Industry Stories

Early Focus on Data – A Key Strategy for CIS Implementation Success

This blog post has been written with inputs from Parmarth Naswa, Principal Consultant, Utilities, Infosys.

From our long association with Customer Information System (CIS) transformation programs across multiple utilities, we have observed that most believe a successful CIS transformation is driven by adopting the right technology and business processes. They often overlook the criticality of data transformation in such programs. Those who have been part of complex CIS implementations will know that data is often the long pole in such programs, yet the focus shifts to data mainly during the design phase. This is too late in the program lifecycle and introduces significant risks to the program’s subsequent phases (testing coming to a halt because of lack of data) and overall success (incorrect bills being sent out to customers, post go live). An early focus on data, while helping mitigate risks, can also accelerate timelines and enhance quality. As we say, a new CIS system can only be as good as the data in it.

How do utilities ensure that their data is ready to play the pivotal role it must, towards successful CIS implementations? By focusing early in the program on all the aspects of data - from strategy and planning, to data quality and conversion.

Making data an early focus in your CIS program

  • The pre-program initiation Period – Data source identification and cleansing

Even before a System Integrator (SI) is selected for a CIS implementation, utility companies should start identifying its sources/systems of data that will be decommissioned. While major sources are often known like a large mainframe billing system, there are often smaller but significant source systems, maintained independently by business users in the form of access databases or excel sheets that hold important data and are required by the new CIS system. Inventorying, documenting and removing redundancies in data sources is the first step for an early focus.

Data cleansing is most effective when done early and directly at the source. Master data like customer and account, or meters and site are often the low hanging fruits for cleansing. For example, if there are email addresses missing for few hundred customers, an outreach campaign can help gather the missing data and update the source systems. Similarly, we often notice Personally Identifiable Information (PII) data gets stored in non-PII fields – like a driver’s license or social security number is saved in the notes fields by call centers. This data, if migrated as-is, can cause audit and compliance concerns in the new CIS system and therefore should be taken up for cleansing early in the process.

Data cleansing at source acts as a pilot for establishing the data cleansing methodology that can then be easily repeated for higher volumes of cleansing during the CIS program.

  • The pre-design period – Initiate data mapping, data conversion strategy, and early extraction

At the planning and analysis phase (pre-design), the CIS product and SI selection is complete and the target data model is known. With newer CIS products from major product vendors like SAP and Oracle, most data models and their foundational data needs are well defined. At this point utilities must start mapping the required data entities based on the target data model, without waiting for the detailed designs. This can free up the CIS program’s focus on business requirements, business process, and solution design. By now, the data cleansing methodology would have matured and a data quality metric framework can be established to monitor the quality and data stewards identified to own the data.

Finalizing the data conversion strategy during the pre-design phase makes sure that all the downstream work is well aligned and there is lesser conflict of activities post-design phase. The early conversion strategy gives enough time for other program workstreams like functional design, testing and infrastructure to plan their activities accordingly.

Setting up the data staging environment and performing a full extraction early in the pre-design phase help in determining the size of the staging environment and improving the extraction performance. Full extraction is often required to avoid any transformation stress on legacy systems. This early extraction provides a sandbox environment for the data team to profile for quality and test the cleansing programs before implementing it back in source.

Key benefits of focusing early on data

An early focus on data allows a lot of activities to be complete before the design phase and gives the CIS program time to focus on key areas. Identifying data quality issues at the beginning of a program, gives the business enough time to rectify them.

Early attention on data also allows accelerated readiness for integrated testing allowing the focus to shift to real issues rather worry about data quality or conversion issues. This also sets the stage for full cycle data mocks feeding into each integrated test cycle, user acceptance testing and other types of CIS testing.

Lastly the real benefit of data transformation is realized when the new CIS system goes live and enters production with minimum data related issues. All these benefits make for a compelling case for upfront investment on data in a CIS program.

To know more about how to build a smart utility, visit our microsite and meet our experts at CS Week in Phoenix from April 8-12, 2019.