Data is an integral part of every organization, and data conversion from legacy to new Enterprise Resource Planning (ERP) is a critical success factor for any ERP deployment. Organizations aim at rapid ERP deployment to reduce time to market. Concepts such as Agile, iterative conference room pilots, developments, and automated release management are being adopted based on an organization’s size, business characteristics, and readiness to adopt.
Although these concepts are successfully used in certain pockets in ERP, they have not yet become the norm in the ERP world. Some organizations have adapted to Agile, but data conversion often seems to be a hindrance in rapid deployment. Some of the challenges the conversion team faces are the unavailability of an end-to-end solution in the early stages; master data cleanup; and close collaboration among the business, IT, and the implementation team.
This article suggests an approach for handling data conversion using Agile methods (Scrum-based) for a rapid and risk-free ERP deployment.
Data conversion challenges
Data is the key enabler for responding to changing business needs. During an Enterprise Resource Planning (ERP) deployment, legacy data conversion to the new is also key to success.
Below are some of the challenges faced during data conversions.
The main activities in an end-to-end data conversion cycle are mapping, cleaning, transformation, loading, and validation. The primary owners of these activities, in general, are from different groups, such as business, IT, and implementation partners. This creates a need for extreme collaboration among these groups to avoid last-minute surprises. A gap is often observed in achieving this much-needed collaboration.
The data conversion plan is typically tightly coupled with the overall ERP implementation plan. Traditionally, data conversion planning activities start after requirements gathering and the completion of a high-level solution design. A late start of data conversion activities leaves little room or contingency for managing requirements changes.
Underestimation of effort
Importance and urgency in data cleaning is sometimes missing or not understood enough by the business because they are not part of the core team. It is assumed that data is in the right shape and that there is no need to clean it. Based on the underestimation of effort and time needed for data cleaning, these activities are planned in the last stages.
Business validation of converted data is an important activity. Any major gaps identified during the validation stage cause risk of a delay in the implementation timeline.
Focus on solutions
The ERP implementation team frequently gets caught in the uniqueness of the implementation and loses the focus it needs to use out-of-the-box (OOB) tools/solutions for automation and reuse.
Below are steps to convert data by applying Agile principles. These steps are based on the Agile Manifesto and best practices.
1. Start on day one
There is a myth that data conversion cycles are linked to requirements-gathering and solution-design activities. During the preplanning stage, we need a focused approach for a possible early start of data conversion activities.
- Kick start with a clear definition of objectives, scope, and a RASCI (Responsibility, Accountability, Conform, and Inform) matrix.
- Form a cross-functional team, independent of the rest of the implementation tracks, with buy-in from key stakeholders.
- Prioritize the product backlog to help identify risks ahead in the cycle.
- Focus on data conversion in standard ERP entity formats in the early sprints, followed by incremental changes during the solution-definition phase.
2. Leverage OOB solution/tools
Conversion teams believe that their data is unique, so they have reservations about using a standard OOB solution. An Agile approach will help jump-start the data conversion activities by using OOB solution/tools in standard formats and by incrementally implementing the delta solution requirements.
- Identify the best suitable OOB tool/solution for data conversion. A good tool should be able to support all data conversion activities, such as mapping, sequencing, loading, and high-volume data management.
- Start the data conversion in standard ERP entity formats in the early sprints, followed by incremental changes during the solution-definition phase.
- Train the team to get comfortable with the tool and maximize its use. The team should automate as many activities as possible to reduce human error and time in subsequent cycles.
- Ensure tool usage for all conversion activities, such as data mapping, order sequencing, and data loads.
3. Form one team
Conversion requires the involvement of stakeholders from various functions; hence, it is important to ensure close collaboration. It has been observed that even after having representation from different groups, the team is challenged with complex, multiway communication requirements that lead to inefficiencies. The team works on assigned conversion activities in addition to existing responsibilities, leading to the deprioritization of conversion tasks. Formation of one
team in the Agile method helps to address these pain points. It will also help reduce rework and effort needed from key business resources in later stages.
- Identify key functions, team members, and buy-in from leadership on the team’s involvement.
- Clearly define roles, responsibilities, and effort required from each member.
- Ensure that the team is trained in the Agile framework and required tools to support self-managed teams.
4. Find a solution for de-risking using an incremental and iterative approach
Solutions that impact conversions typically get finalized after the first testing cycle. This leads to a rework on conversion programs. Because changes are inevitable, and they are received in the last stage, there is minimum float left to absorb the impact. These changes are normally not welcome because of the risk of delay to the overall timelines.
Nonfunctional requirements (NFRs) are overlooked to meet delivery timelines, resulting in data issues during and after the company goes live with a global implementation.
- Prioritize the conversion backlog, including NFRs, to help identify risks ahead.
- Focus on data conversion in standard ERP entity formats in early sprints, followed by incremental changes during the solution-definition phase.
- Automate testing to incorporate changes faster and reduce the conversion cycle.
- De-risk the data conversion activity by starting early.
- Focus your approach with an independent cross-functional team.
- Reduce communication overhead with buy-in from all stakeholders.
- Reduce overall conversion cycle time and human error through use of a standard OOB solution in conjunction with other tools and automation.
- Deliver higher business value early by prioritizing the conversion backlog.
- Reduce data issues during and post-go-live using Agile methods.
Data conversion plays a vital role in ERP implementation. The adoption of the Agile framework can help with a risk-free, on-time deployment. Jump-start this critical track by starting the conversion activities early. Carving it out as an independent component and then building it in incremental, iterative fashion helps. Also, getting early buy-in from stakeholders to adopt the Agile approach, ensuring training of team on the required tool set, and focusing on automation helps the team achieve the set objectives.