Skip to main content

Platforms and Technology

 Amazon Connect Dynamic Menus from a Document Upload

Our consultant team of Amazon Connect experts worked together to creatively solve an interesting business use case for one of our clients. We were presented with a scenario that required business users to dynamically update their IVR menus, while not being able to directly access the call center system. This business team also needed the ability to quickly and frequently change the order of their menu options.

This feature couldn’t be owned by one group, and it needed to support the shared responsibility of three different teams with undesirable permissions:

  • A security team that had requirements that prevented the business team from directly accessing their contact center
  • A business team of product owners (line of business (LOB) owners) with full control over what the IVR menus should offer, and the order that the options would appear in
  • an IT team that supported the Amazon Connect contact center and had the authority to modify the IVR menus. This team desired minimal maintenance responsibilities for the menus and was unable to support the frequency of menu change requests from the business team

Our development team offered a solution that satisfied all three teams:

  • We worked within the constraints requested by the security team and had a strong understanding of the Amazon Connect contact center permissions. We also knew that there was a gap that could not be addressed via configurations.
  • We identified an alternative method to give the business team the control they needed – upload a document that specified the menu order and allow messaging associated with the menus.
  • We provided a minimal-maintenance path to allow for frequent updates to the menus via smartly defined Amazon Connect contact flows.

Discovering and Addressing the Gaps

We worked diligently with the LOB owners to understand how their customers interacted with the contact center and how they intended to deliver information and self-service capabilities to those callers. We were quickly able to identify and recommend the following adjustments and improvements to the IVR:

  • optimizing user experience
  • removing customer-trapping scenarios
  • streamlining the caller’s path towards an agent “self-serve to agent pathing” experience.

Furthermore, in coordination with the IT team, we established a way for the business team to upload a spreadsheet to an AWS S3 bucket related to their menu configurations. We also found a way for the IT team to maintain functional documentation through the well-defined Amazon Connect contact flows.

We made the upload of the spreadsheet as flat as possible while also containing all LOB owner inputs.  The upload consisted of the following fields:

  • MenuId: Specify the contact flow / business context that menu option is related to
  • MenuOrder: Specify the order/position of the option in the menu
  • MenuOption: The special unique identifier of the items in the menu (Process Payment, Agent, etc.)
  • Prompt: Specify the message that the customers will hear

See below for sample data:

MenuId MenuOrder MenuOption Prompt
paymentmenu 1 ProcessPayment To pay your current balance with a credit card.
paymentmenu 2 PaymentFAQ For answers to frequently asked questions.
paymentmenu 0 Agent To speak with an agent.
customerservice 1 SalesMenu To see what services are available for your account.
customerservice 2 Extensions To contact a representative by extension
customerservice 3 MyAccount To get your account details
customerservice 0 Agent To speak with an agent.


Leveraging additional Amazon Web Services

Our development team created a trigger on the S3 bucket to activate a process that would transform the spreadsheet into multiple records that would be loaded into the DynamoDB table, with data validation checks and data cleaning processes. We provisioned Lambdas to retrieve, transform, and output these data points to TTS menu prompts.

The menu choices are then dynamically retrieved and processed in the Amazon Connect contact flow that the IT team maintained:


Contact Flow Configurations

  1. The IVR system collected and populated contact attributes to establish the customers’ context and reason for calling in, thus identifying the menu that should be offered.
  2. We provided a Lambda function that would fetch the menu details based on the established caller context (e.g.: paymentmenu, customerservice). This function would be a multi-step process that includes the following actions: fetch the menu details from the database, stitch the items together in an ordered fashion (0-9), and provide a message prompt that activates in
  3. After receiving the customer’s input (0 through 9), the contact flow would pass the customer input, add menu context to a Lambda function to resolve the choice selected (ProcessPayment, PaymentFAQ, Agent, etc.), and send it back as an external contact attribute.
  4. The contact flow block would then use this external contact attribute to route the caller to the correct path and/or flow.

Final Results

After completion of this development experience:

  • The business team had control over their IVR menu choices and a way to update their menu orders without being restricted by the support team’s schedule.
  • The contact center IT team, through coordination with the business team, was able to minimize their maintenance and support needs for Contact Flow Designer updates.
  • The security team could uphold their goals of minimizing access, giving security roles and functionality only as required.

If you are struggling to find a solution that fits your unique and complex organizational structure and need some guidance on maximizing your contact center’s efficiency, we can help. At Perficient, we are an APN Advanced Consulting Partner for Amazon Connect which gives us a unique set of skills to accelerate your cloud, agent, and customer experience.

Perficient takes pride in our personal approach to the customer journey where we help enterprise clients transform and modernize their contact center and CRM experience with platforms like Amazon Connect, Service Cloud Voice, or our own personalized offering: Perficient Amazon Connect Experience.

For more information on how Perficient can help you get the most out of your contact center needs, please contact us here.




Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Alan Ching

Alan Ching is a Lead Business Consultant and a Project Manager who is focused on Cloud Contact Center implementations powered by AWS Amazon Connect.

More from this Author

Follow Us