Kill the CIO!
Bad budget news for the big boys
Posted in Management, 16th October 2007 00:15 GMT
Watch Now : Virtual Machine Movement with Hyper-V
It's budgeting season, and this year is no exception for big corporate IT departments, and the big IT industry players who feed on them. Pressure is yet again on reducing IT costs, and especially costly IT people.
And as in most recent years, cost reduction ripples will spread across the IT pond, with the major IT industry players, according to Business Week, again bracing themselves for lower corporate spend. And, most worrying for traditional IT company investors is the fact that corporate spend is traditionally high margin, and its reduction therefore profit dilutive.
Outside of acquisition-driven consolidation, the IT cost reduction challenge is really about acquiring technology differently. CIOs who are well past their technology prime often lack the know-how or confidence to meet and manage this challenge. And for a growing number of CIOs, their IT staff is so depleted from historic cost cutting, that they often don't have the horsepower to work beyond keeping the trains running.
This is good for the big IT industry players, but bad for their customers whose IT shops are increasingly on a treadmill of doing the same or more for less - and typically faced with cutting close to, or increasingly into the bone. At some stage, these CIOs crack and outsource, which means passing the IT baton to an external organization that has to provide the same services and levels for less cost, while expecting a 30 per cent profit on the business in the process. Yeah, that really works!
CIOs who deal in reality are faced with managing two different technology cost models - big science and big commodity. Big science is typified by past major IT investment programs, like SAP implementations, that have paid out all the business benefits they are ever likely to, and where change is offset by risk and work place disruption. However, hidden in Big Science are all the highly expensive underpinning enablers that can be addressed, often in isolation of the user community. These include things like Oracle software, the operational/management tools from folks like BMC etc. - all of which are Big Science-priced, and provide huge cash flows for the big players.
Most of these enablers historically justified their existence on optimizing performance, which, thanks to commoditization, low cost hardware, and increasingly low-priced software from Open Systems shops, are no longer a priority. Customers going that direction are worrying for traditional IT industry players who capitalize on expensive proprietary product cashflow - un-bundling and keeping mainframe software expensive props-up IBM's profits, I suspect. And I'm sure this same economic logic is driving Oracle's BEA takeover thinking. (The whole European Community action against Microsoft - which in reality is now essentially a cash cow legacy software provider to the corporate world - is about the company's exploitation of its proprietary products.)
The Big Commodity model, which includes Open Systems, covers all the things that continue to drop in price - network communications, commodity priced data storage, low cost computing equipment and cross-over Internet services (like email). The challenge here is to leverage commodity pricing in the corporate world. And this really is a no-brainer - savvy IT shops with strong technology leadership are already doing it aggressively. Such companies are piloting Google Apps and running virtual private networks over low-cost Internet connections. They typically already:
- Use NetGear for network equipment in place of Cisco, saving huge mark-ups;
- Simplify computer security by using standards like VPN tunneling and secure sockets, shifting system access control responsibility to the user departments, and strengthening company IT policy and its enforcement;
- Act like Google, buying commodity computers and relying on standard open system configurations
- mirror critical data in My SQL databases using cheap-as-chips commodity disk storage in place of Big Science Oracle AND EMC products;
- Out-serviced things like email to Google and Yahoo. Their undeniable economies of scale and up-time motivation, make comparative large corporate IT shops look sub-atomic in cost and service level comparison;
- In support of mobile workers, office hoteling and home-working, are blending standard VPN computer access and commodity VoIP telephony services into their communication mix and exploiting low-cost video conferencing services for companies like Skype for internal use;
- Exiting ownership and support of staff personal computers, mobile phones and other IT equipment - instead providing allowances and funding home broadband.
Their savvy CIOs are fully fledged members of the management team, able to couch technology expenditure in real business terms, and are not afraid to quantify risk, manage and mitigate it. In short, they lead.
For the other guys there is of course the IT "brainer" option - to reduce IT cost by de-layering corporate computerization. First nixing applications that provide no bankable or strategic business benefit. Then killing political refugee systems that were supposed to be replaced by those yesteryear Big Science projects, but whose owners who dodged budget responsibility, refused to let go.
Naah, that would involve working with the business folks and helping them understand real IT economics. Far, far too risky for an endangered species! At least for the short term, I suspect corporate IT will spare the big industry players some of the reality check they need. ®
Cormac O'Reilly is a late sixties IT industry entrant with early developer gigs in London at Abbey National, Unilever & BOC. Senior IT oil field trash in the eighties and nineties; Schlumberger (Houston TX) and Shell (The Hague). Board IT big-wig at Costain (London) before CIO/CTO at Digital and Wang Global/ Getronics (Boston). Non-exec director at two flame-out dot.coms; now spending ill gotten gains and being provocative in Newburyport, MA
Watch Now : Virtual Machine Movement with Hyper-V
COMMENTS
RE: how (WARNING!! TROLL ALERT!! Someone keep Johnny on his lead!)
Well, you could try running HP OpenView Network Node Manager on Linux.... You may also find that intelligent design circumvents many "it's broken, the business is dying" problems. Oh, and those lovely peeps at Red Hat and Novell are very happy to sell you a support service which will (in your case, obviously quite easily) surprise you.
Big hint - look beneath the covers of many of the products you listed and you will find a Linux- or FreeBSD-based OS controlling the product.
how
how does I watch broadcast storm wash back and forth across my cheap ass switches???
How to sink your business in 8 easy steps me thinks.
O linux fan bois - you try managing network security and identity with linux, sure you can do it, but it takes knowing a truck load about ldap and kerberos (well unless you don't want the security part.) There are things linux is godly at (namely running my database) and things it's crap at (network user management).
Wanna know the best thing about our brand name linux, our Cisco switches, our Oracle databases, our Checkpoint firewall, our Mitel phone system and our EMC storage array?
They're safe as houses. If they break we pick up the phone and say "this is broken - make it work" that's if they break. Your paying for reliability, stability, power and performance.
Sacrifice that and your ass will spend the next year running around fixing stupid problems, until eventually it all just falls into a heap and you need to spend twice as much pulling your self out of the hole you've drowned in.
@fanbois
So, we can all agree then that Linux isn't 'free' as in zero-cost? If it's to provide a secure, reliable service It needs to be supported, just like 'Windoze' (LOL). These support costs will (over the lifetime of the equipment) almost always outweigh the initial purchase costs of hardware and software.
Whether Micro$oft (ROFL) or Linux provides the lower cost solution will be case dependent. Anyone who simply claims that Linux will always be cheaper because it's 'like. free, man' isn't going to survive long in a business environment.
Re: "Linux is only free if your time is worthless".
By that very logic then Windows is hideously expensive and all companies would be able to significantly reduce their IT spend by canning it.
Oh wait, hang on...
Hmmm
The guy used to be the CIO/CTO of Wang/Getronics, he must know what he's talking about... (sarcasm)

Enabling efficient data center monitoring
The new Office Garage series:
Top 10 SIEM implementer’s checklist