Your Resource for All Things Apps, Ops, and Infrastructure

Migrating an Enterprise Call Center to the Cloud

A positive customer call center experience requires flexible and reliable contact center technology. Switching to a cloud-based contact center provides cost-effective scalability, increased reliability, and overall improved customer experience.

To ensure high uptime, the migration of the cloud center requires a skilled team of software engineers to build out the system in the cloud. At the enterprise level, it is crucial that the on-premise infrastructure is not interrupted while components are shifted to cloud providers such as Amazon Web Services (AWS).

Basic architecture requirements must be carefully vetted to ensure a smooth transition. A solution must be provided that keeps in mind many of the considerations around stability, cost, and operational efficiency. A shortlist of those considerations include:

  • life cycle manageability
  • flexibility for rapid growth
  • enterprise security requirements
  • long-term efficiency
  • cost-effectiveness

Key Requirements for Migration

A cloud-based customer engagement platform has many components, which must be unified.  Customer-facing employees need to ensure internal data stays internal and that the flow of information across departments remains uninterrupted. Communication infrastructures are company-specific, but many components are common in each organization, such as chat, chat bots, Interactive Voice Response (IVR), automated call routing, inbound / outbound call center, and third-party integrations.

AWS-hosted resources consist of what may seem like a typical set of servers, databases, storage and more, but under the hood, it’s a bit more complex. For maximum fault tolerance and resiliency, these resources must be physically isolated to prevent performance issues and provide additional security as well as must be deployed across two availability zones within each in-use region. Multiple tiers (e.g., development, pre-production, and production) are built and maintained to handle a full pipeline of changes, upgrades, and testing.

For reporting and analytics, specialized internal teams are engaged to build a cloud-based ETL service utilizing AWS tools such as Glue and Athena. This work is performed in conjunction with the “lift and shift” work being performed in phases – this work that affects all business units in the organization. 

In addition to the AWS resources discussed earlier, other key requirements are as follows:

  • EC2 instances deployed to private subnets within a VPC with a minimum of one instance per role per zone utilizing Elastic Load Balancers to load balance web traffic and business telephony application clustering for high availability (HA).
  • DMS replication instances and database EC2 instances deployed to reserved private subnets within a VPC with a minimum of one instance per zone to establish an HA cluster.
  • Latency of 70ms or less is recommended by the vendor to maintain call quality for customers and voice analytics software. This required placement of media servers close to session border controls and telecommunications services.
  • DMS statistics are required for data quality checks made by downstream consumers.  Statistics are only available via AWS CLI, so a Lambda function can be deployed to provide this data.

Common Challenges

There are many challenges, but they can be mitigated by a skilled team of experts aligned with a strong project management framework. A short list of key challenges to consider before taking action is below:

  • Migration plans should be vetted at the enterprise level. Teams should provide UAT sign-off prior to release.
  • The development team might be lacking skills in areas involving various AWS services or they may have no skills in the cloud. This will require training prior to and during development.
  • Third-party support teams might handle support in tiers 1 and 2, so operational alarms (e.g., AWS CloudWatch) must integrate with IT operations management (e.g., ServiceNow) for incident creation and response.
  • Production changes should pass through a change advisory board prior to implementation.
  • Changes from other teams can negatively impact call center changes, possibly jeopardizing service-level agreements.

Business Value

If proper consideration of the common challenges is taken, the recommended design principles are followed, and the myriad of systems and assets involved with unified communication is protected, tremendous business value can be delivered following a successful implementation.

Recently, Vertical Trail successfully completed a call center and voicemail migration from on-premise to AWS, fielding more than 25,000 calls per month in the new system. With the right team in place, you can achieve the same.

Ready to migrate your call center to the cloud? Contact us for a complimentary consultation.

Subscribe to the AHEAD i/o Newsletter