I was chatting with a consultant late last week, and had a great idea for a blog posting about IBM (and specifically Lotus Notes) in the SMB marketplace.

Then Charles Robinson came along and stole a bit of my thunder.

However, whilst I agree with Charles that IBM isn't focused enough on the SMB marketplace, I disagree that it's because of any dogfooding that IBM does. It's just IBM having IBM's focus on the marketplace - they've always been biased towards large accounts and installations, and that's their corporate culture.

Which will be very hard to change, I suspect.

 

Dogfooding controls how you develop a product. As Charles points out, the idea behind it is that by using your own software, you find bugs and fix annoyances in such a way that it improves the software.

Charles goes on to posit that this means that the software is biased towards enterprise installations rather than SMB features. But I can only agree on one specific instance there, which I'll come to later.

I disagree, because this has the fundamental assumption that SMB's won't want the features that enterprises are also interested in.

The best example of this is to be found in IBM's DB2. It's a great database engine. But it was also they very vanguard of IBM's drive towards what they called Autonomic Computing.

The idea is very simple - computers are complicated, good administrators are rare, and why waste those rare good administrators on chasing trivial things on your back-end IT? Why not free them up for researching the oddities and planning/implementing improvements?

So DB2 can pretty much manage itself.

Really.

I'm not kidding here. It monitors its own health, and will adjust buffer pools or indexes almost automatically. You just ask it to do this, and it just confirms these things with you.

That doesn't mean that there aren't going to be fringe cases or oddities where a good administrator isn't needed, or can't improve over what DB2 can do itself. But it does reduce the workload of your administrators and allow you to use their time more productively. (Or to reduce the number of administrators you have, unfortunately. But let's gloss over that unfortunate potential for the moment.)

Do you think this doesn't apply to Lotus Notes/Domino? Think again.

Domino Domain Monitoring (DDM) can flag problems and then remove those flags once they're no longer problems. That's a good quarter of the way towards autonomic computing anyway. We're seeing the influence of autonomic computing coming into Domino, and don't make the mistake of thinking otherwise. It's all about reducing the total cost of ownership, after all...

I just can't accept that a small business wouldn't be attracted by self-managing software. And IBM is king of self-managing software these days.

So it's not that IBM can't succeed in the SMB space - they're very well positioned for it. It's that they're not focused on the SMB space, because their corporate culture precludes it.

(This is what I was going to write before Charles' posting, by the way. It's what I'd call "The Tragedy of the SMB Market" - that IBM has incredibly capable software that could really help SMBs, but they just don't seem to want to sell to that market.)

 

There's only one area where I think IBM's dogfooding is affecting their products aversely for the SMB market, and Charles touches on that in his posting - integration.

The SMB Market prefers (demands?) all-in-one solutions, especially now that they're used to Microsoft Small Business Server. And that means that any solution must be fairly tightly integrated in the user-access area. If you create a user or group for Notes, they must also exist for network file & print and database access.

Now, technically, I think IBM can do this. They have the technology lying around to do it, certainly. But I don't see the willpower.

IBM could give us a Linux-based SMB server that uses Red Hat or Suse, has SAMBA for file and print, Domino for mail and workflow, and DB2 for heavy database stuff. The biggest problem, technically speaking, would probably be whether we use the Domino web server or use the IBM http server (which is based on Apache) - I'd be tempted to try and use Domino, but suspect that Apache might have to run on another port just to make integration easier.

(A number of the non-Domino products will be using the Apache-based webserver, and I'm guessnig that working with the Domino webserver would take quite a bit of effort.)

Pre-configured, out of the box, this could be done. A setup script to create your Domino /O= certifier as well as setting the Samba domain and so forth shouldn't be too taxing for IBM. Especially not when they have experienced developers who integrate these kinds of things for Tivoli software for a living.

This is the only area where IBM's dogfooding might hurt them. Like most large companies, IBM have an attitude that "integration happens". It's part and parcel of running a large infrastructure - there will always be something, be it your phone system, your fax software, your photocopies - something that presents a problem. And that's what IBM wants to free up your talented administrators for with their Autonomic Computing - solving those problems.

Not to mention that IBM's own consulting department and plenty of business partners probably make a good deal from integration jobs - so there's a little lack of incentive for them there...

So I still don't think dogfooding is to blame. It's IBM's focus. I have a sneaking suspicion that their sales team prefers large sales, and that this means the unspoken rule is that if the first ten or so sales don't make you back your development costs then it just isn't worth it.

And in the SMB market, you need to be thinking more along your first 1000 or even 10,000 sales making those costs back.

 

But IBM can't afford to ignore the SMB market forever. Small companies become big companies. And the costs of migrating to any other platform are well documented, meaning that there's only a tiny sliver of a window (probably at the 500 person mark) in which a migration to Notes/Domino to get better ROI makes a compelling argument for a growing business. After that, it's diminishing returns these days - it's not that there's no benefit, but just that the benefits come with significant disruption and risks, which a business may not accept.

With Notes R8, IBM has a chance at this market again - the buzz will be there, and the new client overcomes a number of problems. IBM could take the SMB market seriously at last.

But I'm not going to hold my breath. They'd need to get people on to this RIGHT NOW.

If an SMB offering like the one I described above can't be demonstrated within six months of the release of R8, then I'd say that they've missed the best window of opportunity that they're ever going to have.

It doesn't need to ship within six months. A year would be fine for shipping. But it must be demonstrable within six months. Get the press interested, get Microsoft doing their negative campaigning - get it established that it's a credible threat, and make it clear that IBM is in the SMB market for the long haul.

 

The technology is there. I firmly believe that.

I just don't believe that the corporate focus or willpower is there.

That's the tragedy of the SMB marketplace, and why Microsoft's monopoly of it will continue.

BOOKMARK THIS CONTENT

	Bookmark at:

	Articles
	Blog entries
	Comments