A new Research Management System (RMS) for your fund marks a fresh start for how you create, capture, store and share fund research from here on in, but the treatment of your past data is critical to its success.
If you’re moving to a new RMS in order to get away from the legacy approach of a central research repository (you know, the ones analysts use reluctantly as a data storage facility in order to check compliance boxes) then your migration to the new RMS needs a different approach too – one that means more than ‘getting your data in the new system’.
Without due care and attention to the structure, tagging, sources and data mapping of your existing information assets and your future workflow needs at the migration stage, your research (and your new RMS investment) can quickly lose its value.
In order to create a single, integrated research environment for all your data – one in which analysts work productively in and collaborate from – you need to not only preserve the quality and consistency of your existing research, but also improve it.
That’s true whether you require access to a decade’s worth of research data from your legacy RMS, disparate notes from a collection of consumer tools or information from your own proprietary systems.
Last year we ran this process for a number of clients transitioning onto the Bipsync platform, including migrations of Advent Tamale RMS and MackeyRMS, amongst others. Here are a few recent examples…
How We Migrated Over 60,000 Advent Tamale RMS Notes into Bipsync
One of our most challenging migrations to date involved bringing over ten years’ worth of research data from an on-premise deployment of Advent Tamale RMS into Bipsync. The scale of the data involved set this migration apart from previous projects; with so many notes and documents to migrate a one-off import could run the risk of content being missed.
To combat this we first iterated through a series of ‘mini-migrations’, gradually building up the size of the dataset from a few hundred notes, to a few thousand, and eventually beyond. After each iteration we audited the results of the import and took improvements through into the next run: tags were tweaked, automated tests were updated, new associations were discovered.
By starting small and only moving to larger sample sets once the accuracy of the imported data had been confirmed by the client we were able to arrive with confidence at a final import of over 50GB of data.
A MackeyRMS Migration over a Weekend
MackeyRMS’ ability to export its content in XML format has proven to be useful as we’ve found this a straightforward way to identify and map data for import in to Bipsync. For other vendors we sometimes need to spend additional time up-front reverse-engineering databases to identify what is available to us – without that worry here we can quicky set about making the data available in Bipsync so the client can begin the auditing process almost immediately.
In a recent example we were able to complete the migration process over a weekend. Analysts saved their content in Mackey before leaving the office on Friday night and began Monday with a personal onboarding session on their Bipsync installation pre-populated with all their research.
Getting Your Data into Bipsync: Our Migration Methodology
To get it right requires planning with agreed structure, scope and workflow defined and built into the migration process from the outset. With a number of significant research migrations under our belt, we have developed a clear and proven process to ensure a smooth transition, designed to maximize the value of all your research – old and new – in the Bipsync RMS and expedite user adoption. It goes likes this:
- Requirements Gathering / Discovery
- Define the scope and goals of the migration.
- Identify all sources and systems of data.
- Initial estimates for type, quantity and structure of research.
- Gather schemas, screenshots and explanations of existing data/systems.
- Co-ordinate dates and methods for transferring data, in phases where appropriate.
- Initial Assessment / Review
- Initial import data transferred.
- Profile data for inconsistencies, anomalies, duplicates and cleansing opportunities.
- Initial mappings produced.
- Migration design
- Review mappings and update.
- Update timescales/phasing based on findings.
- Agree on data subset to test.
- Review Subset
- Update mappings and migration.
- Review additional subsets if necessary.
- Full Import
- Final import data transferred.
- Re-profile to check for problems.
- Execute full migration.
- Quality Assurance
- Internal Quality Assurance test in Bipsync
- Fund Quality Assurance tests in Bipsync.
- Ongoing data maintenance / edge-case fixes where necessary.
- Old system retirement or parallel operation period.
If you’d like to learn more about migrating your fund’s research into Bipsync and how to get started, get in touch.