Detailed and Deliberate Planning Was Key to Our Successful Migration to v7.1.2

[Luqmaanm1]

• Why did we upgrade?
• Who was involved in the process?
• How did we approach the upgrade?
• Phase 1: pre-migration/planning
• Phase 2: migration and execution
• Phase 3: post-migration
• What benefits have we experienced with v7.1.2?
• Some final words of advice for customers beginning their upgrade to v7.1.2 or v7.2

The automation team recently upgraded to SS&C Blue Prism v7.1.2. In this blog, Luqmaan Mohammed, head of RPA, shares tips from him and his team — including platform owner Amanda Tatana and production manager Yaqub Asmal — on how to execute a smooth upgrade.

Why did we upgrade?

Since we hadn’t upgraded or optimized our RPA software, we were experiencing performance issues on version 6.5. It was an old version, and we hadn't upgraded or optimized it for a long time. We wanted to increase the stability and performance of both the application and our infrastructure to reduce infrastructure costs and take advantage of new functionalities. We especially wanted to leverage the new browser extensions, the browser automation agent, access some new security patches and utilize the API support.

Who was involved in the process?

We had quite a big and supportive team. The core group was me, the head of RPA/business owner; Yaqub, the production manager; and Amanda, the platform and program manager. We also had a technical architect and a senior process controller who understood the production environment in the application well. From the IT side, we had database support from a database administrator, virtual desktop infrastructure support and senior staff from the delivery and automation teams. We also had great support from SS&C Blue Prism’s professional services team via their Product Upgrade Assurance (PUA) offering — Chris McGowan, Mike Lowe and Diane Prempeh were amazing.

How did we approach the upgrade?

The SS&C Blue Prism professional services team suggested that we divide our upgrade into three major phases: pre-migration and planning; migration and execution; and post-migration. Each of the large categories was then subdivided into specific milestones, with detailed individual checklists.

Phase 1: pre-migration/planning

We planned for the upgrade in a very detailed and deliberate manner. Planning was likely the most crucial phase — and it took us the longest amount of time; longer than we expected, to be honest. We did a lot of knowledge-sharing since it’s difficult to plan without context or without knowing what you're getting yourself into. This knowledge sharing set us up for success in the migration phase. If we hadn’t spent time planning, we probably would've spent more time trying to fix and remediate at the end. So, we took the time upfront to plan.

SS&C Blue Prism provided us with templates, like the Runbook, that helped us plan and execute the upgrade. In the end, we combined a lot of different templates and documents, which served as our guide. We designed gates for each phase, which had to be passed before moving to the next phase, and then we required a signoff for each phase from all the stakeholders. We aimed to have minimal impact on business processes and current automations and didn’t want to breach any service level agreements (SLAs).

Professional services gave us a great roadmap to success. We had specific stakeholders attend professional services sessions so that they understood the project phases, and so we were all following the same methodology.

Phase 2: migration and execution

Instead of doing an in-place upgrade, we opted to utilize new infrastructure and do a controlled migration into the new environment. We have 250 SS&C Blue Prism digital workers running 85 automations across 130 processes, so we needed to avoid any downtime and maintain the system’s stability. Even a few minutes of delay can compound to create hours of downtime. By moving to a new system, we started with a clean slate and migrated processes into the environment in small batches. Since it was a migration, we didn't have to provision for downtime because we were moving to a new environment. We had a built-in contingency plan because if anything unexpected occurred, we could revert to the old environment until the issue was resolved.

The actual migration was very smooth. We executed our plan, and we diligently tested according to the previous configuration. We worked through the SS&C Blue Prism checklists line-by-line. I cannot stress how important it is to go through that at every stage. That checklist saved us because there were things we had missed. It's a long process, because you literally have to go through every config and document that you've updated. But that's where the success came from — from those provided templates. It was the control for us, to say that we've signed off that phase, we've gone through the checklist and we’re happy.

Phase 3: post-migration

This part of the process was the smoothest. We simply monitored the new system and decommissioned the old program. We completed the entire process in three months. We didn't see any significant business process impact during the project because of our controlled methodology and solid contingency plan.

What benefits have we experienced with v7.1.2?

We’ve seen a significant improvement in the stability of the solution. We have fewer incidents and fewer errors in processes. We didn't anticipate that some processes would become more stable and successful because of the smaller optimizations. In the past, the solution would run, but from time to time we might get little errors in stages. Now, in addition to faster delivery due to an overall more stable system, the developers can work without interruption. They're not losing any progress or productivity because the solution is always available and there are no errors.

We’ve been able to cut costs as well. We reduced the infrastructure needed to run the solution — and because the application is optimized and running smoothly, we anticipate reducing the current infrastructure by half in the coming year, which will lead to additional cost savings. With the load balancing functionality — which dynamically spreads the load in the most efficient way — we’ve increased resiliency by spreading our infrastructure across two availability zones to meet both our high availability and disaster recovery requirements.

Some final words of advice for customers beginning their upgrade to v7.1.2 or v7.2

Trying to upgrade an application within the constraints of time and cost can be a daunting task. Before embarking on this journey, the outcome, or what you want to achieve, should be clear. Take the time to plan to get there. If you plan adequately, the execution becomes second nature. And by no means will this be without without issues or delays. Learn from them and use them to your advantage by building safety measures so they don’t happen again.

There is no single recipe for an upgrade but stick to what works for your organization and be sure to take everyone along the journey with you.