Off-Shore Outsourcing Best Practices

This paper was developed to provide general background to assist clients who are considering engaging an off-shore outsourcing partner, and wish to explore concerns and fully evaluate their options.
We should note that many of the matters discussed in this paper can be equally applicable to domestic outsourcers, as well.
Please note as well that this paper presents professional opinions intended to apply generally and that clients must take appropriate care to evaluate them in light of their specific needs. TBI makes no representations, warrantees or guarantees of any sort as to the applicability of the opinions presented in this paper to the specific needs of any client.

Overview – Best Practices

This paper is primarily concerned with off-shore outsourcing of services, and within that category, technology related and business-process related services. Accordingly, we do not take up here the very specialized concerns of other types of outsourcing, such as computer infrastructure outsourcing or manufacturing and distribution outsourcing.

Outsourcing Provides Organized Skills

Outsourcing is the contractual use of low-cost external suppliers to provide organized skills, Usually for business processes not regarded as core competencies by the contracting firm. Such skills typically include Information Technology (IT), financial and benefits transaction processing and accounting, manufacturing and various aspects of customer service, such as outbound and inbound telemarketing.

Off-shore outsourcing is simply the sourcing of such skills from non-domestic suppliers, usually at rates far below what one might expect to pay domestic suppliers.

The compelling interest behind off-shore outsourcing originally was and continues to be cost savings:

• Developing countries offer lower labor costs for highly educated, English-speaking professionals providing commodity IT and other business process skills; and
• Labor costs can range between 30 to 70 percent less than what would be paid domestically.

As the global IT outsourcing market has matured over the past twenty years, competitive pressures have led to increasingly professional and very high quality capabilities among off-shore providers. Many of them possess respected certifications attesting to high levels of professionalism and quality, including, among others, Carnegie Mellon’s Software Engineering Institute’s (SEI) Capability Maturity Model Integration (CMMI) certifications, which range from Level 1 (the most basic) to Level 5 (the most sophisticated); and the Swiss International Standards Organization’s (ISO) 9001:2000, which certifies attainment of very high quality standards. India has as many firms at CMMI Levels 4 and 5 as the U.S. does but of the approximately 250 firms that have achieved level 5 certification globally, most are in India.

Besides basic cost savings and high quality, other factors driving increasing acceptance of off-shore outsourcing include the following:

• Flexible staffing and basing models – full outsourcing (outsourcer takes over), co-sourcing (combination of outsourcer and client staff), staff augmentation (provision of specific, highly specialized skills), on-site in client’s domestic facilities, completely off-shore, a combination of on-site and off-shore, and near-site basing (basing of outsourcer’s staff at a location near but separate from client facilities);
• Ready availability of highly specialized skills without a need to commit long-term to individuals – promotes the ability to form and dismiss highly specialized teams quickly and cost-effectively; and
• A more intense focus by off-shore outsourcers than by domestic providers on developing technologies that make software development and implementation increasingly cost-effective.

While off-shore outsourcing is currently dominated by Indian firms, others will offer competition. But India will probably continue to dominate.

• A number of areas other than India currently offer off-shore outsourcing services – Eastern Europe (Czech Republic, Poland, Rumania), Russia, China and Southeast Asia (Philippines, Vietnam, Singapore)
• However, other than China and Russia, none of these alternatives offer India’s huge pool of indigenous talent with mature educational infrastructures for turning out large numbers of professionals annually.
• Russian and Chinese capacities, of course, while increasing in popularity, do not offer India’s advantage of an English-speaking professional class.
• Consequently, while Russian and Chinese capacities are expected to grow, and China in particular may evolve into a significant competitive threat to India, the smart money remains on India for dominance of the outsourcing market.
• Traditional U.S. based providers, including IBM, CSC, EDS, etc., are expected to open (some already have opened) their own off-shore facilities to rationalize their own cost structures and better compete with the Indians and others.

India currently possesses approximately eighty percent of the global IT market, while other nations split the remainder:

• Israel, Eastern Europe (Latvia, Czech Republic and Poland), and Russia account for most of the remaining twenty percent.
• Asia (Philippines, Vietnam, Singapore and to a lesser extent China), while showing promise, has not as yet made a serious dent in market share.

Of the types of off-shore outsourcing that are available, some are quite mature while others are less so.

