Client
VERGLEICH CH – localsearch (a brand of Swisscom Directories AG)
Contractor
Panter AG
Year
2024
A successful Google Cloud migration for localsearch’s comparison platforms
localsearch, leading marketing and advertising partner for Swiss SMEs, operates over 50 B2B2C comparison platforms with over one million visitors per month. Due to the outdated architecture in a legacy data center, the company decided to migrate to the Google Cloud. The aim was not only to modernize the infrastructure, but also to make optimum use of the advantages of cloud services.
In order to minimize sources of error as far as possible, the project was divided into several phases with clearly defined control gates. In addition, several simulations were planned in order to be prepared for various scenarios. The migration was carried out during ongoing operations. It ran smoothly within the planned maintenance window and was completed successfully.
The migration also brought significant business benefits:
- Long-term savings in operating costs
- Increased efficiency thanks to a standardized infrastructure
- Improved working environment for software developers
About
localsearch is the leading marketing and advertising partner for Swiss SMEs and operates over 50 B2B2C comparison platforms. The localsearch product portfolio helps companies to be found online, attract new customers and retain them in the long term. With simple and effective solutions for online marketing, competent local advice and an attractive price-performance ratio, localsearch contributes to the success of Swiss SMEs in the digital world. The high-reach directory and booking platforms local.ch and search.ch, which reach 2.7 million people in Switzerland, are particularly well known. High availability with as few failures as possible is key for operators. After all, any downtime not only means dissatisfied visitors and customers, but also a direct loss of sales.
Initial situation
The platforms were operated in a legacy data center with a legacy architecture that was not based on common cloud standards. The imminent replacement of the existing environment required a rapid migration to a future-proof operating platform. The modernization of the architecture presented a strategic opportunity for further optimization. With this in mind, localsearch decided to migrate the on-premise solution to the Google Cloud. With a clear deadline and the goal of not simply carrying out a “lift-and-shift” migration, but rather making the most of the possibilities of the native cloud offerings, the project team faced a demanding challenge. The core business, the processing of thousands of customer inquiries per day, was to be affected as little as possible. In addition, due to some legacy platform functions, targeted architectural adjustments had to be made to enable the migration and make it future-proof.
Solution
In order to successfully complete the migration before replacing the existing environment, we opted for a multi-stage approach. This approach ensured that all risks were minimized.
- Analysis of the current situation (1): Before you can migrate a system, you need to know it very well. The first step was therefore to gather all the information about the current system. All findings about the target architecture, peripheral systems and dependencies were documented and served as the basis for further planning.
- Design of the solution architecture (2): Based on the analysis, we examined how the systems used could be transferred to the alternatives offered by Google. We followed a clear principle: wherever possible, we used standardized Google services without changing fundamental technologies within the application. Where no suitable alternatives could be found, we outlined which code and architecture adaptations were necessary.
- Migration from as-is to solution architecture (3): With a clear solution architecture, we developed a plan to carry out the migration with minimal disruption. We not only worked out the steps for the platform changeover in detail, but also planned the communication with the business and support team as well as the previous provider.
- Setting up the target environment (4): Once the design was complete, the new infrastructure was set up and put into operation. This ensured that several environments were available again, that all access to the peripheral systems worked smoothly and that the various user groups (Dev, Ops, Support) had the necessary authorizations.
- Application adjustments (5): The identified adjustments to the application architecture were implemented together with the developers and extensively tested and verified in the new test environment.
- Main rehearsal and test (6): All use cases that were part of the analysis were fully tested after the main rehearsal (including test data). Some minor bug fixes were identified that would otherwise only have been discovered during the productive migration.
- Implementation (7): For the actual migration, the procedure tested in the previous step was repeated exactly. All use cases were also fully tested again to ensure that the B2B2C platforms functioned correctly. After successful testing, the DNS entries were migrated to the new infrastructure.
- Conclusion (8): Finally, clean-up work that was recognized during the migration but not absolutely necessary for success was carried out.
The plan described above enabled the application to be migrated smoothly within the planned timeframe and without disrupting customers. The implementation was completed on time and even under budget. No outages occurred outside of the defined three-hour maintenance window.
”Thanks to our collaboration with Panter, we were able to successfully migrate a heavily used B2B2C online platform from an aging infrastructure to the Google Cloud on time, on budget and without any interruption to operations. Panter's experts carried out the planning, project approach and technical implementation support from our core team in an exemplary manner.
Gabriela QuadriSenior Product Manager Verticals localsearch
Technology
A before-and-after comparison shows that neither the access figures nor the response times or the number of support tickets have changed significantly. Thanks to a well thought-out maintenance window, customers did not notice any restrictions.
All legacy components have been completely overhauled and optimized for use with standard Google Cloud components. This has future-proofed the infrastructure and extended the lifecycle by several years.
To minimize potential sources of error, the project was divided into several phases with clearly defined control gates. In addition, several simulations were planned in order to be prepared for various scenarios. In the end, everything worked right from the start.
A multi-stage process with precise go/no-go decisions and careful testing of all components strengthened the confidence of all those involved in the smooth running of the migration. The database migration was secured by several test runs with copies of productive data. This ensured data integrity at all times. The switch to state-of-the-art cloud components also led to a further increase in security standards
Thanks to intensive testing and planning phases, everyone involved was able to start the productive migration with absolute certainty. The mix of flexible fallback scenarios, precise planning and comprehensive tests led to a pragmatic and confident implementation.
Flurin Capaul, VP Clients
Interested in working with us?
Contact Flurin and get support from a partner with many years of experience.
