By Tarun Yalavarthii, AWS Developer/Architect at BizCloud Experts
By Rupesh Kothapalli, AWS Architect/DevOps Engineer at BizCloud Experts
By Nagesh Kunamneni, Chief Technical Officer at BizCloud Experts
Due to lockdown restrictions, the global COVID-19 pandemic brought an unprecedented challenge to small, eat-in restaurants. Many turned into “ghost kitchens” as they were expected to transform into a carry out and delivery model overnight.
The food service industry is the second largest private employer in the United States, counting more than 1 million restaurants, including food chain outlets. Of these, 90 percent are small businesses with fewer than 50 employees. Most are independent or family owned, according to the National Restaurant Association.
The pandemic forced millions of diners to call in for carry out and delivery orders, and telephone lines at many of the nation’s eat-in restaurants could not handle call volumes during rush hours. Customer service dipped rapidly as people experienced busy tones and dropped calls. It was difficult for staff to keep track of who was calling in for orders.
Margarita’s Mexican Restaurant, a family-owned eat-in based in Justin, Texas, faced a similar situation during the pandemic’s early stages. At first glance, this looked like a telephony problem, but our team at BizCloud Experts understood the situation from a customer experience view.
We quickly designed a solution to transform Margarita’s staff manning the point of sale (POS) stations to become agents on Amazon Connect. Within 48 hours, the restaurant was handling its carry out and delivery business much more efficiently.
Amazon Connect is a cloud-based customer contact center solution built by Amazon Web Services (AWS). It helps businesses provide superior customer service at a lower cost, with minimal configuration and little time to deploy required.
In this post, we’ll walk you through the solution we implemented and AWS services we utilized, and the results we achieved for Margarita’s.
BizCloud Experts is an AWS Advanced Consulting Partner with the Amazon Connect Service Delivery designation, meaning we have been validated by AWS Partner Solutions Architects as experts in delivering this specific AWS service for customers.
Objectives of Margarita’s Contact Center
Many businesses look to outsource their call operations because managing a typical contact center is complex. Doing so includes managing call routing, scripting, integrated software, call distribution handled by employees responsible for quality assurance, and software customization.
During the pandemic, it has been especially hard for small businesses operating without a contact center to transform their company into a carry out and delivery-optimized model.
The key pain point BizCloud Experts addressed for Margarita’s was the ability to handle inbound call volumes during peak business hours.
Apart from that, the solution we implemented had the following objectives:
- Provide inbound call treatment for customer engagement.
- Overflow inbound calls to outbound team during off-times.
- Handle agent non-availability and heavy call volume with low wait times.
- Notify management via email or text of contact center conditions.
- Design reports to support daily and summarized activity.
The following diagram gives an overview of the solution BizCloud Experts implemented for Margarita’s.
Figure 1 – Implementation of contact center using Amazon Connect and capture of workflow.
The solution we implemented covers the following aspects:
- Store and retrieve call records from Amazon DynamoDB table.
- Define a custom call flow and integrate workflow checks.
- Insert the missed call records into customer relationship management (CRM) software of choice and generate leads.
Basic Call Flow and AWS Services Utilized
Let’s walk through the custom call flow process we defined for this solution, the AWS services we utilized, tools we integrated, and results we achieved for Margarita’s.
After provisioning the contact center, BizCloud Experts configured the business operating hours and created a couple of agents with a basic routing profile. Margarita’s was then provided the URL endpoint for their agents to log in to their virtual desk.
The next step was to create custom call flows to handle incoming calls with a custom welcome message. Call flow checks for the business operating hours we specified during the initial configuration. If there’s an incoming call outside regular business operating hours, the call flow can be configured to play a custom message.
The workflow checks for any available agent during the regular business hours. If the agents are busy, the call goes to the queue until it’s answered by an agent. A custom music tone is configured while the customer is on hold until the call is answered.
Calls are automatically escalated to a mobile number during business hours if unanswered. Also, if the wait time in the queue is more than 10 minutes, there are different call-back options that can be automated.
We created an Amazon DynamoDB table to store call records that contain items, such as customer phone number and history (timestamp and flow). A sample call record from the DynamoDB table is shown in Figure 2 below.