• Greatest degree of maturity is in software application migration and maintenance as well as legacy applications support. All are substantial sources of revenue today, but legacy maintenance is declining as legacy environments are replaced; however, production support and maintenance of state-of-the-art environments is rising.
• More caution is shown buying new application development, such as mission-critical applications requiring high involvement by client personnel and greater business impact. While packaged applications (e.g., ERP, SCM, CRM) and business intelligence (e.g., data mining) are not currently mainstream offerings, they nevertheless offer substantial potential for the future.
• Least mature offerings currently generating the least revenue are call centers, business process operations (e.g., billing, claims processing) and technology infrastructure (data center services) support.

All that having been said, and despite the substantial benefits associated with off-shore outsourcing, risks exist to such use, including the following:

• Global political instability that could jeopardize key operations or key projects;
• Different labor laws, tax implications and legal frameworks;
• Operational factors that are difficult to control by clients, such as disaster recovery, voice/data speed/robustness, security policies and personnel attrition, to name a few;
• Cultural issues that hamper effective cross-communication, of particular concern as development tends to be conceptual in nature and therefore harder to define accurately or precisely – communication bottlenecks can occur among cross-functional teams, exacerbated by sheer distance (distance removes more than half of the visual cues we rely upon for effective communication), that are difficult to manage effectively;
• Technology used often is new and its full features and interactions are sometimes unknown or unclear to all participants; and
• In the U.S., increasing scrutiny by Congress and regulatory bodies on displacement of American workers, which speaks more to how outsourcing is delivered rather than its inescapable reality, today and in future.

In order to effectively manage the risks while exploiting the potential of off-shore outsourcing, the first challenge is to be aware of the potential traps, then to address them. We take these matters up at length, below.

High Valued-Added Capabilities

As IT and commodity business process costs continue to rise, many competitors are rationalizing such costs through outsourcing of commodity skills, presenting a competitive threat to those who do not.

• By the end of 2003, 70% of commercial enterprises adopted some form of outsourcing5, driven primarily by cost advantages. By 2004 its worldwide size was $163 billion, and $297 billion in 2007.
• The worldwide IT outsourcing market is estimated to grow to $450 billion by 2012.
• We believe that beyond this year and for the next 10 years, such enterprises will move toward a model that strictly distinguishes core competencies from commodity processing, and organize accordingly.
• The primary driver will be cost advantage with high quality. The strategic goal will be increased competitiveness while the tactical objective will be increased profitability.
• This model will have a far-reaching affect on barriers to competition as the winners, with rationalized cost structures, will dominate their markets. New entrants may be faced with requirements for organizational complexity that they are not prepared to provide, while mature competitors will scramble for years to catch up, losing market share along the way.

That’s the picture: we believe that Corporate Governance will evolve into management of a set of Core Competencies, High Value-Added Capabilities and Commodity Capabilities. Corporate Governance and Core Competencies will be internally managed, High Value-Added and Commodity Capabilities will be outsourced in a manner that dramatically reduces cost while protecting quality. Every company that wishes to remain competitive will need to organize this way.

Where will companies go for outsourcing partners? Wherever the best balance of low cost and high quality can be achieved reliably. Anywhere in the world. Indeed, as the trend evolves, the commodity exchanges we see gaining popularity today may well be established for such services as well, where specific jobs or even long-term responsibilities of any size will be posted globally on an exchange, and pre-vetted (by the exchange) vendors will bid on the jobs, allowing the client to choose among prices and other vendor and bid characteristics.

If this is the future, then we have seen above some of the competitive consequences of not embracing it early. But what are the implications if a company wishes to embrace the model now, and how can a company prepare itself for the future?

The implications are organizational – you need to evolve new control mechanisms to effectively manage outsourced work on a large scale. You prepare for the future by searching for partners who can serve you well in this transition, and by aligning with them.

There we have the premises: you will need to outsource in a big way, probably at least partly using off-shore suppliers; you will need to reorganize to do it effectively; and you will need to do it soon.

We take up how to go about doing this and what to avoid in the process in the remaining sections.

How To Proceed

In order to proceed responsibly with outsourcing, a company first needs a roadmap. What business processes should be targeted for eventual outsourcing? Which should be outsourced to off-shore providers, which to domestic providers? Which of those processes offer, if outsourced, the best mix of high payback and low risk? The first step in answering these questions is in determining who in the organization will answer them, how the effort will be organized and, if necessary, how it will be monitored.

