Organizations endeavoring to architect and implement a data lake program are moving in the right direction, as they know data lakes are generally less expensive than traditional platforms and allow a deeper level of analytics for driving new insights. No matter how well intended, though, companies often set themselves up for failure by not including their number one internal customer—the business.
Here’s the challenge: Data lake programs are generally led by IT teams, leaving business leaders disengaged until post architecture design and initial execution. This “build it and they will come” mentality may prompt business users to challenge justification for incremental investment in big data and, more importantly, wonder what’s in it for them in terms of real business value. While it remains important, re-platforming existing business intelligence reports and ETL offload alone do not lead to long term adoption of a data lake. Organizations are seeking more concrete and measurable reasons, like the need for an environment from which to drive new insights and improved analytics with expanded data—all on a less expensive platform.
Inviting Business to the Table
As with any major IT initiative, cost-savings underpins the start of many data lake projects. However, the larger return on investment is ultimately realized in the potential it offers for real business value. For this reason, business should be invited to the conversation early on in development—and arm IT with use cases that have clearly defined success criteria.
A successful data lake design should be built around measurable and anticipated business outcomes, and ultimately answer the question, “what value are you trying to drive?”. For example, are you seeking to improve efficiencies, gain additional customer insight, evaluate marketing assets, or mitigate downtime due to equipment failure? No matter what your unique business goals, they should be directly tethered to the foundation and development of your data lake venture. Otherwise, you’ll spend a lot of time and effort; but in the end, no one will be equipped to use or understand what they can get out of it.
Proof of Value
As with any IT initiative that leverages new technologies, a data lake goes through the typical proof-of-concept stage to evaluate tools and finalize architectures. However, demonstrating proof of value—and repeatable value—is a necessary early milestone for long-term success of a data lake. The ability to demonstrate new value that a data lake can deliver, and repeating results, encourages business groups to evaluate and leverage it for even more use cases. Engaging business users to deliver early success stories creates the internal demand a data lake platform needs for sustained growth.
If business is on board, investment can be shared—and that’s important. Sharing the investment in strategy with IT and business, along with the expenses and resources, helps ensure greater and more meaningful engagement between the builders and users of the data lake. In the end, the goal is to create an environment where business analysts and data scientists can effectively mine information—leveraging both internal and external data—to drive efficiencies and valuable insights.
More importantly, growth and expansion of the data lake becomes a shared goal—where business demands more data and faster access, and the expected incremental value becomes the justification for the investment. And let’s face it, when everyone’s on board the faster and more intentionally development can proceed, and the quicker your organization can realize proof of value.