The real cost of offshore outsourcing

THERE APPEARS to be a general assumption that dramatic savings can be made by sending software development to countries where IT salaries are 10% or 20% of similar IT professionals in the USA or Europe. That assumption is quite invalid according to those who have tried it, in fact they claim that savings rarely reach 30% and it can take years before even that level of savings materialise. According to them, a multitude of other costs, beyond the obvious cost of salaries, are often forgotten when considering sending work offshore.

In this article I want to outline some of these forgotten costs so that we have a better understanding of the complete picture. Of course many of the costs described here will vary between projects. Any given situation might include costs which are not mentioned here and conversely, because some of these costs might apply to mutually exclusive situations, some can probably be omitted when considering a specific project. On the whole however, this article should go some way to explaining why the anticipated savings fail to materialise.

Specifying the project
From the very beginning, the move to offshore outsourcing will incur a number of costs. The effort to prepare a project specification not trivial by any means because the document must be sufficiently detailed for the outsourcing provider to clearly understand the requirements.

In all probability it will be necessary to provide detailed information about the interfacing of the software to other software, to existing file or database structures and often the screen layout for the user. This kind of detailed work takes time and so the costs are either lost opportunities for more-productive by one's staff or those costs associated with additional IT professionals to actually perform this work.

Selecting an outsourcing provider
Once the specification is suitably developed it may be sent to several alternative outsourcing providers in order to obtain their responses and proposals. When these are received it will be necessary to spend some time evaluating those responses and negotiating a contract. Various staff members from both IT and the rest of the company may be involved with this evaluation. At some point during the preparation and negotiation of the contract it will be necessary to talk to legal professionals.

A report here talks of a project where evaluating the outsourcing vendors took five months to complete, took 500 hours in total and cost $20,000 on top of normal salaries for a CIO and three senior managers who all worked part time on this evaluation.

By this time not one line of code has been written and yet significant costs have accrued. Just as importantly time has moved on and opportunities may have been missed or requirements been modified.

Travel
Before signing the contract and probably several times during the contract period the client will want to visit the vendor for face-to-face meetings and to reassure themselves that all is well.

Travel to somewhere half-way around the world is both time-consuming and expensive. Once there, the client's staff reportedly work at less than optimum levels for a variety of reasons and yet their frequent presence there is seen as highly desirable in order to monitor the progress of the project.

Transfer of work
If the new work is to integrate with or somehow modify existing code then it is more than likely that there will be costs involved with educating the vendor's staff about those existing applications. This might seem excessive but those who have tried to do otherwise have sometimes found that the cost and difficulties of correcting misunderstandings is greater than the costs of ensuring that the vendor's people know what they are dealing with.

The most practical method of doing this is for some of the client's people to visit the vendor, or vice versa. This visit can take anything from a few days up to several months depending on the complexity of the situation and the relevant staff of both companies have to be paid during that time. Again this incurs costs with no immediate return on that investment.

Vendor's costs
The vendor's costs can be a major hidden expense in these outsourcing projects and unless payment for these clearly defined in the contract these costs can come back to haunt the client.

Experienced offshore clients talking of having to pay for essential infrastructure like networks and high-speed communication, or perhaps for hardware, software and licences which are compatible with the platforms on which the final code will be executed.

If there are delays with the provision of any of these items it will usually mean even greater costs for the clients.

Communication
The direct and indirect costs of communication with an offshore outsourcer can run quite high. Firstly there are the direct charges associated with frequent phone calls and any permanent leased lines for data networks or voice.

The difference between the time zones adds indirect costs to the equation because one cannot expect staff to make regular contact on business matters far outside normal business hours and not be compensated for doing so. Even a time difference of less than eight hours may not be ideal because there may still be some period of the day for which communication is difficult or at least unreasonable on a regular basis.

Software Testing
It is fundamental that if work is being done by another party then the testing of the resultant code has to be very thorough in order to ensure that it is compliance with the contract. With in-house software developers one has the ability to quickly call upon them to fix any bugs that may be discovered but this task is far more difficult when dealing with an outsourcer, even more so if the contract has been fully paid.

Testing will require the involvement of various people within a business or organisation, particularly the users and some local IT professionals who can probably better describe the faults to the outsourcing vendor's people. All of this requires time and effort, and so this testing incurs greater costs than would typically exist for an in-house development.

Costs of Failure
Many potential clients of offshore outsourcing give little thought to the consequences of failure and the costs of correction. The problem is that offshore projects do sometimes fail to deliver on their promises but admittedly the extent is difficult to gauge because few companies publicise the incidents.

