These data points spanned multiple domains, including Location, Menu, Nutrition, Pricing, and Supply Chain. The specific data elements managed included:
- Assets: Images & PDFs
- Classifications: Menu Levels, Ingredients Folders, Business Units, Reporting Assets & Assets Library Folders
- Products: Business Units, Products, Ingredients, Sales Areas, Sales Channels & Buy Side / Sell Side Suppliers
Data Challenges
Mitchells & Butlers were already using Stibo Systems STEP and had a wealth of validated data. The primary challenge was to preserve this approved data while migrating to the new solution and extending the data model to include additional business areas, such as pubs.
A significant part of the challenge was integrating multiple external sources into a consolidated STEP system while ensuring data quality and consistency.
The migration project
To ensure a seamless migration, Amplifi structured the project with parallel teams: one dedicated to building the PIM and MDM systems and the other focused on managing the data migration. This division allowed the implementation team to concentrate fully on system development, while the migration team ensured accurate and efficient data transition to support these new systems. Whilst in this specific case both the implementation and migration teams were from Amplifi, we’re also well experienced liaising with internal teams, GSIs and other specialist consultancies during large-scale projects – read more about our experience of co-existence with other teams, here.
The migration project itself was divided into three main stages:
Stage 1: Source
In this stage, we identified and assessed the 10 distinct data sources, with Stibo Systems STEP being the primary repository. We meticulously labelled each piece of data, categorising it for clear migration mapping. A key development was the creation of a STEP XML Accelerator, designed to transform and output data into STEP XML files, facilitating the efficient loading of data into the new system.
Stage 2: Transformation
This stage focused on cleansing and transforming the data to align with the new target system’s model. Using ETL tools, we consolidated semi-structured data, removed inconsistencies, and mapped it to the new data model. Complex business rules were applied to ensure the data was ready for migration, with a strong emphasis on maintaining data quality throughout.
Stage 3: Target
The final stage involved executing a "big bang" migration over a single weekend, transferring all the transformed data into the new system. Extensive dry runs were conducted beforehand to identify and resolve potential issues. The migration was followed by thorough validation checks and documentation reviews, ensuring the new system was fully operational with minimal disruption. High data quality was maintained throughout, resulting in a smooth and successful transition.
Data Quality within the Migration Project
Data quality was a key focus throughout the migration, from discovery to cutover. As the client continued using their original data sources until fully operational on the target system, we maintained continuous monitoring and management of data quality.
We began with a comprehensive review of data sources to determine the migration scope. Rigorous data quality tests were conducted, and we collaborated with the client to manually verify results against business rules. Standard approaches were used to address issues in the source, migration process, or target system.
Mitchells & Butlers encountered complex data quality issues due to their intricate data model, particularly with recursive relationships in recipes and sub-recipes. Challenges included improperly linked ingredients and inaccurate nutritional values, leading to cascading data quality problems. Our team collaborated with the client to develop custom rules to ensure accurate data hierarchy relationships.
We also focused on aligning data quality with new business strategies, including the creation of new business rules to meet specific criteria. By proactively addressing potential issues with existing data attributes, we ensured a smooth transition to the new model.
The project results
Our iterative approach, combined with multiple layers of validation, was fundamental to the smooth execution of the migration. From the outset, our team prioritised meticulous planning and rigorous testing to ensure that every potential issue was identified and addressed well in advance. Several dry runs were conducted, each designed to simulate the migration process as closely as possible, allowing fine-tuning of strategy and ironing out any possible challenges before the final cutover.
It is important to highlight the key role client engagement plays in driving the success of a migration, alongside the expertise of the consulting team. Whilst the team provides a high level of skill, active collaboration from the client ensures the best results. Mitchells & Butlers migration achieved particularly great success due to their strong involvement throughout the testing and design phases, demonstrating the impact of close partnership.
As a result of this thorough preparation, the actual migration was executed flawlessly, with no data-related issues emerging post-migration. This seamless transition stands as a testament to the effectiveness of our team’s approach, marking the project as a model of a perfect migration. The absence of post-migration issues meant that the transition was so smooth that it went largely unnoticed - precisely the outcome desired for a data migration!
The value of working with Amplifi
Specialist Consultants: Amplifi’s expertise and methodical approach provided significant value to Mitchells & Butlers, not only ensuring the success of this migration but also establishing a foundation for future projects. Our team of seasoned consultants brought deep domain knowledge, effectively navigating the complexities of the data and business requirements for a smooth migration.
Validation First Approach: We employed a rigorous "validation first" methodology, conducting thorough testing and validation at each stage of the migration. This approach minimized the risk of errors, ensuring the integrity and readiness of the final migrated data.
Separate Migration Team: By deploying a dedicated migration team, we maintained high-quality standards without straining the implementation team. This clear division of labour allowed both teams to focus on their specific tasks, leading to an efficient and effective migration process.
Internal Coordination: Strong communication and coordination between our teams and those at Mitchells & Butlers were crucial. Our collaborative approach ensured alignment on project goals and timelines, allowing us to proactively address potential issues and adapt to challenges.
STEP XML Accelerator
In addition to the immediate benefits, the STEP XML Accelerator developed during this project has become a valuable tool for future clients. This innovation highlighted our forward-thinking approach to data migration, providing a robust solution that can be leveraged in subsequent projects, further demonstrating our commitment to excellence in data management.
Facing similar data challenges? Speak to our experts to discover how Amplifi can help you streamline your data migration or MDM strategy, and to learn more, download our guide As if by magic... The expert guide to Data Migration below.