Once these questions have been answered, the roadmap begins to take shape at its highest level – a set of targets in priority order, each with a direction for further analysis. At this point, it is advisable that the roadmap be socialized with key management, in order to determine general feasibility prior to performing more detailed analysis and preparation.

The further analysis deals with how to go about achieving the target, which calls for answering a new set of questions. How must the organization and the process change in order to achieve the target? What steps need to be taken, in what order, to effect the necessary changes? Which candidates should be considered for outsourcing? What should be the detailed objectives of outsourcing the target process? What should be the evaluation criteria driving the outsourcer selection? By what mechanism(s) will candidate selections be made? Once such answers have been developed, it makes sense to re-examine assumptions to determine if priorities have changed with more knowledge.

We would strongly advise performing this more detailed analysis only for the first few, highest priority process targets, as you will become better at this activity with experience, and as it is unlikely that you will attempt to achieve many of the targets concurrently.

Once these questions have been answered, the roadmap takes on substantial detail. At this point, the consultant assisting you to organize this effort or the internal people performing the activities themselves, should contact candidate outsourcers to begin discussions and to have them make initial presentations.

The next step is to prepare a Request For Proposal (RFP) for each of the target processes, which would be given to each candidate outsourcer for each target process. An RFP should clearly define the business process(es) to be outsourced, expected service levels and commitments and ceiling prices expected to be paid, as well as other, more general requirements such as financial viability of the candidate, etc. If the target is not a business process but a specific project of large scope, then such commitments would include considerations such as specific deliverables, methodology used to secure objectives, project duration, degree of permitted operational disruption, and other matters. If an actual process is to be outsourced on-going, other issues take on more importance, including modes of on-going internal and outsourcer interaction, the extent of ramp-up required by the outsourcer to familiarize him with the process, and other matters.

Each candidate outsourcer will provide a response to the RFP, which then needs to be analyzed, applying selection criteria previously agreed to, resulting in a tentative selection. This is followed by detailed negotiations to set the terms of a service contract. If the target is on-going outsourcing of a business process, then a Service Level Agreement (SLA) formalizing each party’s commitments, must be negotiated as well.

Roadmap Checklist

1. Establish organization for planning outsourcing analysis;
2. Define process targets, in priority order by business benefit as balanced by business risk;
3. Further analyze highest-priority targets to determine impact to the business, candidate outsourcers, mechanisms for evaluation and selection;
4. Validate assumptions and priorities;
5. Contact candidate outsourcers, initiate preliminary discussions;
6. Prepare and deliver RFPs;
7. Analyze responses, apply selection criteria, tentatively select the outsourcer or outsourcers;
8. Conduct detailed negotiations on price, commitments and other service levels, ending in formalizing of selection(s);
9. Formalize the Service Level Agreement (SLA) and other contracts, which formalizes the selection;
10. Begin the outsourcing; and
11. Measure the effectiveness of the SLA, taking remedial action as required in the event of non-compliance or unacceptable performance.

Once these matters have successfully completed, the selection is formalized. If the results of negotiations are unacceptable, then the next most attractive candidate may be tentatively selected and better results attempted.

The work of constructing an RFP and analyzing responses, not to mention negotiating agreements, is sufficiently complex and technical that it strongly suggests use of an experienced consultant to assist in the process.

Once agreements are in place, the actual outsourcing commences. Its effectiveness relative to the client’s expectations should be monitored continuously, and contracts and SLAs should reflect penalties, escalation mechanisms and contingent actions for performance that is regarded as sub-optimal.

Two points of particular importance deserve elaboration here:

• How a company should organize to effectively manage outsourced capabilities, and
• How such capabilities should be monitored.

Organization

If your interest is in a project rather than an outsourced business process, organization, while important to impose an effective span of control, need not change dramatically, except in how you communicate. However, if on-going outsourcing of a process is the objective, clients often underestimate the nature and extent of indicated organizational realignment.

Outsourcing a process rarely means that a partner takes over all aspects of the process. Even when the process is expected to be substantially outsourced, the partner must be managed by internal personnel accountable to management, who set direction and priorities and assure quality of service. And when the relationship is one of co-sourcing, in which client and outsourcer share responsibilities, the organization required to effectively manage the blending of roles needs to be robust.

