Discussion on:

Sanity check: Is IT moving toward an inevitable split between in-house strategy and outsourced operations?

I wrote about the potential split between strategy and operations that may be coming for IT:
http://blogs.techrepublic.com.com/hiner/?p=531

What do you think about IT being separated between strategy and operations? Have you seen examples of this already? What would it mean for your career?

I've worked in situations as you've described in the verizon example; The problem is that too often, the only persons who can provide the strategy ARE those in operations/support. These people who really know the systems are those who live in it every day, and had input in the design and execution of the Strategy/Infrastructure/Network/Application. So, what will happen is this, for example: Suppose a backup fails at 2:30 am; the overnight lackey at the monitoring centre will call the person who is 'ON CALL' i.e. the person who spent his whole day the in the server/network room, battled his way home, had dinner and snuggled into bed to be woken by a pager buzz/phone call at 3:00 am that says "The backup failed...What should we do? The only thing you can do is try to go back to sleep and deal with it in morning when you get to work. While you are begging for sleep to overtake you, wonder out loud why management bothered to outsource in the first place if all the documentation in the world is not read by the 'Outsource Team'.

I would be very dubious to have the provider managing my WAN Links; it is too easy to borrow bandwidth from here and put it over there behind the customers back. I know of one situation where we have caught telco providers doing this red handed even IF they are not charged with support of the WAN link...

Strategy and Operations go hand in hand; it is a feedback loop that managers/Supervisor's and VP's will ignore at their peril. How will you know what strategy to follow if you do not know your tools and how they work in various conditions, including adverse ones? CIO's who are 'hands on' with staff do not have enough support in the first place.

Separation of stategy and operations leads directly to commoditization of IT.

Organizations forget that the WHOLE organization has a role of input and feedback from IT and VICE VERSA.

Programmers, it is a high value ticket and you may have no other choice but to outsource that.

it can work if you have people in strategy who take the time to learn what operations does and establish strong two-way communications. That's an absolute necessity in a split setup. It requires people in strategy that have a strong technical aptitude.

That being said, I appreciate what you're talking about. You have highlighted one of the real risks in making a split between strategy and operations.

Thanks, Jason

I agree with both of you and I've experienced this as an employee and a business owner. Day to day operations are just that - they continue the operations the way they are currently. I find they have no time, nor have it allotted to their daily schedule, to come up with ideas for the future well being of the organization. For my business, which is a technology provider, we rely on new innovative technology to drive our offerings and our strategy. We, as owners, are technical, yet have the need to outsource technical functions such as programming, to external sources. We cannot outsource the friendly faces our customers see each day. We become project managers ourselves though, so I don't see that as being a viable outsourcing solution at this time either. It's the tasks that one would assign in a development or system admin project that can be, as long as we have a project manager on staff. We are a smaller company.
I was in consulting for 10 years and found mainly big corporations were willing to hire us for the staff on a project, but felt safer to have their own in house staff manage the direction and the projects.

"...if you have people in strategy who take the time to learn what operations does and establish strong two-way communications." All the communication in the world with Operations doesn't help if the players are not focused and committed to the priorities and goals of the company they are supporting. In a well run company, these individuals get recognition and reward for going beyond the call of duty when they recognize and stop threats that directly jeopardize the company's bottom line. These company's also provide incentive to behave in a manner that improves the company's bottom line.
Outsourced support people are focused on keeping their overall personal "measures" looking good across all customers so they can keep their job and (hopefully) rewarded for outstanding measures. If you were in the position of an outsourced operations person, what would your priorities be? For that matter, would you even know (or care) about anything beyond the measures being place on your own head? How close can those measures be defined and redefined to match what is really important at the moment for each individual company?

The counter argument is; "We align the outsourced operations priorities by defining company's service priorities and establish measures to enforce it with penalties if they don't meet those measure". My response?; Good luck keeping those measures defined that so that it actually has any meaningful influence on support behavior, while your company's priorities change in a dynamic competative business environment.

Think of it as the "IT Outsourcing Uncertainty Principle". (My apologies to Heisenberg.)

Sounds like the Outsourcing version of the Peter principle. 'Promote people to their appropriate level of incompetence.'

