[Webinar Highlights] From Pilot To Full Scale RPA Implementation: The Raymond James' Success Story

On our most recent webinar "From Pilot To Full Scale RPA Implementation: The Raymond James' Success Story" we had the pleasure to host a guest presenter, Kemal Kvakic, the RPA head & CoE chair of the Raymond James organization, a leading global financial services company.

Below are the highlights from the webinar, we highly recommend watching the webinar recording to get the full context of the summarized content in this blog.

Raymond James RPA program overview

Kemal shared an overview of Raymond James' global RPA program, which currently has 100+bots in production, producing 20-25 new bots every 8-10 weeks with the support of a dedicated operations team and a hybrid CoE (Center of Excellence) that manages the infrastructure, SDLC (Software Development Life Cyle), implementation frameworks, entitlements, access, etc.

RPA vendor selection

He also walked through the different aspects they evaluated during their RPA vendor selection process, including pricing, features, training/documentation, active community and academy resources as their main criteria. Which resulted in the selection of UiPath as their main RPA platform.

Infrastructure

There are many criteria that goes into the infrastructure design and hardware selection to support the RPA deployment, Raymond James used virtual machines to host the bots initially, but switched to Windows servers later based on their on-going needs. Kemal reviewed the different technical aspects of infrastructure design which included:

  • Virtual Machine vs Windows Servers
  • Dual or Single Orchestrator
  • Password Vault
  • CyberArk
  • Monitoring / Dashboards
  • SPLUNK (He mentioned that UiPath now has an integrated analytics feature that is very robust and new users should consider utilizing that instead of an external analytics, monitoring, and dashboard app)
  • Disaster Recovery
 
Picture3

Incident Remediation

Raymond James created a very comprehensive 'operations guide' to help their command center re-mediate RPA related incidents, with process guidelines to follow for a wide array of scenarios. (Please refer to min 0:20 of the recording for details on this subject)

WIKI & Documentation

Kemal shares the different Wikis and documentation (Please refer to min 0:22 of the recording.) they created for their RPA program that is offered across the organization to ensure users have access to all the information needed to understand and troubleshoot RPA related topics, including:

Guiding Principles

  • Think Time

Best Practices

  • Performance Testing
  • Peer Reviews

Coding Standards

  • Naming Standards and Annotations
  • Retry Logic and “Network Blips”

Framework

  • With Orchestrator Queues and Without

Entitlements and Access

This is one of the major hurdles they needed to overcome, as BOTs can not have the same type of profile access as humans as they can't have the same HR accounts, with benefits, vacations, health insurance, performance reviews, etc. They chose to use service accounts that will allow them to have access to data but nothing else. This led to exception handling, governance structure, and now they are in the process to creating robotic accounts that will help restrict more the BOTs access and help with audits and compliance for reviewing work performed by BOTs vs humans.

He also shared issues they ran into the security, licensing access, testing, etc. Please refer to min 0:26 of the recording.

Incident avoidance

To protect the organization from having incidents in production, Raymond James have structured their environment to make sure changes/patches made to different applications and software that affects the BOTs do not go into effect without proper testing, this includes setting up triggers and notifications for App owners and making sure regression testing is performed before anything goes live into production, please refer to min 0:34 of the recording for more details.

SDLC

The software development life cycle for Raymond James BOTs production can be summarized this way:

-8-10 weeks – Reqs to Go Live

-4 teams – 4 processes

  • 5 Developers
  • 2 Technical BA’s
  • PM
  • Manager

-140-150 min/major tasks

  • Naggy Nancy – BOT

-QA and UAT cycle

Picture4

Center of Excellence & RPA across the organization

Kemal considers grassroots engagement to be the number 1 aspect to ensure the success of a RPA program, making sure that everyone in the organization will support the implementation program including HR, App Admins, App owners, release managers, PMO office, Security, business users, etc. Reminding us again why change management is such an important issue for organizations to pay attention to when adopting a new technology that can be disruptive to the employees.

He also took a deep dive into their CoE design and team structure, please refer to 0:45 of our recording.

 

Picture5

What is next?

Kemal mentions that they are still working on many aspects to ramp up their existing RPA implementation, as this is an ever-evolving ecosystem which drives the larger digital transformation initiatives in the organization, these include:

•Re-Asses SDLC

•DigitalLab

•MicroBots

•Shared Libraries and Activities

•Attended BOT policy

•RPA tools / Usage and control

•Industry Watch

---

We want to thank Kemal for sharing an amazing story about the RPA journey that Raymond James embarked a few years ago and is quickly scaling up to be a highly advanced automation program. We hope this webinar provided great benchmark points for our audience.

Whether you are just starting your RPA journey or looking to expand your existing RPA program, JOLT is here to help. Feel free to contact us if you have a specific use case you’d be interested in automating with RPA, we are happy to help you assess the opportunity and build a robust business case for your executive team, and if you are looking to scale your current implementation, please feel free to ask us about our RPA pulse check program.

 

New Call-to-action
comments
0