Outsourcing: A symptom of poor Governance!

In their book entitled “IT Governance: How top performers Manage IT decision rights for Superior Results,” Peter Weill and Jeanne Ross list some of the symptoms of poor IT Governance.

Notably they cite “Senior Management Sees Outsourcing as a Quick Fix to IT Problems” as one of these symptoms.

The thrust of the book is that IT Governance mechanisms should be aligned with the desired outcomes of the business itself. The inability to articulate these desired outcomes in a clear way is one reason why the mechanisms may be misaligned in the first place.

Outsourcing of specific IT functions ironically requires that these outcomes be defined in a greater degree of clarity than would be required if they were provided by an internal department.

So we find that when we approach an outsource provider for a report that “we always used to get from the internal department” they give us a quotation for providing that report. Now this is as a result of not being unable to specify exactly all of the detailed specifications of the tasks being outsourced.

So in order to effectively outsource a function, a very detailed analysis of the steps and responsibilities for that function needs to be done – by the enterprise, before outsourcing that function. Failure to do this will result in the enterprise paying more for tasks that were left out of the outsource deal, and should have been included.

That detailed analysis of the function may very well be what is required to enable the old internal provider to justify needed training or additional resources to be able to provide it effectively… or simply to realize that it is providing that function in a way that is incongruous with the organization’s needs. What prevents that from happening is called poor IT governance.

Senior management can no longer live with the function performing poorly, so they outsource it. What that achieves (if they are lucky) is the surfacing of all the issues that should have been dealt with by effective leaders and managers within the organization.

Oh dear!

Now the enterprise is worse off than it was before, and it probably costs more to get what they had before they outsourced.

So my counsel is to get IT Governance right before doing any outsourcing, having noted that the actual desire to do outsourcing is a possible symptom of poor IT Governance!

Getting IT Governance right starts with clearly articulating what the Enterprise needs from its IT function and translating that into IT Governance mechanisms. Doing that is a lot easier than sufficiently articulating every detail of even one outsourcing contract, and will pay much greater dividends in the long term.

Technorati Tags: ,

Posted in ITSM, Management | 1 Comment

Happy New Year! … Now get busy.

2007 is here(!) and hopefully the 007 part indicates a year full of amazing new tech toys.

The rumours of the iPod Phone are everywhere, the Playstation 3 is looming, Windows Vista and Apple’s Leopard as set to challenge each other for the desktop among many things we have to look forward to in 2007.

2007 is the Chinese year of the Pig, but will 2007 be the IT year of SOA or the CMDB, or will it be the year of Virtualisation? Virtualisation is less tangible (no joke) but its effects will be dramatically felt.

Probably the CMDB needs to come first. It is astonishing how often I hear about organisations that do not know their environment to the level of detail that configuration management denotes.

Standards must have benefits
Standardisation is the precursor to both of these in my opinion. With a clearly set out schedule of standards, the IT organisation can move forward in the knowledge that it will converge on a set of chosen technologies.

Standardisation will unlock the benefits of virtualisation. Standardisation implies a finite set of technologies which can be tested and deployed in well-known environments. Once a specific technology has been tested and a “best” configuration has been determined for a given environment, the process of deploying a new instance of that technology should be dramatically cut-down from the process used to deploy and test a new technology – (otherwise why make it a standard?)

Now Add Virtualisation

Once something is a standard and is deployable using a subset of the usual lengthy deployment process we have an opportunity to virtualise it – make it so easy to deploy that you can create and destroy instances of it on-the-fly. Go one step further and give projects the means to create and destroy instances themselves, from a menu of tested options and well defined variables. I take it for granted that if you were to do this, that the instance created would automatically register itself with your CMDB and notify the Change management process of its existence etc – all the required processes would be followed, but AUTOMATICALLY.

It an environment like this we have derived REAL benfit from Standardisation and Virtualisation.

What about the CMDB?

Yet again the CMDB seems to be implied without being too explicitly mentioned. The fact of the matter is that if you allow on-the-fly virtual deployments of standard solutions, then you will quickly lose track of them and be on a wide and swift road to capacity management hell, UNLESS you have a solid grasp on your configuration situation. This is the key that allows IT to manage the chaos that the users have been empowered to create. Don’t even try to do on-the-fly virtual deployments without a CMDB. What is more, on the journey to standardisation, the CMDB will be your yard-stick to measure progress towards standardisation.

