Genesys Cloud - Main

 View Only

Sign Up

  • 1.  How are you managing promotion of changes from Dev to Prod?

    Posted 15 days ago

    Hi all! I thought I would post this question in the community to see what great strategies I could stir up!

    I'm working with a client who is looking to better understand how other Genesys Cloud customers manage the promotion of changes across environments, specifically from development/non-prod into production.

    They are also interested in hearing how others handle multiple dev or non-prod orgs, especially when different teams, business units, or workstreams are making changes in parallel.

    A few areas they are curious about include:

    • How you track and document configuration changes before moving them to prod
    • Whether you use one dev org or multiple dev/test orgs, and how you manage consistency across them
    • Whether you use manual processes, APIs, Terraform, CI/CD, or another approach
    • How you handle approvals, testing, and rollback planning
    • What types of changes you promote regularly, such as Architect flows, routing, queues, integrations, bots, etc.
    • Any lessons learned or best practices that have helped reduce risk

    Would anyone be willing to share a bit about how your organization approaches this today, what has worked well, and what you would do differently?

    Thanks in advance for your thoughts!


    #System/PlatformAdministration

    ------------------------------
    Kendall Swartzell
    Senior Customer Success Manager
    ------------------------------


  • 2.  RE: How are you managing promotion of changes from Dev to Prod?

    Posted 15 days ago

    Below is my experiance:

    • How you track and document configuration changes before moving them to prod
      Answer: We will prepare document "Method statement" which consists of all the steps related to deplanement plan , Change window, Roll back, testing, configuration procedure.
    • Whether you use one dev org or multiple dev/test orgs, and how you manage consistency across them
      Answer: Yes we have both Dev and Prod instance in the client side. Initial development will be in Dev lab and followed by Prod.
    • Whether you use manual processes, APIs, Terraform, CI/CD, or another approach
      Answer: Manual as of now but planning to implement Terraform in future.
    • How you handle approvals, testing, and rollback planning
      Answer: We are using agile methodology through Jira we will raise request and track everything including approvals.
    • What types of changes you promote regularly, such as Architect flows, routing, queues, integrations, bots, etc.
      Answer: We will touch everywhere based on the requirement
    • Any lessons learned or best practices that have helped reduce risk
      Answer: Checklist and approvals are important along with Dev test results and Post Prod results.


    ------------------------------
    Satish Huzuru
    Infra Transformation Associate Manager
    ------------------------------



  • 3.  RE: How are you managing promotion of changes from Dev to Prod?
    Best Answer

    Posted 13 days ago

    Hopefully the following helps...

    How you track and document configuration changes before moving them to prod

    All our changes are committed to version control (GitHub) via a Pull request (PR). It is within this PR that we add a description of the change.

    It is only when a colleague has reviewed the changes, which includes ensuring they can understand the PR description is it then merged into our main codebase. This merge triggers a deployment to our prod Genesys Cloud org.

    This is a high-level diagram from my newsletter article of all we do in our process to get a change from dev to prod:

    image

    Whether you use one dev org or multiple dev/test orgs, and how you manage consistency across them

    We use dev, nonprod and prod. Consistency is managed by an automated pipeline that will overwrite any manual changes. Meaning people must make them via our CX as Code route.

    How you handle approvals, testing, and rollback planning

    Since changed must be contained within a Pull Request it means we can rely on GitHub's approval process, which includes a check that the automatic tests have passed.

    For testing we use my open-source testing tools e.g. Genesys Cloud Chatbot Tester.

    Rollback is performed by reverting a change within the GitHub project, which causes the old change to be deployed again (over the top of the last change).

    What types of changes you promote regularly, such as Architect flows, routing, queues, integrations, bots, etc.

    We regularly promote flows, integrations and bots.

    Any lessons learned or best practices that have helped reduce risk

    In our deployment pipeline the automated tests must pass before a change can be merged. Originally these tests were static and very scripted. The scripted tests are reducing in favour of AI based tests, which adapt with the changes, which still being strict on degradations in the CX experience.



    ------------------------------
    Lucas Woodward
    Winner of Orchestrator of the Year, Developer (2025)

    LinkedIn - https://www.linkedin.com/in/lucas-woodward-the-dev
    Newsletter - https://makingchatbots.com
    ------------------------------



  • 4.  RE: How are you managing promotion of changes from Dev to Prod?

    Posted 8 days ago

    Hi Lucas,

    This is only working for Flow and data actions? If so, how do you manage to keep the rest clean i.e. Queues, skill groups, etc.

    thanks

    Simon



    ------------------------------
    Simon Dufour
    ------------------------------



  • 5.  RE: How are you managing promotion of changes from Dev to Prod?

    Posted 12 days ago
    Edited by Bogdan Simaciu 12 days ago
    • How you track and document configuration changes before moving them to prod

    We require Jira tickets for any change request. Bug fixing is based on user CRM support tickets.

    • Whether you use one dev org or multiple dev/test orgs, and how you manage consistency across them

    DEV also Test and Prod. Genesys is very inflexible with regards to dev/test licensing compared to many other vendors (maybe smth to look into)

    • Whether you use manual processes, APIs, Terraform, CI/CD, or another approach

    TFE VCS (CXasCode)  + Github --> PR requests with Jira/Ticket numbers and detailed description.

