IT companies and organizations moving to new systems often find themselves facing the need for data conversion. Before a system is replaced, teams must assess the downstream impact to other applications, and this can often identify the need to convert data within other applications. Depending on the quantity of data and the number of changes, data conversion can be extremely complex, and the project managers leading the initiative don’t always have the technical background needed to be successful. To simplify the process, we’ve outlined a four-step process project managers can leverage regardless of their technical know-how.

# 1. Prepare

Crafting a core team of essential data specialists, including redundancies in the team structure, is essential for success. Having a well-defined team provides peace of mind that every data set will be clean and consistent – two crucial parts of data management. Critical team members include 1) a primary and secondary technical point of contact (or developer) that understands the current and new application; 2) testers to validate the converted data and functionality; 3) enterprise and solution architects and database analysts to support the infrastructure; 4) a release manager to support environmental preparation activities (e.g., backups, code deployments); 5) business leads to provide guidance on the user impact; and 6) stakeholders accountable for decision-making.

Define Requirements

The next step is to define the data that will be converted and the ones that will not, and justify the whys and why nots along with the rules. Get stakeholders to buy off on the data chosen for conversion.

Identify Environments

Before production, define the process of data conversion before production, including migration paths and environment, and whether a code freeze (i.e., preventing any further changes to code) is necessary. The integrations can then be re-evaluated post-conversion if additional testing is needed.

  • • The environment used for practicing data conversion should be production-like in order to be most representative of the go-live experience.
  • • For multiple applications, multiple environments may need to be prepared (i.e., copy or export data on the same day and time prior to conversion).
  • • If environments are integrated, plan to isolate applications and turn off integrations prior to conversion.

Clean the Data

After reviewing the data, conversion and cleaning can begin. Removing manual errors and fixing mistakes before converting or analyzing data helps eliminate potential issues that may arise in the production phases. Make sure to review the data to be converted in the source application.

Define Control Points

Control points, or metrics for data conversion (e.g., account, customer or product identification numbers), should be clearly defined at project inception so that key stakeholders understand the use cases and intention behind data conversion. This includes identifying the control point names (and percentages) that will be converted, analyzing what data from the source is needed, and ensuring alignment with previously defined requirements.

Create a Project Plan

Project plans should include details for every step of the process and be able to roll up to the milestone level. The following formula designates critical elements to include in the project plan:

Tasks:

  • Environment preparation: Back-up environments, copy or export production environments, isolate applications, turn off integrations, and execute code deployment (as necessary).
  • Execute data conversion.
  • Output control points.
  • Test to validate data conversion.
  • Activate integrations (if applicable)
  • Provide downtime between each practice for reconciliation, reflection, and applying lessons learned.
  • Repeat the steps above and continue to refine the process based on the learnings gleaned following each conversation.

Helpful MS Project Plan fields and columns include:

  • Percentage complete
  • Task names
  • Duration of time (hours if applicable)
  • Baseline, actual start and finish dates and times (hours if applicable)
  • Predecessors and successors
  • Roles and resource names

# 2. Practice

Once the plan is in place, the fun can begin. It’s time to execute everything from start to finish. Before moving swiftly into execution mode, however, it’s essential to discuss the plan with key stakeholders to ensure that everyone is in lockstep on strategy. The following tips (in chronological order) will help ensure a successful data conversion process:

  • Confirm resource availability.
  • Coordinate location logistics (will the team be co-located or on multiple sites?) and any transportation needed for team members.
  • Plan for meals as necessary.
  • Have equipment backups ready to go (e.g., phone, laptop chargers, batteries).
  • Set up a conference call or screen share during the data conversion and execution of the plan.
  • Verify whether the team is ready.
  • Execute the plan.
  • Track and record the durations of tasks.
  • Keep an issue log and document problems as needed during the practice.
  • Ensure constant communication to ensure that everyone is aligned.
  • Update stakeholders with progress reports during the execution of the data conversion.

# 3. Reconcile and Reflect

Reconciliation and reflection allows stakeholders to make micro-adjustments along the way and ultimately ensure that future data conversions are executed successfully.

Reconciliation

Reconciliation allows the data conversion team to identify the data that was not converted, and why it wasn’t converted, after each practice.

  • Create a control point reconciliation process. This should be a repeatable process for each control point and each team/application. Include an action tracker that can be maintained to track the progress of details within each process step as needed.
  • Apply a timeline to the process to mitigate scope creep and ensure that reconciliation is completed in a timely manner. The goal is to estimate the number of working days based on process steps in order to provide a target resolution date for each control point by the team.
  • Create a control point dashboard and align it to the process.

Reflection

Reflection involves a holistic overview of the process. It involves assessing how the team felt about the data conversion – everything from their physical location to the meals provided. Share positive feedback, identify what went wrong, and brainstorm ways to improve.

# 4. Repeat

As you complete each of the predetermined number of practice runs, implement your learnings by updating the appropriate documentation (e.g., requirements, project plans) as you go.

Prepare, practice, reconcile and reflect, and repeat. Then you’ll soon be ready to execute the final plan and celebrate a job well done.