When considering outsourcing requirements (when developing an RFP), take care to carefully distinguish commodity from high-value activities. The highest payback from outsourcing is to retain high-value activities closely related to overall success while shedding commodity activities that can better be performed by those who focus exclusively on them. So, part of the internal organization needs to take accountability for the high-value activities, and that part needs to be defined. Because high-value and commodity activities can be organized very differently prior to outsourcing (as in the case where both are performed by the same people), the change in an outsourced environment may need to be significant.

Management and monitoring are other areas that require attention. In a purely internal organization, the degree of required monitoring is often very different from that required in an outsourced environment. This is because those in an internal organization are presumed
to have aligned interests because they all answer to the same authority and share a company culture; so, monitoring of performance often is not rigorously performed.

However, in an outsourced environment the culture is not shared and the interests are not necessarily the same. The extent and intensity of indicated monitoring needs to rise. Trust but verify.

And management often needs to change in nature. In a purely internal organization, because interests and company culture are aligned, direction can often be collaborative, as can be priorities. Both can bubble up from the ranks since the ranks are intimately familiar with what needs to be done and the political and other constraints on what can be done. Such is rarely the case in an outsourced environment, particularly one with a large off-shore component, which will tend to be far more reactive. Management will need to drive the process rather than simply coordinate it.

So, understanding organizational implications is very important before commitments are made to a relationship and a partner, and those implications should be considered carefully in selecting the partner.

Monitoring

Monitoring begins with choosing the right metrics by which performance can be determined on-going. Such metrics will vary by organization. The best general rules are:

• Don’t make the number of metrics so numerous that you add significantly to cost or make the measurement environment too complex; and
• Define metrics that motivate the right behavior.

What are the most important performance factors to you? Reducing costs? Reducing defects in a programming environment? Increased production/productivity? Improved time-to-market? You will need to select a set that reflects your priorities while remaining within the control of the outsourcer. It should be a limited set that motivates the right behavior without imposing excessive and costly burdens of data collection and analysis, on you or on the outsourcer (who will, of course, pass the cost on to you).

But metrics need to be meaningfully motivational – measuring numbers of lines of code that a programmer generates will motivate a programmer to generate more code, not necessarily to produce elegant, flexible, easily modifiable code.

However, when developing an SLA, clients often make the mistake of defining how metrics are to be satisfied, along with the metrics themselves. You should leave the how to the outsourcer and focus on objective measurement of the results, which are important – the outsourcer will determine his methods in such a way that maximizes performance in achievable ways for his organization, because a successful relationship means a long-term one.

You should also choose metrics that can be easily measured. For example, measuring whether a newly written program meets published IT standards can require extensive manual review, and thus might not be an effective metric to implement. However, an off-the-
shelf metric analysis tool can automate the same process, securing the same motivational result at a fraction of the cost.

Once the metric set is established, the next step is to establish reasonable baselines for the metrics. They need to be achievable, although an SLA may require that performance relative to the baseline improve over time, with increased familiarity. Unless a great deal of relevant empirical data is available for the metric in your environment, be prepared to revisit the baseline standard of performance, perhaps frequently, until it becomes appropriately aggressive but still achievable with the right behavior.

For instance, in a software development environment, Function-Point Analysis is an often-used technique for estimating complexity of a task. A number of function-points are assigned a generic task, each point is assigned attributes such as effort required (person-hours), duration required, cost required, etc. You add up the points associated with all the tasks you contemplate and you have an estimate. You measure the actual experience against the estimate. The problem is that the assumptions underlying the number of assigned function-points and their attributes are almost always valid only for a stable organization (your organization, not all organizations), and even that assumption is challenged once personnel turnover begins to accelerate or systems are dramatically changed. Porting the assumptions to an outsourcer and measuring him on the basis of your experience may well create an unreasonable expectation (conversely, he may dramatically outperform your organization, which might create other types of problems and fail to motivate him to improve).

In general, the categories of metrics from which you might fashion a reasonable metric set to consider in an SLA include the following:

Volume of Work

Usually the most visible metric category in an outsourcing relationship, Volume is the exact level of effort to be provided by the outsourcer within the defined scope. Any effort expended outside of this scope usually requires separate billing or re-negotiation of the terms of the SLA. Broadly defined as the number of units of a work product or the number of deliverables produced per unit of time, Volume of Work metrics should be specified for every major deliverable cited in the SLA. Examples include number of outbound calls per day, number of maintenance requests per month, etc.