    • How you handle approvals, testing, and rollback planning

    Test in DEV, approval per Jira requests. Nothing gets promoted without testing.

    • What types of changes you promote regularly, such as Architect flows, routing, queues, integrations, bots, etc.

    All types, even small configuration changes (e.g. ring rotation timeouts)

    • Any lessons learned or best practices that have helped reduce risk

    Using Terraform already is a big plus for keeping control on configuration and flow code. Adding checks and pipeline will be our next step.



    ------------------------------
    Bogdan Simaciu
    ------------------------------



  • 6.  RE: How are you managing promotion of changes from Dev to Prod?

    Posted 10 days ago

    @Jared Jaeger, see this thread from what I mentioned last week!



    ------------------------------
    Kendall Swartzell
    Senior Customer Success Manager
    ------------------------------



  • 7.  RE: How are you managing promotion of changes from Dev to Prod?

    Posted 5 days ago

    Gostaria de acrescentar uma perspectiva de desenvolvedor a essa discussão.

    Ao longo dos últimos anos trabalhando com Genesys Cloud, já tive a oportunidade de atuar individualmente e com diferentes equipes, tanto em projetos prioritários quanto em múltiplos projetos executados em paralelo.

    O único cenário com o qual ainda não trabalhei diretamente foi uma esteira totalmente automatizada de CI/CD. Mesmo assim, em todas as experiências que tive, o fator mais importante para uma implantação segura sempre foi possuir uma documentação clara, detalhada e previamente validada pelo time.

    Essa documentação pode assumir diferentes formatos, dependendo do projeto, como:

    • Documento de visão e requisitos.
    • Diagramas no Miro ou Draw.io.
    • Protótipos no Figma.
    • Especificações funcionais e técnicas.
    • Inventário das configurações e dependências.
    • Plano de implantação e rollback.
    • Evidências e cenários de teste.

    Quando existem ferramentas capazes de promover fluxos e configurações automaticamente entre organizações, o uso de ambientes separados de desenvolvimento, homologação e produção oferece ganhos evidentes de governança e segurança.

    Por outro lado, quando essas ferramentas não estão disponíveis, manter uma organização exclusiva para desenvolvimento ainda traz benefícios, mas pode reduzir significativamente a velocidade de implantação. Nesse cenário, muitas configurações precisam ser recriadas, ajustadas ou validadas manualmente na organização de destino.

    Já trabalhei com planos detalhados de troca de fluxos nos roteamentos, tanto para executar pessoalmente uma mudança e seu possível rollback quanto para permitir que outro desenvolvedor conduzisse a implantação.

    Para mim, um bom plano de mudança precisa permitir que outra pessoa consiga:

    1. Entender o estado atual da produção.
    2. Identificar exatamente o que será alterado.
    3. Executar as etapas na ordem correta.
    4. Validar o resultado após a implantação.
    5. Retornar ao estado anterior caso algo falhe.

    Uma das principais lições que aprendi trabalhando com Genesys Cloud é que devemos sempre manter documentados tanto o estado atual da produção quanto as alterações propostas.

    Também considero essencial possuir uma cobertura de testes elevada. Os testes não devem validar apenas a nova funcionalidade, mas também incluir cenários regressivos para garantir que jornadas, roteamentos e integrações já existentes não tenham sido afetados.

    Na minha experiência, ferramentas e automação aceleram bastante o processo, mas documentação, revisão e testes continuam sendo a base para reduzir riscos durante a promoção de mudanças entre ambientes.

    ________________________________________________________________________________

    I would like to add a developer's perspective to this discussion.

    Over the past few years working with Genesys Cloud, I have had the opportunity to work independently and with different teams, both on high-priority projects and on multiple projects running in parallel.

    The only scenario I have not yet worked with directly is a fully automated CI/CD pipeline. Even so, in all my previous experiences, the most important factor for a safe deployment has always been having clear, detailed documentation that was reviewed and validated by the team.

    Depending on the project, this documentation may include:

    • Vision and requirements documents.
    • Miro or Draw.io diagrams.
    • Figma prototypes.
    • Functional and technical specifications.
    • Configuration and dependency inventories.
    • Deployment and rollback plans.
    • Test scenarios and evidence.

    When tools are available to automatically promote flows and configurations between organizations, using separate development, non-production, and production environments provides clear governance and risk-reduction benefits.

    However, when these tools are not available, having a dedicated development organization still provides value, but it can significantly reduce deployment speed. In that situation, many configurations must be manually recreated, adjusted, or validated in the destination organization.

    I have worked with detailed plans for switching flows within routing configurations, both when I was responsible for executing the change and potential rollback, and when another developer needed to perform the deployment.

    In my opinion, a good change plan should allow another developer to:

    1. Understand the current production state.
    2. Identify exactly what will change.
    3. Execute the steps in the correct order.
    4. Validate the result after deployment.
    5. Restore the previous state if something goes wrong.

    One of the most important lessons I have learned while working with Genesys Cloud is that both the current production configuration and the proposed changes should always be documented.

    I also believe that high test coverage is essential. Testing should not only validate the new functionality but should also include regression scenarios to ensure that existing journeys, routing behaviors, and integrations have not been affected.

    In my experience, tools and automation can greatly accelerate the process, but documentation, peer review, and testing remain the foundation for reducing risk when promoting changes between environments.



    ------------------------------
    Matheus Mendonca
    ------------------------------