Freigeben über


Is IT a commodity that a corporation would be well-justified to outsource?

This question has been asked before, but not often of people like the readers of this forum... the folks who write the code:  Should IT be outsourced?

Reasons to outsource IT:

1.  In some organizations, IT can never reach a size to specialize sufficiently, therefore permanently relegating the IT staff to generalist roles that are usually difficult to fill or may cost more to perform.

2.  Many of the functions of IT provide no strategic value to the business, and therefore the business should seek the lowest cost way to provide those functions.

3.  If you outsource IT, then the technical investments that are needed by any good team are not scrutinized by the business.  Instead, they are hidden in 'service fees.'  This allows the IT team to spend the money it needs on prevention rather than fighting forest fires.

4.  Most IT shops are terrible at what they do.  Rather than keeping a bad organization running, a company can do better by allowing the competitive market to pick the winners.

Reasons not to outsource IT:

1.  Information buried in the corporation is difficult to secure if it is hosted at an external IT provider's site.  The company is still liable for breaches in trust, even if the loss occurs because of an error at the outsourced provider. 

2.   If you outsource IT, you get non-strategic software.  Any potential that IT has for providing strategic value is lost, because outsourced providers, by definition, are unable to deliver strategic value at commodity prices.  (This statement may not apply to companies that simply outsource the staffing and management functions, but keep a team dedicated and on-site to deliver business solutions).

3.  Outsourced IT workers are less familiar with your business, and may be less dedicated to the success of your business than regular employees. 

4.  Investing in IT is, for a large part, investing in people.  Any growth of knowledge and skill earned as a result of IT money spent by a corporation is intellectual property accrued to the outsource provider, and not the company itself.  Therefore, if staff changes or, heaven forbid, the company switches outsource providers, then all that investment in human minds is lost.

What do you think?  Should IT be outsourced?

Did I miss an argument that you think should be in there?

Comments

  • Anonymous
    June 27, 2006
    Well I believe I.T. should be outsourced when a young start up ocmpany is attempting to save money. That is to say if the company is not in of itself a information technology based company. For instance many resorts and hotels would really benefit to outsource their operations to one company that specializes in that market. So yes I.T should be outsourced in many cases where that type of service is more of a cost and not an investment. However as a company grows I believe a dedicated staff of I.T. workers should be a strategic part of every large corporation. I also believe that customer service should not be outsourced if you want to maintain your customer base. The reason is simple, your customers are your revenue stream and should be treated as an key asset. Just my two cents.
  • Anonymous
    June 27, 2006
    If you look at this question through the lens of the "economics of specialization", I think you can see the answers more clearly. Consider a spectrum of IT outsourcing that goes from building your on computing hardware to leasing a dumb terminal (those used to be pretty much your only choices, now that I think about it).

    Nowadays, I don't think anybody needs to build their own hardware. Even if you have highly specialized needs (everybody say hi to the NSA guys), it's better to have somebody do that for you. I doubt there are very many organizations who even need to build their own operating systems, relational databases or network stacks. There are a few organizations that need to customize things that layer though (although mostly for security reasons I think).

    As soon as you get above that layer, things get very interesting in most organizations. Do you need a specialized accounting system? What about payroll and human resources? For most organizations the answer is clearly no, but for some organizations there will be good strategic reasons to "roll your own". At every level you have to ask, does the cost of maintaining this expertise deliver enough value to the core business? Do the risks of outsourcing outweigh the benefits? Can anybody do this job better than we can?
  • Anonymous
    June 28, 2006
    Hi John,

    I appreciate your post.  You ask a question that I find interesting: "does the cost of maintaining this expertise deliver enough value to the core business?"   That sentence is very hard to quantify, yet it looks like it could be quantified.

    First off, how do you measure the cost of maintaining not only an application, but the expertise that it represents?  

    Secondly, how do you measure the value it provides to the core business?  If this is truly a unique approach to the business problem, there is no yardstick by which to measure whether it is a better approach without intelligence on similar costs in other companies that took another approach.

    Third, you ask to compare the numbers and say "does [IT] deliver enough value..."  In other words, it is not a simple comparison.  There is a formula there for comparing value to cost.  What is that formula?

    These are the basic questions of application portfolio management.  They are three difficult questions to answer.  
  • Anonymous
    July 01, 2006
    I'd say that within my current firm, IT is definitely a source of strategic competitive advantage.  As such, some portions are certainly commodity.  Others are the bread and butter of what we do.  Ultimately, the practice of bringing all of this together efficiently and effectively is what brings about value for the firm.  Determining the value of IT is more about 'willingness to pay' for that efficiency and effectiveness than cost.  And that WTP should provide a good indicator as to how important a role IT (or a particular system) is to the enterprise.