Been there, done that. Now I work for a small non-profit. Public companies pay better, but it just doesn't make sense to Outsource IT here. I am the solution to the problems caused by having too many consultants in house.

Small organizations can't afford the communication overhead of using outsourcers (unless the outsourcer is a single point of contact who takes care of all the IT issues). It's better to have an all-purpose in-house person to take care of IT, development, strategy, and training. If it's not someone dedicated to the task, then someone paid to do it part-time.

Your comments are so generic that, whilst they have an emotional pull, upon examination they offer nothing.

ANY supplier of ANY service to ANY client runs the risk of inappropriately matching their individual motivations and rewards for their own team, versus the day-to-day or periodic needs of their clients.

This is nothing new.

Indeed, it is the basic business plan of millions of companies in thousands of industry sectors in every country around the world.

To say that this is something scary and that needs to be thoroughly planned and executed is 100% correct.

To think that there are not thousands of organisations who have the proven capacity to do exactly that, is naive.

And, on top of that, there are millions upon millions of individuals who work in these companies. For want of a better word, they simply care. Some of them, perhaps like you or me, actually always try to consider the customer's needs first.

What would you do?

It seems you reap the benefits of our various opinions...

To quote Wayne M. if I may;

"Should Not Separate Thinking from Doing"

What would you do Jason?

... techies are the 'dime a dozen' Outsource the **** out of them. Much cheaper in the long run.

Service level agreements make for a nice night's sleep.

If you don't think IT ops is a commodity already then just wait for the earth-shattering news.

I can run a large and sophisticated operation, warehousing, shipping etc and all I need is a network and internet access.

I can buy them on the street corner. The Yellow Pages has thousands of them and, guess what, not all of them are useless. Trick is, as always, finding a good one and making sure you all have the same idea about what is required and what they are going to deliver.

IT Ops delivers on strategy in the same way that the receptionist and switchboard operator do. Vitally important roles to the company, but at the end of the day they just do a reasonably basic job in response to delivering upon the strategy.

Now a programmer - that's different. That's strategic.

SLA and SLC does not necessarily make for a nice night's sleep. Handing over your IT to an outsource company constitute a very huge risk for your organisation and/or your customers.

Without necessary controls you are doomed.

further, those "necessary controls" are more commonly including things that are more difficult to measure- like customer satisfaction.

I can pound any company's statistics into the ground. And that is all they are showing me with up-time/down-time monitors and calls resolved SLAs. When the lion's share of the down-time happens during my peak hours, or calls resolved means that the tech hung up or transferred the ticket, they are showing me that they can't get the job done.

There are times when outsourcing is a good thing. There are times when it isn't.

It really is just that simple.

Our outsource model based on our complex IT infrastructure where most our resources are focused on our core business functions and information management and ONLY outsource IT operations.

the phrase:

"Without necessary controls you are doomed"

... that is such a 'catch-all phrase' that it can be applied to any aspect of any business.

The opposite remains as true:

With the necessary controls I can reduce my operating overhead and focus on the business at hand.

I said it was a commodity, which I believe it is. I didn't say it would be successful without due care and responsibility.

I do agree with you.

Over a period of time it will be difficult for the strategist to know what they need to strategise on if they are away from that technology.

How does the management tackle this??? Currently am being informed directly or indirectly to give away my solutions to the strategists so that they can implement it and I do find this to be affecting my Intellectual Property rights!!!

And moreover its a defacto to consider the offshore units as cost saving operational centres...but I would definitely like to know what should the resource do when he is in a position to provide solutions but then not allowed to implement the same??

Split operations and strategy, admin and maintenance, hardware and software. In house software and product, management and implementation..
Doesn't matter, given you can find someone to provide the service, you can outsource any part of it.

Whatever you do, you must keep some competent people in place to manage and facilitate the service in house, otherwise the third party will ream you in order to increase their profit. Worse still in many cases, they can make it extremely expensive to go elsewhere or to reverse the decision.

What would happen to my career, nothing, my current employment might change, but that's hardly a new thing is it?

Yes, thats true. Though outsourcing IT operations will reduce cost for all the known reaosns,it will however bring stingent application of laws such as DPA as well. Many organisations have no outsourcing clauses as well as they dont want non employees to deal with their confidential information. So to what extent, will outsourcing go is still a question?