Quality of Work

Quality metrics are very diverse. Covering a wide range of deliverables, they seek to measure the conformance of those items to specifications or standards. When deliverables fail to meet the acceptance criteria in the specifications or standards, quality problems arise. Examples include:

• Defect Rates — number of errors in major deliverables.
• Standards Compliance — e.g., Internal standards for application source code, documentation, reports and other tangible deliverables, including number of enhancement tasks passing standards reviews, number of documented programs, etc.
• Technical Quality — Measurements of the technical quality of application code, normally produced by commercial tools that look at items such as program size, degree of structure, degree of complexity and coding defects.
• Service Availability — The amount of time/window of time that the services managed by the outsourcer are available, such as on-line availability of 99% during working hours.
• Service Satisfaction – extent of client satisfaction in service provision as measured by client surveys.,

Responsiveness

Such metrics measure the amount of time that it takes for an outsourcer to handle a client request. These are particularly visible and important from your perspective. They include:

• Time-to-Market or Time-to-Implement elapsed time from the original receipt of a request until the time when it is resolved.
• Time-to-Acknowledgement how responsive the outsourcer is by determining when a request is acknowledged, and accessibility of status information.
• Backlog Size such as the number of requests for service awaiting action during some specific time period, or the number of person-hours required to process the requests.

Efficiency

Such metrics measure success at providing services at a reasonable cost. However, be careful: don’t disrupt a desirable relationship between volume of work and delivery effectiveness just to measure something. For example, a commitment to process 1,000 telephone support requests per day for a fixed price of $ 10,000 per day may miss the point — if that is all that is measured, and if the outsourcer doubles his effectiveness, he still handles 1,000 calls and charges $ 10,000. Hardly optimal from your perspective. However, if you measure instead cost per call, you could see a drop in that metric from $.20 per call to $.10 per call, particularly if performance incentives are included.

Examples of efficiency metrics include:

• Cost/effort Efficiency — an efficiency metric, such as programs supported by one person or cost per support call, that can be sued to document cost reductions or productivity increases.
• Team Utilization – which measures workload of specific individuals in a team and aids in effective resource utilization.
• Rework Levels — a metric which measures the percentage of work products that are returned to a previous step for additional work, correction or completion, and which aids in assessing the quality and efficiency of a process by pointing out patterns of wasted time and effort.

Cautionaries

Commitment to a course of extensive outsourcing, particularly off-shore outsourcing where language, culture and methods may differ extensively from yours, is non-trivial and should be approached carefully. This section deals with the issue by presenting what we believe to be the primary risks associated with off-shore outsourcing, and our judgments as to how the risks may be effectively managed.

Inappropriate Speed

One of the most serious risks is proceeding too quickly into a commitment. With time and exposure come familiarity between the partners, which can avoid misunderstandings and the perception on the part of the client that outsourcing is too complex to succeed.

So, if initial interest lies in a specific project, the project should not be mission-critical but relatively low-risk and certainly manageable; and if the interest is in on-going outsourcing of business processes, the first process to be outsourced should not be too complex and it should not be mission-critical.

In this manner, you shake out the modes of interaction with little risk and build the mutual familiarity with each other’s culture and methods that will serve you well in more important undertakings, when the price of failure could be much higher.

Global Instability

India and Pakistan are nuclear powers and adversaries. Russia and a number of the more significant elements of the former Soviet Union also are nuclear powers, as is China, and all are under assault by Muslim and other minorities. Other potential partners also lie in unstable regions dealing with their own problems.

These are facts, and facts that are unlikely to change soon.

However, risks associated with instability can be minimized. If you align yourself with partners who possess domestic facilities, and negotiate commitments that key personnel will be moved to those facilities in the event of difficulties in order to keep service as transparent as possible, then much of the risk can be mitigated.

Different Labor Laws, Tax Implications and Legal Frameworks

Responsible firms protect themselves. While other judicial systems can be quite mature and not too dissimilar from ours, such as India’s, they all remain different, which could present problems in the event of disagreement.