“But a CMDB is such a huge and complex project,” I hear you say.”We’ll never finish,” you complain. You can’t start yesterday so my counsel is that you had better start doing config management today. The longer you delay, the less likely you are to survive as an IT organisation in a virtualised future; And I have very little doubt the future will be virtualised.

So welcome to 2007. Now get busy.

Technorati Tags: , , ,

Posted in Challenge, ITSM, Management, Technology, Trends | Leave a comment

Never Give Up!

There are many occurrences in IT management that continue to amaze me.
They happen so often that I have decided to call some of them “Laws of IT Management.”

Some Sober laws of IT Management

  • The Law of the Surprise Victory:
    • Just when you’re about to give up is when you suddenly succeed
    • The Law Unpacked:
      • Even if you think you are shouting against thunder and that you are the only person who can see the disaster looming, don’t give up on raising the issue. If you are sure you are right, then don’t lose confidence. Eventually sense will prevail – and usually it will happen when you least expect it, or when you feel particularly low about the fact that no-one is listening.
  • The Law of anti-Halation:
    • As an IT Manager, you will very seldom be the hero or the saviour of the day.
    • The Law Unpacked:
      • Just days before the plans you have been working to motivate for
        years can actually be implemented, a directly related crisis occurs
        which requires that you rush the implementation to resolve the crisis.
        You are then in trouble for not having implemented the solution earlier.
      • Accept it. Your wonderful solution to a problem you have known about for a long time will be turned into a damage limitation exercise at the last minute.
      • When I miss a turn or take another route, the Navigation system in my car simply recalculates the new route to my destination based on where I happen to be at that moment. It doesn’t get angry with me for missing the turn.
      • Do the same. Don’t get overly upset, clearly communicate the facts, then replan based on the newest turn of events.

Technorati Tags:

Posted in Challenge, Management | Leave a comment

New Site

I have renamed and moved “The Smell of Fresh Technology” to its new home at sobercounsel.com where it will be known as “Sober Counsel” (surprise surprise)

If you have subscribed via Feedburner your subscription will continue, if not – better go and have a look at www.sobercounsel.com

Posted in Announcements | Leave a comment

Arbitrary Important Decisions

The problem is: A decision is needed. Almost everyone agrees that it isn’t really important what the outcome is, as long as an outcome is reached. The decision is therefore arbitrary, but is important.

Who will take responsibility?:
This doesn’t change the problem, it simply makes someone responsible
for resolving it. The success would depend on the authority they are
perceived to have in driving a decision.

An Executive Decision:
The decision is escalated to the Executives to make. Some balanced information could be presented to the executives and possibly some short listed outcomes. The reason the decision hasn’t been made in the first place could possibly reflect a conflict of opinion between the executives themselves and so this may simply be moving the problem and not drive out a speedy result.

A management mandate:
Someone is mandated to obtain a consensus and achieve an agreed decision.

Ongoing avoidance:
Everyone avoids the decision until it becomes so critical that something breaks and an outcome is forced.

Someone sees the issue as important and takes it on themselves to try to drive out a decision.

So now someone is responsible or at least everyone agrees that a decision must be made quickly. Now this person has several options to achieve an outcome

Unilateral Decision:
The responsible person could attempt to make the decision on their own and then sell the outcome to all the stakeholders. Depending on the percieved authority of this person, this may be completely unsuccesful or may achieve limited buy-in. It is unlikely that anyone else will feel a degree of ownership sufficient to care if the decision turns out positively or negatively.

Management Decision:
The responsible person could convene a forum of the affected managers and facilitate a deicisonmaking process to drive out an outcome. This is significantly better than the first option, but only the management team are likely to be driving the outcome.

Stakeholder Negotiation:

Put all the Stakeholders in a room, and facilitate the debate until an
outcome is agreed, either by consensus or majority or in the worst case
unanimous decision. This could take a long time, but hopefully an
outcome has broad based buy-in. (This may not have to be a physical meeting if a facilitation process exists to support a remote method of achieving consensus.)

Where the decision is not arbitrary (i.e. there is a right decision or
a wrong decision) then the process would need to be supported by a
carefully planned process and be informed by appropriate research.
What is critical then is that the criteria for making the decision are
agreed in advance.

So really this applies to any important decision, where buy-in is important. Try to include as many people as possible in the process and plan carefully so that the facilitation method used ensures that an outcome is achieved. The important thing is that everyone identifies the outcome as one they participated in achieving and those who disagreed with the outcome acknowledge that the process used was valid and agree to support the decision.

Posted in Management | Leave a comment