In the 20 years I've seen the outsource model appoached
umpteen different ways. I've seen very cyclic models
where the mantra was outsource, and then turned around
to in-source or repatriate the jobs, then back to
outsourcing (offshoring, etc.), then back, etc., etc.

Bottom line seems to be if it's a long-term permanent
full-time work, insource it. If your environment is very
process heavy with lots of Project Management overhead
(a.k.a. paperwork), new people take a long time learning
how to get things done and this too should be insourced.

Now we're finding with the close tie of Problem
Mangement funneling into Change Requests, etc. that one
unified work group seems to be a more efficient model
with less 'hidden' costs.

Good points...
The main thing is that Outsourcing to an external organization brings significant internal costs in oversight.

Speaking as a former Outsourcing manager, our strategy was never to reduce the customer's cost. It was to show a significant ROI at the onset (win the sale), and to charge back to make up for it later on.

From the perspective of the corporation, outsourcing was a power play to reduce employment in IT - and to centralize IT across the divisions. It also looked good to stockholders. But, it was a disaster to operations.

A shell game overall. Its all about the benefits costs...

There's always a lot of stuff to externalize, and a lot of stuff you have to DIY. The more strategic IT becomes, the more you have to insource talent, not outsource.

Outsourcing is like buying off-the-shelf software -- a lot of tech, for cheap, probably standardized, but it's not exactly what you need. You have to adapt your operations to the software. Likewise, you have to adapt to outsources a bit, and they to you.

Eventually, if you outsource enough, you have to decide on whether you hire a specialist to manage all that stuff, or you outsource even that work. If IT is strategically important, then you obviously have to insource that, or lose the competitive advantage.

It's like a small business using some accounting software. As you grow, you eventually need a bookkeeper and accountant both outsourced. Then, you have to insource the bookkeeper. Eventually, you may need to insource the accountant. Then, you may again need to hire outsourced specialist accountants to augment the internal staff. It necessarily goes back and forth between needing expertise quickly, and needing to have control over strategy (and organizational stability).

It seems to be the same with IT - always shifting. In fact, it seems to be the same with everything - companies and organizatiosn have a mix of internal and external workers. They need this mix, just as you need a balance between change and stability.

I read the article with interest having seen a number of outsourcing contracts come and go.

The one thing that is clear to me in all my experience is that outsourcing companies never attract top IT professionals and if they do it's an incredibly small percentage. Instead they pick up cheap resources generally treat them like slaves and expect to have staff turnover to rival the England Managers job! (that's football, not soccer by the way!!! I don't care what Beckham says :-))

Outsourcing companies are not interested in attracting the best IT staff, they are interested in their bottom line and the cheaper the saff they can put on site and get away with it, the better. After all they have shareholders to feed.

So all that happens is that the operation service declines, the outsourcing contract is not renewed and the services are then insourced, things get better for a couple of years until management think that it's costing too much and since the pain of the outsourcing has been forgotten mostly companies are dragged straight back in and the cycle starts again.

Having been around when corporations were a too tight on the inside and micro-management ruled the day, this seems logical in a business where ROI rules.

When the term, "desaster recovery" became, "BCP" (Business Continuity Planning) it was a
top-down process and in most of my experience
has been for Audit purposes only.

It is going to be that the seperation will
happen, just timing and org. changes need to
be signed off on.

But, if the two units don't have some solid
purpose mandated by the BCP bible, this could
lay down really slow and tedious response to
planning and execution when it is needed most!

"IBM, HP, and Verizon all have thriving managed services divisions."

If it?s solely about huge multinational companies aka "investments" with 1000's or more people that is one thing but for a company in the U.S. or any single country a lot of the statements that you quoted fall apart or just don?t make sense. In fact the closer the locations of a company are together the less it makes sense. To Out source in house work to a source that performs that work outside the company is going to be limited to remote access. You can have Help Desk performed in India but unless they have a magic carpet they can?t perform tasks that require a physical presents. As far as Help Desk goes you can do that anywhere and support multiple companies no big deal there. I can see remote offices having service Out sourced when it makes more sense then having IT folks and others fly to Tim Buck To and back every time they have a need.