AWS Lambda was integrated to insert records into the DynamoDB table. During the call flow, a Lambda function is triggered multiple times and input parameters are passed from the Amazon Connect call flow with the destination key as “flag.” The value is based on where the Lambda function is triggered (inbound/outbound/routed, for example).
Based on the input parameters passed to Lambda from the call flow, the path is decided. Even though the Lambda function is triggered multiple times with the same input parameter destination key “flag” and different values during a call, all flag values are recorded to the same record in DynamoDB based on Contact Id and the flow is decided accordingly.
Figure 2 – Sample call record from the Amazon DynamoDB table.
Another Lambda function is triggered at regular intervals (every four hours in this case) using Amazon CloudWatch Events. This function is responsible for sending alerts related to missed calls to the CRM of choice. Margarita’s has utilized Zoho CRM, as shown in the architecture diagram in Figure 1 above.
The Lambda function scans the DynamoDB table and checks for the records that are four hours old. It verifies the last value in flow and if “answered,” the record is ignored. Else, it will create a task for the record in Zoho CRM as a missed call that contains a custom subject, phone number, and the call flow.
Notifications and Reporting Techniques
Apart from the configuration of call flow, setting up routing profiles for agents, and the integration of CRM software, BizCloud Experts also implemented the following notification and reporting techniques to help Margarita’s measure their key performance indicators:
- Defined custom overflow conditions like call escalations based on the number of calls in queue, volume of calls during rush hour, and agent availability.
- Call process report on a daily basis to understand the number of calls blocked, average time handled, and calls answered within specific time.
- Notify management about extreme wait times so calls can be escalated based on overflow conditions specified.
- Agent productivity reports to determine the occupancy rate, turnover rate, and first call resolution.
By implementing the Amazon Connect solution for Margarita’s, we achieved great results:
- Go-live in 1/2 day.
- Order process increased by 91 percent.
- Customer complaints reduced by 87 percent.
- No change to POS operations.
- Simplified call handling by eliminating the need to hold telephone lines.
- Take orders remotely, enabled ability to social distance, and reduced congestion in restaurant
As a small business ourselves, BizCloud Experts understands the challenges faced by many companies during the pandemic, and we are committed to helping SMBs rethink their customer experience.
This quote from Margarita’s captures the essence of our partnership: “We make good Mexican food, but we are not experts in technology,” says Co-Owner Jovany Sandoval. “BizCloud Experts quickly observed our operations and seamlessly integrated the call queueing mechanism with our single land-line right at our point of sale system. We now can handle increased call volumes, allowing us to turn our focus back to preparing and delivering good food.”
With Amazon Connect, Margarita’s Mexican Restaurant was able to provide a seamless experience across voice for the customers and agents. It includes a single set of tools for skills-based routing, powerful real-time and historical analytics, and easy-to-use intuitive management tools.
Amazon Connect simplified the restaurant’s contact center operations, improved agent efficiency, and lowered operational costs with its pay-as-you-go pricing model.
BizCloud Experts utilized AWS services such as AWS Lambda to store and retrieve records, Amazon DynamoDB to store information records related to incoming calls, and Amazon CloudWatch Events to run a scheduled cron job to trigger the Lambda functions.
BizCloud Experts can help SMBs set up a contact center, build smart serverless integrations, and support your migration to Amazon Connect from current legacy solutions.
“In these unprecedented times, we believe small-medium brick and mortar businesses have to ReThink how to service customers,” says Nagesh Kunamneni, President and CTO at BizCloud Experts. “We can simplify and improve voice handling for SMBs by building and implementing solutions on Amazon Connect.”
The content and opinions in this blog are those of the third party author and AWS is not responsible for the content or accuracy of this post.
BizCloud Experts – AWS Partner Spotlight
BizCloud Experts is an AWS Advanced Consulting Partner that helps customers accelerate digital transformation projects through consulting, development, collaboration, and integrations.
*Already worked with BizCloud Experts? Rate the Partner
*To review an AWS Partner, you must be a customer that has worked with them directly on a project.