand IBM's messaging and collaboration strategy. It appears that much of the projected migration away from Notes/Domino and Workplace is predicated this lack of understanding. If (a) the lack of understanding is removed, and (b) the strategy aligns with the value that customers are trying to achieve, then one can conclude the migration won't be so severe. Finally, if, according to Radicati, customers are confused by IBM's strategy, are they equally confused by Microsoft's one? Eg, what role does Groove Virtual Office and the Groove Servers play vs. SharePoint, and will customers will continue to be happy to invest in a growing roster of servers and associated CALs from Microsoft?


• Microsoft has found it difficult to get existing customers of Exchange Server to upgrade to later versions, with many organizations still running Exchange 5.5. Perhaps it has finally put the worst of its pull-based upgrade methodology behind it, and customers have cheque books lying open to Seattle.


• I'm surprised that Radicati isn't being more bullish on pushing for adoption of Mirapoint, given that it published a report in July 2004 that demonstrated that Mirapoint Server is four times less costly to run than Exchange Server 2003. That is, the three-year total cost of ownership for Exchange 2003 is $317.21 per user, but only $87.54 for Mirapoint. One would have expected that Radicati would proactively advocate product adoption based on its total cost of ownership analysis.

In conclusion, don't put too much faith in this report for your business planning. Take it as one input, but don't base your messaging strategy on it. With all of the changes coming down the pike on the collaboration front, there's much more to consider that just whether or not to use Exchange Server. I think IBM has a strong ongoing story with Notes/Domino and Workplace, and customers have derived value from those over many years. I don't see that stopping. Microsoft has definitely seen some level of market success with Exchange, but the jury is still out on whether (a) customers will continue to upgrade to its new-new stuff, and (b) what its collaboration offerings will look like in the Office 12 wave and beyond. You know my position ...

Comments

[Simon Barratt]

Good summary Michael.

It amazes me that Radicati are still operational.

The way you correct their MS numbers, by using the actual MS published numbers is highly comical!

[Ed Brill]

it can't be coincidence that this report has faded from view on the microsoft.com/exchange website. It can still be found, but it's no longer highlighted on the top level page, and is only visible on the IE version of the 'competitive comparisons' page.

[Michael Burry]

Hi Michael,

Have just been reading your response and one thought that has occurred to me is that it depends on how one defines a mailbox. If, for example, a mailbox is defined as a repository associated with a unique email address, then a public folder would qualify. Maybe they expect the growth not so much in the number individuals but from growth in use of such entities as public folders. This would be consistent with your suggestion of the growth in collaborative workspaces (CW). In MS speak the CW would be the public folder. Certainly we have clients that use public folders in such a way and I wouldn’t call them nor I leading edge!!!

Who knows on the future directions of both IBM and MS. I guess both companies are trying stuff hoping that something will appeal. I wonder whether Bell, for example, had a very clear idea about the impact of the phone when he was busy inventing it. I think the real issue which is valid for both IBM and Novell is that they are fundamentally poor at marketing their products whereas Microsoft clearly excel (sorry about the pun) at this activity. It doesn’t mean that Microsoft’s products are necessarily better (nor worse), it’s just that they have a better pitch. The IT world is littered with companies whose products are probably better (the Mac being one example) yet have failed to win the important test of the cheque book. The standard response of these organisations is to bleat about how unfair it all is, yet they have had exactly the same opportunity; they just failed to execute.

As I think I have said to you previously I am no fan of any of these research companies because there is always the suspicion of bias either through overt research sponsorship or personal bias. At the end of the day there is really only one test and that is the test of the chequebook.