Now a days you have companies getting flipped like pancakes. So with the rush to incorporate all the new parts and jettison the unwanted parts I could see Out sourcing those processes and for that matter all the IT jobs since the companies are so volatile that the jobs are much closer to being a temp job that anything else. I just don?t see much room for it in companies that are long term bed rocks of their community where jobs can span many years and even decades or small and mid size companies and companies with offices in relatively close proximity. If the company has a lot of IT diversity then it will be difficult to find an Out source company to handle everything. They only want to deal with companies of a certain size they will only accept a minimum amount of revenue if you don't meet that standard they are not interested in doing business with you. Imagine having servers managed by one company and workstations by another. You will have one to your right and one to your left pointing their fingers at both you and the other guy.

I would like to point out the service companies mentioned have many different divisions and divisions with in divisions. They get highly specialized and restricted. We had one for an ISP and managed firewall and of course they supply our T1 lines and various other telco lines. First of all the ISP is a different division from the managed firewall which has present some serious challenges to us over the years when one division feels that the other is at fault and leaves to us to deal with it. They can't say anything that could be viewed as stepping on the others turf. Second we have Cisco Switches and Routers and they both sale them and the Cisco warranty support. Of course it?s 2 different divisions. I figured that I would get a quote for new equipment and for a warranty support contract on existing equipment since we already had a relationship. Wow what a trip that was. First off our Rep changes about 3 to 5 times a year and we often are not notified who the new Rep is and given the contact information. So to find out who it was at the moment was quite the challenge. Then I found out how micro managed everything is. One item per division. Then I found three people need to get involved just to get a warranty support contract. You have a sales person for that product then a person who keys the serial numbers in and a go between with Cisco. I needed to talk to all three often via the Rep. They don?t communicate well with each other let alone the customer. Needless to say I decided not to buy anything from them. I just could not imagine having them manage our T1 lines and Cisco equipment. We took the firewall in house when we went with the MARS security system. We did Out source configuration and installation of the equipment but it will be managed and serviced in house. I can get training as we go while having the system in place. I could not imagine Out sourcing management of our T1 lines Routers, Switches and Security to those folks (Yes I know but it needs said twice). We have them for ISP and telco, we have to have somebody for that.

As I have said some things make sense to Out source but no where near the scale mentioned. You can do one time deals and things that are needed and in house can't get to and odds and ends that in house is burdened buy.

My point is I think the article needs to define Operations with a fine tooth comb. Exactly what is it? If you think that everything except for one or more big brains who make the decisions are going to be Out sourced dream on. We have software that interfaces with a vendor in real time and the vendor does most of the support but the software manufacturer gets involved at times and we provide some and that is Operations. How would you Out source our portion? It's performed on the fly. Are the IT jobs they refer to that are left in house merely dealing with the 800 pound Gorillas that everything gets Outsourced to?

We are outsourcing our software to a 3rd party, (Not Overseas) but still outsourcing. We will outsource the project management for the implementation to a group out of Chicago. But strategy and direction remain in house as well as operational support for the hardware.

Software support will be split with 1st line to the vendor, with onsite help from in house staff.

We arrived at this point because it is not cost effective to manage system change in our legislative envioronment. The vendor can spread that cost over 1500 installs. We have to do the same thing with one install. Up to now, we have done it, but the changes keep coming. And those changes are required by law, not just market forces. We have those also, but they can be delayed and worked in. The legal requirements have to be done by government deadline.

So we are outsourcing. With outsourcing we will lose a lot of flexibility, some safety, and some tactical advantages. With it, we will gain standardization, additional support resources, change management, and stability.

I don't agree that the outsourced company always only has a few experts. We found a company that has a solution to the revolving door and keeping talent. If you are interested, google "Meditech" and you'll see both the good and bad of doing it their way. Job satisfaction is in direct inverse proportion to length of service. Their way is not fair to everyone, and they don't care. They still have 1/3 of the market and are relatively stable. So it works for them.

