Before you dismiss this blog because you think whether or not to outsource is an old question long settled in many businesses, consider this: outsourcing is making a dent in the BI environment. I’m talking about outsourcing analysis and generating insight, not just admin functions. The driver in many cases is the lack of in-house data scientists and/or analytical skills. Organizations believe both can be a critical distinction for success.
I posed the “O” word as a topic in one of my recent #BIWisdom tweetchats, as the tribe has been privy to outsourcing deals from every angle (as the buyer, the vendor, the consultant) and definitely have opinions. Like these, for instance:
- “I’m not a fan of outsourcing in general. Too much information bleed from an organization.”
- “I know outsourcing is in vogue and makes money for vendors; but, to me, BI and analytics are the ‘family jewels’ that you want to keep close.”
- “There are lots of data scientist firms out there to lend expertise to data discovery. Outsource the hard stuff short term.”
- “This gal says keep the high-value information assets analytics in house!”
- “I have seen this and have seen nothing tangible come back as the biz context is lost.”
- “I think outsourcing should only be for starting a proof of concept or a fringe scenario.”
- “Data science professionals are so expensive. There are companies that provide them as a service, which enables charge-back and controls.”
- “Outsourcing this service can cost a fortune and, at the end of the day, you’re left with less than if insourced (skills sets are gone).”
The discussion narrowed down to two camps: One side tweeted: “I’ve seen this: ‘Send us your data and we’ll send you some insights.’ For some, this makes sense. But are we really comfortable with sending so many gigs/teras of data to others for analysis?” The other camp’s opinion is reflected in this tweet: “But what about companies that don’t have the resources to do their own analysis?”
It’s an investment either way. A company can invest in hiring in-house resources or invest in using outsourced resources, and staying in business always requires investments.
I asked the #BIWisdom participants what factors they advise for decision making regarding outsourcing. First is the value (perceived or real) to the business. And of course the in-house vs. external skills quality and availability as well as data sensitivity are factors.
Budget is also a factor. And internal politics if the results will hold people accountable for policies or decisions based on the insights generated.
Most important is the issue of whether the development work and data models are transferred back to the buyer or remain with the third-party outsourcer. In some cases, the third-party uses proprietary software that makes the models not be exportable. A participant tweeted that “with more consulting companies creating software products, outsourced deliverables may exclude the technology and models.”
The goal is that the client should be self-sufficient afterwards, and the group tweeted strong opinions about how to ensure that goal:
- “Knowledge transfer should be written into the Statement of Work, or you will be left with a result set and no real learning.”
- “It’s a poorly worded outsourcing contract if the deliverable doesn’t include all the created assets.”
- “Active knowledge transfer needs to be factored into the price.”
Bottom line: Given the existing lack of adequate data scientists and the rapidly growing use of analytics, many companies must consider the advantages and risks of outsourcing part of their BI functions.
But if your organization decides to go this route, make sure you take this journey with a surer step. “Mistakes often point to areas in which we lack self-awareness of a weakness or threat,” wrote Robert Bruce Shaw in “Leadership Blindspots: How Successful Leaders Identify and Overcome the Weaknesses That Matter.” Decades of studies about outsourcing reveal that the biggest culprit in outsourcing failures is the buyer not stating in the contract exactly what it expects the outsourcing company to deliver. Make sure your contract specifies deliverables that include created assets and knowledge transfer.
Howard Dresner is president, founder and chief research officer at Dresner Advisory Services, LLC, an independent advisory firm. He is one of the foremost thought leaders in Business Intelligence and Performance Management, having coined the term “Business Intelligence” in 1989. He has published two books on the subject, The Performance Management Revolution — Business Results through Insight and Action, and Profiles in Performance — Business Intelligence Journeys and the Roadmap for Change. He hosts a weekly tweet chat (#BIWisdom) on Twitter each Friday. Prior to Dresner Advisory Services, Howard served as chief strategy officer at Hyperion Solutions and was a research fellow at Gartner, where he led its Business Intelligence research practice for 13 years.