Data Warehousing - Offshore or On?

What is the right answer?  Should you use an onshore or offshore vendor for your next data warehousing project and which vendor is really cheaper?  There is not a one size fits all answer and there are a number of factors to consider when trying to decide which model to use.  Unfortunately, more often than not, the decision is made using a single metric – hourly rate.  It is true that using an offshore vendor will give you significantly lower hourly rates, which can look very attractive through the procurement lens, but cost savings are not always what they seem.  Lower rates do not always translate to lower total project cost. 

When evaluating the use of an offshore vendor for a data warehousing project it is important to consider the following:

• What level of productivity will you receive for each resource?  In many cases, you can expect to receive less skilled resources that have lower overall productivity, increasing total cost
• What are the hardware and software costs associated with each resource?  More lower skilled resources means more hardware and software costs.  If the development software being used is licensed on a per seat basis, this can have a significant impact to the bottom line cost of the project
• Do you have clear requirements?  If not, the communication gaps and distance of the offshore provider can create significant challenges
• How much institutional knowledge will be lost by offshoring the project?  Is that knowledge critical to the success of your business?  Is there a transition plan in place to transfer knowledge back to the organization when the project is complete?
• Do you have the right data security and data protection infrastructure in place to support offshore development?  Does the offshore facility also have this infrastructure in place?

In many cases the answers to the above questions leave you with good reasons to use an offshore model and when used correctly it can deliver significant cost savings.  Offshore development provides staffing flexibility and many offshore models are based on units of work that can flex up or down as needed giving you as much or as little bandwidth as you need.  If you have a clearly defined project with detailed requirements that can be easily handed off, the offshore model becomes very attractive.

But what happens when requirements are changing, you’re not sure what you want and you need something done quickly.  If you have worked in the data warehousing environment, you know this is the norm.  As data is consumed, new insights are found and requirements change, creating a moving requirements target that is difficult to hand to an offshore development team.  Changing requirements and ambiguity are like kryptonite to an off shore team, it requires a level of agility and adaptability that stresses the offshore model. 

In these cases it makes sense to move away from the offshore model, but the alternative doesn’t have to be flying in high dollar consultants weekly from around the country to deliver your solution.  Costs can be lowered and off shore risk mitigated by using an off site but onshore consulting team.  With this model, severe time zone differences and communication challenges are mitigated, and if resources need to be on site for short periods of time, it could be as simple as a short drive or a last minute flight.  In this model, the team can be sourced from regional firms that have highly talented resources with lower cost structures than the national consultancies.  When done properly an onshore model can result in a highly dynamic, flexible and easy to work with delivery team that can handle rapidly changing requirements.  At first glance the onshore cost may look higher than the offshore estimate, but when you factor in the quality delivered, the reduced amount of rework and the flexibility provided the costs become at least comparable and potentially lower.