Travel and Leisure (MDM) Client Case Study
Onepoint implements a MDM for global tourism group
The Client
- Our client is in the midst of a major Customer Transformation and Experience programme. They are looking to transform how front end agents in call centre and holiday resorts interact with their customers. Key to this transformation is to have a reliable source of accurate and good quality customer data.that can be easily accessed and distributed.
- Transform the business to a customer centric organization as opposed to transaction centric.
- Improve the customer experience across multiple touch points.
- Improve the quality and accuracy of the customer data used by business functions and processes.
Main Requirements
- Design and build a flexible, scalable and high performing Customer Master Data Management platform that can support multiple geographies and is a trusted source of quality customer data (data that is validated, cleansed, enriched and de-duplicated).
- The platform must be able to integrate with multiple sources of customer data and be able to distribute cleansed, enriched and deduplicated customer data to interested systems.
- Allow role based access to the ‘golden’ customer record.
- Empower the business stakeholders to control rules around validation, standardization, matching and survivorship.
Specifically our client wants to put in place a Customer Master Data Management solution that can act as a single source of truth for more than 100 millions customer records and can seamlessly integrate with systems that can provide customer data and with systems that are interested in receiving the cleansed customer data.
The Solution
In order to implement a performant and scalable system our solution comprised of:
- Mongo DB with a minimum of 3 nodes as the main MDM data store.
- ElasticSearch with a minimum of 3 nodes as a fast search database for detecting duplicates and potential duplicates, with a data model containing the searchable subset of customer information fields and prepared to use the supporting the Jaro Winkler algorithm.
- Multiple Camel route message driven services, and each of these routes is a part of the overall data pipeline – inter-connected by message queues – which receives data from the client CRM system, pushes the data to MDM and eventually merges customer records in the original CRM system.
[Route: A route is a process which gets triggered by messages typically coming from enterprise messaging systems, like e.g. Tibco Enterprise Message Services] - Multiple Talend ESB server for synchronous services.
- Onepoint architected and built the entire infrastructure on AWS (EU Region) for our clients based on stringent non-functional requirement and we are currently managing the entire application, data and infrastructure. The infrastructure comprises of a Production and number of non-Production auto-scaling environments.
Outcomes
- scalable customer master data management platform that can be extended to cover multiple geographies and can manage 100 million + customer records.
- A platform that can provide good quality customer data to systems and processes that need it.
- A single reliable source of truth for customer data that can be leveraged to support marketing and customer analytics.
- Confidence in the quality of customer data whilst interacting with customers.