I have worked in many areas of IT--tech support, programming, web development, consulting, etc. I was also part of the military at one point. What the numbers folks fail to realize is that you can't commoditize people without regard for how they really contribute, just like you can't commoditize large sections of IT operations at will. Our director and bean counters tried to save money by handing over some of our operations and databases to a third party consulting company against IT's wishes. After two years of struggling, our IT group was handed control back again.
Why did we revert back to IT operations having control? The consulting company was robbing us blind and doing some really crazy stuff. They were giving subcontracts to their friends without considering how it would affect our business. The backup plan they suggested for a couple of our servers was overly complicated and cost us a lot more than it should have. Hardware, databases, operating systems were also bought based on who the third party was in bed with. Not only that, since we were downsized to nothing, we couldn't handle what the third party gave us without hiring more people. Hiring more people involved training , etc. We then had to go through what the third party did with a fine tooth comb. It was a mess. Four years later and we are still recovering. Our HR and customer service systems have never worked right since then.
Fact: people don't care about data until they don't have it. Another fact: the people responsible for accurate, reliable data should be put above politics.
Since IT operations is responsible for that data, they shouldn't be treated like a stray dog on the days that all hell isn't breaking loose. As they say in the military,
without beans, bullets and bandages, you lose the battle. We learned that lesson the hard way. The people that provide the beans, bullets and bandages to your front line people should never be outsourced. The lines of communication between the two should also be very good. Trim the fat if you have to, but never the muscle.

I think it is a better framework to work in, both for the IT Pros and the Operation IT, if done right. If there is someone involved in operations i.e. has extensive technical experience yet has enough time to spend studying and devising strategy then you hit the right mix. However if the guy making the strategy does not know how it would apply and affect the day to day operations then you the Queen of England scenario where she asks people dying of hunger, why they simply can't eat biscuits.

The King Of France's squeeze, and it was Let them eat cake"

Though others have stated it was another Marie, or perhaps even the Loius himself and "Let them eat pastry"

Anyway you look at it the only thing english about it iss the mis-translation.

Good point, bad allegory

If an organization sees the need to separate strategy from operations, they need to examine the leadership in both groups. As a operations leader, I would not allow myself to be shut out of the strategic loop. As a strategic leader, i definitely want my operations people to know the company vision.

As for outsourcing, the providers often speak of a poolof talen that they have, but in reality, they keep a few experts around and fill the rest with novices. That is how they make the financials work.

You cannot keeep operations out of the strategy loop or vice versa for that matter.

Operations is a real time implementaation of a strategy, coming up with one that can't or shouldn't be implemented is stupid, when all it would take to highlight any problems is two people who know their aarse from their elbow , having a chat.

For some reason best known to ******** beancounters, IT outsourcing is different. It's no different to getting in any other third party. Someone on 'your' side of the fence has to know the requirements, make sure they are met, make sure they stay met and above all make sure the service is still saves 'you' money, which was the only reason to do it in the first place.

This is the bit that is not happening.
I've watched helplesly as third party providers took the absolute urine out of their customers, because they knew they they had them by the nuts.

Just utterly simple things like writing very expensive to maintain software and then selling it to someone who didn't know the difference.
Can't blame the third party, it's just a good short term business strategy and everyone likes them don't they?

Your writer must be new to IT, or very young and wide-eyed. This strategy-vs-ops split has been around since I started in IT 20 years ago, and even then, my seniors had seen it all before.
The fact is that not everyone in IT can be a strategy-setter; there has to be someone to do the donkey-work. There has always been the discussion from 'operations' staff having to the monkey tasks, occasionally shouting loud enough to be heard that they are underskilled (read: bored) of effectively routine maintenance, whilst the perceived Top Gun architects, systems development or the similar roles get all the interesting, sexy design-and-implementaion tasks. What has been identified is the monkeys cost unnecessary money for repetitive tasks; spin those tasks out to a company who set their employees' expectations from the outset and of course you'll get a cheaper service, because the outsourcer can set their rate according to demand and supply of people willing to do those jobs. It's nothing new; I left an outsourcing company 10 years ago because this was happening, and as I mentioned, it was happening 10 years' prior.