Your legal representatives should review all contractual agreements with this in mind, and provision should be made in such contracts that any disagreements are to be adjudicated in domestic courts or mediated by domestic authorities. Your legal representatives should satisfy themselves that a potential partner can be effectively challenged in a domestic court.

Cultural Issues That Hamper Effective Communication

Different cultures value different characteristics and possess different sensitivities. Westerners, for example, anticipate pleasantly aggressive personalities and a high degree of candor, particularly among technical professionals, while other cultures emphasize less direct methods of communication. Unless anticipated, this can cause misunderstanding and jeopardize a relationship and expected outcomes.

And language difficulties are always a problem, even with Indians who employ English as a Lingua Franca allowing over one billion people with dozens of languages and dialects to communicate – and others for whom English is at best a secondary language learned in a school setting can present far more significant problems than Indians.

Relationships can (and often do) founder on misunderstandings, so particular attention needs to be paid by both partners to assuring that understanding is complete. Contractual terms, requirements, methods, deliverable forms and all other pertinent matters need to be discussed in detail and placed in context, so as to minimize chances of legitimate misunderstanding based on dissimilar perspectives and assumptions. While this is important when employing any service provider, such rigor is particularly important when visual cues, values, assumptions, even language, are different.

Operational Factors That Are Difficult To Control By Clients

Such factors include disaster recovery, voice and data speed and robustness, security policies, personnel attrition, and many other aspects of supporting infrastructure and how challenges are addressed internally that in turn can have profound effects on client service. They can also include the extent to which your own and your partner’s equipment platforms, operating systems, tools, etc. are in sync. And the sheer time difference (Eastern Time U.S. can differ from India by 10-11 hours) is a major operational consideration requiring careful thought before commitment.

Make a list of such issues. Discuss them candidly with potential partners, arrive at mutual agreements and formalize understandings in contracts and SLAs. Visit partner facilities as appropriate to monitor compliance and to communicate how seriously you regard the matter. Contracts and SLAs should have exit clauses and defined remedial actions that can be taken if compliance is unsatisfactory.

Increased Scrutiny by Regulatory Bodies

The U.S. Congress has recently taken more intensive interest in the matter of displacement of American workers, a trend that is unlikely to diminish.

And this issue is already of significant regulatory concern in other Western nations.

We believe that such concern will result not in diminishing exploitation of off-shore opportunities but in how such services are delivered.

• Those off-shore outsourcers who provide very large numbers of personnel to clients who require long-term on-site domestic presence are at greatest risk, particularly when they rely to an excessive degree on U.S. L-1 visas (which specifically prohibit displacement of American workers). This is an increasingly visible problem.

Off-shore providers who offer more balanced mixes of on-site and off-shore capabilities will be at less risk of damaging protective legislation.

Accordingly, interested companies should seek partners with methods that emphasize the off-shore component of any work force, and have operational models that can be shown to work with little on-site presence. Such emphasis also drives down cost dramatically.

• A concept that is gaining increasing currency is American intervention, what ThoughtWing calls Mediated Outsourcing Services?. In our model, an American firm mediates the outsourced relationship. In addition to providing an interpretive and quality assurance buffer between the client and outsourcer, we add high-value management consulting capabilities that usually are necessary but unavailable from outsourcers, who typically focus on commodity activities, such as programming.

We believe that dependence on such arrangements will avoid regulatory problems and, indeed, foster creation of domestic jobs while allowing exploitation of off-shore opportunities.

Conclusion

Off-shore outsourcing of projects and long-term maintenance and other service relationships is happening and will continue to happen for competitors; and it will need to happen for you as well, sooner or later, if you are to remain cost-competitive and if you resolve to focus on your core competencies. In order to successfully exploit the opportunities and effectively counter the competitive threats, you will need to re-organize to effectively manage the new paradigm.

Often, a company can put off such disruption until the threat is imminent. However, in this case, significant outsourcing suppliers may be forced to forego broad service to an entire industry in order to secure most-favored relationships with specific companies within industries, much as auditing firms currently are forced to choose between Pepsi and Coke. If such is the case, those companies who have not aligned themselves early may find themselves frozen out of the most stable and capable part of the industry. All companies should examine their strategic plans early, make efforts to understand how such a dramatic impact to ways of performing commodity activities will affect them, and take responsible action to protect themselves and their stakeholders.