Over the years the LiftShift team has learned that the success of a data migration starts with asking the right questions. The right questions can uncover hidden challenges and mitigate project risk.
Consider reviewing the following questions with your migration team as you plan the next move between systems:
- What is the current system (legacy system) you are migrating from?
- What is the new system (dayforward system) you are migrating to?
- On a scale of 1 to 10, with 1 being bad and 10 being perfect, how would you rate your data quality in the legacy system?
- Who will be responsible for loading the extracted data?
- Will the dayforward software vendor provide proprietary tooling to import data?
- Will the dayforward software vendor provide the import data requirements? (i.e. fields, types, constraints, formats, database backup, csvs, etc.)
- Will you lose access to the legacy system data?
- When will you lose access to the legacy system data?
- When does the migration need to be completed by?
- How many databases does the legacy system use?
- What is the type of each database used by your legacy system? (i.e. Microsoft SQL Server (aka MSSQL or T/SQL), MySQL, Oracle (aka PL/SQL), MongoDB, Snowflake (ANSI), Google BigQuery, Cassandra, Mainframe (aka DB/2), a proprietary format, etc.)
- How large is each database in the legacy system?
- Is database information encrypted? If so, how is it encrypted? (Column level, row level, entire database, etc.)
- Is the legacy system database stored on-premise or in the cloud?
- Will the legacy vendor provide a backup of the database?
- How will the data be delivered? (i.e. AWS, Azure, etc.)
- How many databases does the legacy system use?
- What is the type of database used by your legacy system? (i.e. Microsoft SQL Server (aka MSSQL or T/SQL), MySQL, Oracle (aka PL/SQL), MongoDB, Snowflake (ANSI), Google BigQuery, Cassandra, Mainframe (aka DB/2), etc.)
- How large is each database in the legacy system?
- Is database information encrypted? How is it encrypted? (Column level, row level, entire database, etc.)
- Will you be migrating all legacy records or specific records?
- How are you deciding which specific records to migrate? (i.e. all records in the last two years)
- Do you need to retain the excluded records that are being left behind? (i.e. you are migrating the last 7 years of data and leaving the remaning data behind)
- What fields will the dayforward software support?
- What fields will no longer be supported that you use in your legacy system?
- Do you need to retain any unsupported legacy data?
- If necessary, how do you plan to retain unsupported or historical legacy data that is left behind? (see PDF Migration and Data Archival)
- What are the table names and row counts in each legacy database?
- What "objects" and how many "objects" will be migrated from the legacy system? ("objects" will be specific to your industry)
- Does the legacy system allow users to delete or purge records, and should deleted files be migrated?
- Does the legacy system store version history of the data, and does the version history data need to be migrated?
- Does the legacy system currently store files or attachments that need to be migrated? (see File Migrations - Asking the Right Questions)
- How are permissions determined and stored in your current system (database, Auth0, Okta, active directory, etc.), and what will they look like moving forward?
- Has your organization performed a migration from a previous system into your legacy system in the past?
- How will the development of the data migration occur? (i.e. by form, report, object, module, system, etc.)
- How will the validation of data be handled? Is there a limit on the number of rounds?
- How will the data be reconciled and what are the risks with this approach?
These questions will not cover the entire scope of your migration, but they will help uncover unique challenges, open further conversations, and can help mitigate unplanned risks.
Are you planning your upcoming data migration or stuck untangling a failed migration? Reach out directly to brendon@liftshift.io or schedule a free consult to learn how LiftShift can help.