Outsourcing is nothing new and the split has been happening at the companies that outsource part or all of their IT department. However, it has always been the exception to the rule. I'm talking about the split between in-house strategy and outsourced operations becoming the standard model for organizing an IT department.

most Companies that use Network Operations Centers (NOCs) have contracts that are restrictive in what the company can or can not change and how that change takes place. The ability for a CIO to take advantage of new technology becomes a lot slower process because they outsourse their day to day operations. the NOCs must review the plan and whether it will fit with their day to day operations, whether they have skilled personal to implement the change. Then provide pricing. The Company may or may not be able to afford the change. Most NOC contracts are long term.

So what the split really does is create a slow moving entity simply because the CIO can't tell the IT department to make the change. The whole change process is far more complicated involving contract changes with subcontractors etc.

The IT field has become far more specialized which in turn creates many more jobs. But with that specialization comes inherent risk that those one hires can only do their specialized job.

I talk with several hundred companies a year about outsourcing. The approaches run the gamut from simple co-location to shedding entire IT staffs.

The ones who get it realize that they need to concentrate on business not technical problems. Their choice, given a budgeted pool of permanent talent, is to bring in resources at the strategic level or at the infrastracture level. If they hire at the top end and keep the low-level tasks internal, they are marginalized at their company -- they're just guys in the basement with propellers on their heads keeping the lights blinking. If they outsource the mundane, common tasks, they give their staff an opportunity to move into roles with higher business and career value.

If you think that your career at your company depends on changing tapes, patching OSs, looking at firewall logs or coding dialog boxes, you are already overpaid and are just waiting for the inevitable.

I've been in IT for 20 years myself and continue to wonder how strategy in IT can be so difficult. I can see if IT is the foundation for your business in which case I would keep strategic functions separate from operations. IT and IT work is becoming to piecemeal, which explains why so many projects fail. The loyality of anything outsourced only runs as deep as the contractual agreements. Meanwhile my career seems to have become about 50% breaking in IT managers and training contractors.

Well I am glad my children don't want to go into IT. They saw their old man get long term layed off in 2002 and decided that was it. My painful experience showed them that IT does not have the stability that would cause my kids to obtain a 4 year degree in any of the engineering professions.

This article doesn't condone outsourcing on all of IT however the increased volatility in operations jobs, which were the only available jobs during the 2001 - 2004 depression, puts pressure on the whole job market.

The author addresses IT from a purely job supply view point and doesn't look at demand at all. My children aren't alone either, many of the college students in the USA feel the same way. Why go into engineering or IT? We don't treat our engineers and programmers particularly well in this country. In this scenario they layed off at the drop of a hat.

Though I see the market shifting towards outsourcing, I don't feel this to be an efficient approach. One simply cannot separate strategy (thinking) from operations (doing). This was a lesson we supposedly learned during the quality revolution of the 1980s - rely on integrated teams.

Outsourcing provides an additional level of indirection between strategy and operations, requiring, as noted by several above, higher commitment to communication. Although this may work in the very short term, as operations knowledge becomes lost to the corporation, the ability to oversee operations effectively also becomes lost. Refer to the ongoing difficulties of the US Federal Government in overseeing contracts as a prime example of this problem.

By outsourcing operations, these duties become a commodity and are largely driven by cost rather than quality. In fact, poor quality is often beneficial to the outsourcing provider. It is much more effective to bill lots of poorly qualified staff than a single knowledgeable person. It is more profitable to maintain inefficient manual processes than automate.

The separation of IT into an independent department has already lead to major disconnects between IT and business. By outsourcing IT, change requests become contractural negotiations and "scope creep" becomes the standard mantra. At a time when IT needs to be ever more closely aligned to business, we should not look to make it even more independent through outsourcing.

I have to totally agree with you on the part of poor qualitly being beneficial to the outsourcing provider. I can not tell you the number of times in my 20 years of experience in IT that I have seen companies outsource projects where milesstones aren't met, deliverables weren't even close to requirements, etc.. Instead of getting rid of the outsource company and hiring a different one, their contracts are renewed because "too much time and effort have already been invested to turn back!". Ridiculous! So what incentive do oursourcing companies have to actually finish the job? None! When they finish the job there's no more money so they drag it out as long a possible. They hire inexperienced, untalented people so they can pay them as little as possible. They do the minimum amount they have to do for the maximum amount of money. Outsourcing is not the answer.

