This put up was co-written with Tony Momenpour and Drew Clark from KYTC.
Authorities departments and companies function contact facilities to attach with their communities, enabling residents and clients to name to make appointments, request companies, and typically simply ask a query. When there are extra calls than brokers can reply, callers get positioned on maintain with a message similar to the next: “We’re experiencing larger than traditional name volumes. Your name is essential to us, please keep on the road and your name will likely be answered within the order it was acquired.”
Until the maintain music is especially good, callers don’t usually get pleasure from having to attend—it wastes money and time. Some contact facilities play automated messages to encourage the caller to go away a voicemail, go to the web site, or name again later. These choices are unsatisfying to callers who simply need to ask an agent a query to get a solution shortly.
One answer is to have sufficient educated brokers accessible to take all of the calls instantly, even throughout instances of unusually excessive name volumes. This might remove maintain instances and be certain that callers obtain quick responses. The important thing to creating this strategy sensible is to enhance human brokers with scalable, AI-powered digital brokers that may handle callers’ wants for not less than among the incoming calls. When a digital agent efficiently addresses a caller’s enquiry, the result’s a cheerful caller, decrease common maintain instances for all callers, and decrease prices. Gartner’s Buyer Service and Help Chief ballot estimates that reside channels similar to cellphone and reside chat value a median of $8.01 per contact, whereas self-service channels value about $0.10 per contact—a digital agent can probably save $7.91 (98%) for each name it efficiently handles.
A digital agent doesn’t should deal with each name, and it most likely shouldn’t attempt—some portion of calls are possible served finest with a human contact, so digital agent ought to know its personal limitations, and shortly switch the caller to a human agent when wanted.
On this put up, we share how the Kentucky Transportation Cupboard’s (KYTC) Division of Automobile Laws (DVR) decreased name maintain time and improved buyer expertise with self-service digital brokers utilizing Amazon Join and Amazon Lex.
KYTC DVR’s challenges
The KYTC DVR helps, assists and gives data associated to car registration, driver licenses, and industrial car credentials to almost 5 million constituents.
“In a latest survey carried out with Kentucky residents, greater than 50% really needed assist with out talking to somebody,” says Drew Clark, Enterprise Analyst and Challenge Supervisor at KYTC.
There have been a number of challenges the KYTC staff confronted that made it essential for them to exchange the present system with Amazon Join and Amazon Lex. The shortage of flexibility within the current customer support system prevented them from offering their clients the perfect person expertise and from innovating additional by introducing options like the flexibility to deal with redundant queries by way of chat. Additionally, the introduction of federal REAL ID necessities in 2019 resulted in elevated name volumes from drivers with questions. Name volumes elevated additional in 2020 when the COVID-19 pandemic struck and driver licensing regional workplaces closed. Callers skilled a median deal with time of 5 minutes or longer—an undesirable scenario for each the callers and the DVR contact heart professionals. As well as, there was an over-reliance on the callback function, leading to a under par buyer expertise.
To deal with these challenges, the KYTC staff reviewed a number of contact heart options and collaborated with the AWS ProServe staff to implement a cloud-based contact heart and a digital agent named Max. At present, clients can work together with the contact heart by way of voice and chat channels. The contact heart is powered by Amazon Join, and Max, the digital agent, is powered by Amazon Lex and the AWS QnABot answer.
Amazon Join directs some incoming calls to the digital agent (Max) by figuring out the caller quantity. Max makes use of pure language processing (NLP) to search out the perfect reply to a caller’s query from the DVR’s data base of questions and solutions, and responds to the caller utilizing a pure and human-like synthesized voice (powered by Amazon Polly), supplemented when acceptable with an SMS textual content message containing hyperlinks to webpages that present related detailed data. With Amazon Lex, the division was in a position to automate duties like offering data on REAL IDs, and renewing driver’s licenses or car registrations. If the caller can’t discover the specified reply, the decision is transferred to a reside agent.
The KYTC DVR reviews that with the brand new system, they’ll deal with the identical or higher name volumes at a decrease operational value than the earlier system. The decision dealing with time has been decreased by 33%. They persistently see 90% of the QnABot visitors routing via the self-service possibility on the web site. The QnABot is now dealing with near 35% of the incoming cellphone calls with out the necessity for human intervention, throughout common enterprise hours and after hours as properly! As well as, agent coaching time was decreased to 2 weeks from 4 weeks on account of Amazon Join’s intuitive design and ease of use. Not solely did DVR enhance the client and agent expertise, however additionally they averted excessive up-front prices and decreased their general operational value.
Amazon Lex and the AWS QnABot
Amazon Lex is an AWS service for creating conversational interfaces. You should utilize Amazon Lex to construct succesful self-service digital brokers on your contact heart to automate all kinds of caller experiences, similar to claims, quotes, funds, purchases, appointments, and extra.
The AWS QnABot is an open-source answer that makes use of Amazon Lex together with different AWS companies to automate query answering use instances.
QnABot permits you to shortly deploy a conversational AI digital agent into your contact facilities, web sites, and messaging channels, with no coding expertise required. You configure curated solutions to continuously requested questions utilizing an built-in content material administration system that helps wealthy textual content and wealthy voice responses optimized for every channel. You may broaden the answer’s data base to incorporate looking current paperwork and webpage content material utilizing Amazon Kendra. QnABot makes use of Amazon Translate to assist person interplay in lots of languages.
Built-in person suggestions and monitoring present visibility into buyer queries, considerations, and sentiment. This allows you to tune and enrich your content material, successfully educating your digital agent so it will get smarter on a regular basis.
The KYTC DVR contact heart has achieved spectacular buyer expertise and cost-efficiency enhancements by deploying an Amazon Join cloud-based contact heart, together with a digital agent constructed with Amazon Lex and the open-source AWS QnABot answer.
Curious to see if you happen to can profit from the identical approaches that labored for the KYTC DVR? Take a look at these quick demo movies:
Attempt Amazon Lex or the QnABot for your self in your personal AWS account. You may comply with the steps within the implementation information for automated deployment, or discover the AWS QnABot workshop.
We’d love to listen to from you. Tell us what you assume within the feedback part.
Concerning the Authors
Tony Momenpour is a programs advisor throughout the Kentucky Transportation Cupboard. He has labored for the Commonwealth of Kentucky for 19 years in numerous roles. His focus is to help the Commonwealth with with the ability to present its residents an important customer support expertise.
Drew Clark is a enterprise analyst/challenge supervisor for the Kentucky Transportation Cupboard’s Workplace of Data Know-how. He’s specializing in system structure, utility platforms, and modernization for the cupboard. He has been with the Transportation Cupboard since 2016 working in numerous IT roles.
Rajiv Sharma is a Area Lead – Contact Middle within the AWS Information and Machine Studying staff. Rajiv works with our clients to ship engagements utilizing Amazon Join and Amazon Lex.
Thomas Rindfuss is a Sr. Options Architect on the Amazon Lex staff. He invents, develops, prototypes, and evangelizes new technical options and options for Language AI companies that improves the client expertise and eases adoption.
Bob Strahan is a Principal Options Architect within the AWS Language AI Companies staff.