In the last six months I have heard of a simple data-entry job that was so poorly done offshore that it needed to be redone by local workers. In another case I have heard a strong rumour that an Australian bank was badly burnt by offshore outsourcing and has repatriated future coding work.

Another report spoke of two US companies that are generating good income from correcting bugs in the code produced by outsourcing vendors. In some cases outsourcing companies have produced code with glaring errors in routines that were probably never tested in the rush to meet a deadline.

All of this corrective work comes with a price-tag and in some cases the total cost exceeds that of having the work done locally in the first place.

Other Factors
There are a variety of political and geophysical risks that may also affect the vendor's ability to deliver according to the terms of the contract. For some there is a chance of war or some form of conflict and for others the threat may be from earthquake or climatic factors such as hurricanes. Larger client companies may wish to deal with a number of outsourcing providers who are not exposed to the same risks but this incurs the costs of each separate outsourcing contract. Smaller client companies are left only with the option of trying to arrange insurance to cover any problems that might be experienced.

A cost that is usually hidden in the contract but is ultimately paid by the client is the cost or partial cost of any local representatives of the outsourcing vendor. The extent of this payment will depend on whether the vendor has people permanently on the client's site or simply operates a local office to administer existing contracts and seek out new work. In the case of a small vendor the entire cost of a local office or representative could directly or indirectly fall back onto the client company.

In-house costs
If an organisation is engaging an outsourcing vendor to replace employees then the costs associated with this will depend on the outsourcing arrangement.

Where the outsourcing vendor is taking over the local facilities, the common practice is to transfer those employees to the vendor and provide them with guaranteed work roles for some agreed period of time. For this situation the original employer usually incurs relatively minor costs.

The situation of transferring jobs offshore is an entirely different matter because it will typically involve some kind of retrenchment payment to each employee who is being displaced. Of course the morale of displaced workers will usually fall and it may be necessary to pay them some kind of retention bonus if they are to share their knowledge with their replacements.

It is entirely plausible that the total paid to these employees will be equivalent to a sizable proportion of the money paid to the outsourcing vendor and that any financial benefits of the outsourcing will take some time to appear.

When any organisation engages an outsourcing vendor they will need to ensure that the terms of the contract are being followed and that any vendor's requests for further information are promptly dealt with. Staff with a mixture of business and IT skills are required for this work and while it might be tempting to re-assign employees that are being displaced, they will often not have the necessary skills to immediately step into these roles.

Intangible Costs
With any outsourcing arrangement there are intangible costs that "bean-counters" tend to ignore because they cannot be quantified. All too often these are matters that can effect the operations of the client company and any proper analysis should attempt to take them into account.

The most obvious example is a loss of in-house knowledge about the relationship of the business to software and IT operations. There may be a loss of information about dependencies and why certain IT functions operate in a particular manner, or what to do when certain circumstances arise, such as the failure of some component or overload in another.

The other problem with knowledge relates to the understanding of the software. In an outsourcing scenario there are certainly no guarantees that the vendor will still have good knowledge of the application in a few years time because software engineers will probably have moved to other employment. There is a better chance that in-house knowledge would continue to exist and that either the software engineers or their documentation can be located, or perhaps other people who are familiar with the code. Certainly guarantees cannot be made about this in-house knowledge but to a large extent such a matter is under the control of the employer.

Outsourcing will inevitably introduce delays when changes are required for whatever reason. These changes might occur during software development when it is difficult enough to incorporate them into the outsourcing contract, or they might occur when the contract is complete and demand even more effort and greater delay in their implementation.

IT might not be core business to many organisations but it is the lifeblood of any modern company and any disruption can have very serious consequences. If there is a conflict in the priorities of the client and the vendor it might have serious consequences for the client especially when the vendor is only prepared to match that priority by increased charges.

Finally there is a risk of data and software security. There are reports of software modifications going offshore and a short time later a competing and very similar product appearing from that same country. This might be an extreme case but there are no guarantees about loss of data or even of data security and the cost of its loss may be quite substantial.

Conclusion
All up, offshore outsourcing is rarely as beneficial as one may be led to believe. There are obvious savings to be made on salaries but the full picture includes a variety of costs which can make a very significant difference to its effectiveness. Certainly the costs will vary between projects but even for short-term work of less than 12 months many of these costs will apply to some extent. As well as the tangible costs there are the intangible costs for issues that will often have a negative impact on the client's business and the whole picture needs to be viewed somewhat more realistically than it has to date. µ

L'INQS
How Offshore Outsourcing Failed Us
There's More to Consider Than Cheap Labor