Sorry WayneM but there are thousands of examples where your underlying principle is not correct.

One that immediately springs to mind is Cisco.

This was originally a manufacturing company.

Now, all their manufacturing is outsourced (or certainly the great majority).

On the other side of the equation are companies such as Flextronics. This is a giant global company who most of us in the general population have never heard about.

Their very existence is completely based upon outsourced manufacturing. That is what they do.

Now, when you talk about the bond between strategy and operations, there are not too many closer than a manufacturing company and the products that it manufactures, distributes and markets.

On that basis, I'd suggest that the IT operations component of a company's infrastructure pales into insignificance.

The risks that you discuss are totally real but all can be managed. I'd suggest that the risks you describe already exist in most internal IT infrastructures and are usually poorly managed on both the business and IT sides anyway.

Remember, one of the (many) drivers for outsourcing is the US's focus on quarterly performance of public companies, and in particular the ratios of revenue-per-FTE, and EBIT-per-FTE.

A company's expense line may or may not actually improve with Outsourcing but that is not the only important factor. As long as the FTEs are not on the balance sheet, the analysts will give better performance ratings.

You can separate who implements a strategy, given that how they implement is understood or documented.
Why and what though is the bond you speak of. I guarantee Cisco don't ring up PartsRUs and ask for a new model of router to sell. They simply don't assemble the things.

They have to define what they want, understand what that means, source suppliers who can do it, make sure they have done it...

Outsource IT, catering, snail mail what ever you like, you must have control over what you want and what you are getting, otherwise you'll get what they want to give.

I know all this sounds really obvious, but apparently not in IT.

... I reserve the right to copy that title at a time far in the future.

Yes, I agree with your points. Hopefully within the bounds of language differences that is what I have been trying to say.

I am not actually for or against outsourcing. I am for what works best for a company in their circumstance.

To suggest outright that Outsourcing is, on balance, the 'worse option' is simply and demonstrably incorrect. And I don't think I am drawing a long bow by saying that is exactly what a number of people have suggested.

The "bonds" of which I speak are the implicit ties suggested by WayneM in his discussion of the need to keep strategy and operations tightly coupled.

It seems to me that, in arguments against outsourcing, smart and experienced people say things that other people use as the very basis for outsourcing!!

it will work.

However you won't get that intial saving, that guarantees a new chair somewhere else before it all goes nipples up.

Outsourcers, who will now be expected to meet some very specific and long term criteria, will see their margins reduced, don't have much scope for cost reduction, so prices go up.

At the moment outsourcing mostly seems to be about low hanging financial fruit. There is no future in this, not for the service providers, not for the consumers.

You hit it dead on. Our company is just beginning this switch. The hard part is implementation because in the short run it will take a lot more money to run the business. In the long run with the complexities of the business our company will be better off. I must give manangement a node for understanding this as this is truly the first major effort in a long, long time, that I believe is worthwhile and will pay off in the long run.

Face it folks. Any process that can be easily digitized within a company will be outsourced (most likely to India). In the next 5 years or so I would expect the number of American companies doing business with India to triple and India is already ready for that. I agree with the author about the split because I am seeing it happen first hand on the operations side within the fortune fifty company I work for. The cost per employee for a given support task is so much cheaper on the spreadsheet when you transfer it overseas that virtually every company in America and Europe realizes that if they expect to survive on the global market in the future they need to follow their competitors. Countries like India have successfully built their infrastructure and are advertising to the world that they will gladly perform the "monkey" jobs that companies can't afford to have their own employees do. While first and second line managers complain with us front liners that the Indian resources do not always posses the same level of understanding and quality of service, I can assure you the top brass is and will continue to focus on the bottom line to investors. Unless a major breach of security causes companies to take a second look or the overall standard of living quickly increases in India I seriously doubt the splitting the author mentions will stop. The average Indian worker makes $200-600 a month and that is considered excellent wages by their own standards. I'd say it's going to be a while before the prices to do work in India will be on par with America or other parts of Europe. Rather than fight this inevitable trend we'd all be smart to start becoming more versatile. Humans do not like to change from the equilibrium they strive for and live within but this change is already upon us. It's time to start seeing where you can take advantage of the new markets.