DevRev Demo

Summary

The video script showcases a demo of the support capability on the DevRev platform, emphasizing the importance of customer support for retention and satisfaction. It highlights features like multi-channel support, SLA targets, AI-powered deflection with 'DevRev Turing', and seamless ticket escalation. The script also demonstrates the efficiency of using AI for summarizing conversations and clustering tickets, improving visibility and prioritization in customer support.

Takeaways

• 📊 The script is a demo of the support capability within the DevRev platform, emphasizing the importance of customer support for retention and satisfaction.
• 🔍 The support platform integrates multiple communication channels, allowing for a unified inbox view that consolidates emails, chats, and other forms of customer inquiries.
• 🤖 An AI capability called DevRev Turing is utilized for intelligent deflection, providing automated responses and recommendations to customer queries based on context and external information.
• 🔑 The platform highlights the significance of Service Level Agreements (SLAs) in customer support, with different targets depending on the customer's tier and the nature of the problem.
• 🔗 The demo showcases the ability to link tickets and issues, allowing for a streamlined workflow from customer inquiry to problem resolution.
• 🔄 The script demonstrates the use of a chat widget, called the plug widget, which can be embedded in applications to provide in-app support.
• 👥 The platform enables easy switching between user personas, such as from a support engineer to an end user, to understand and manage the support process from different perspectives.
• 📈 The script illustrates the use of AI for grouping similar tickets, reducing noise and providing clear visibility into common customer issues and priorities.
• 🛠️ The platform includes a feature to escalate customer inquiries into tickets, which can then be linked to ongoing work items or issues that developers are addressing.
• 🗣️ The demo includes a 'huddle' feature for quick communication and collaboration between support engineers and developers working on related issues.
• 📝 The script concludes with the ability to automatically summarize conversations and tickets using AI, providing efficiency and clarity in the support process.

Q & A

What is the critical aspect of the support capability discussed in the script?

The critical aspect of the support capability discussed in the script is its importance in customer retention and satisfaction, which in turn influences customer loyalty, product purchases, and promotion.

What are the different mediums through which customers can seek support according to the script?

Customers can seek support through various mediums including email, phone, chat, and potentially others, highlighting the omnichannel approach to customer support.

What is an SLA Target in the context of the script?

An SLA (Service Level Agreement) Target is a predefined service level that the support team aims to meet, which can vary depending on the customer's tier of support and the nature of their issue.

How does the script describe the integration of AI in the support platform?

The script describes the integration of AI, specifically 'devrev Turing', which uses context about the user's query, user details, and external information to provide intelligent recommendations and automate responses.

What is the purpose of the plug widget mentioned in the script?

The plug widget is a chat widget that can be deployed in applications, making it easy for users to initiate support conversations within the app they are using.

How does the script illustrate the process of escalating a customer issue to a ticket?

The script illustrates the process by showing a support engineer reviewing a customer's issue, recognizing it as something outside their control, and then escalating it to a ticket for further action by creating a new ticket and linking it to the customer's inquiry.

What is the significance of linking tickets to issues in the script?

Linking tickets to issues allows for better tracking and management of customer problems. It also ensures that updates on the issue's status are automatically propagated to the related tickets and customer conversations.

How does the script demonstrate the use of AI for ticket clustering?

The script demonstrates the use of AI to automatically group tickets based on similarity, reducing noise and providing clear visibility into common problems and priorities among customers.

What feature of the platform allows for automatic summarization of customer conversations?

The platform features an AI-powered 'summarize' command that leverages 'devrev Turing' to automatically generate a synopsis of the conversation history and resolution.

What is the script's stance on the importance of visibility and clarity in the support platform?

The script emphasizes the importance of visibility and clarity as key factors in making the lives of support engineers and customers easier, by providing automated updates, intelligent ticket clustering, and automatic summarization.

What is the 'hangout' functionality mentioned in the script and why might it be considered hidden?

The 'hangout' functionality appears to be a feature within the devrev app that allows for quick, in-app communication with team members. It might be considered hidden because it's not yet publicly disclosed or is in a testing phase.

Outlines

00:00

📘 Support Capability Demo Overview

The speaker introduces a demo of the support capability within the devrev platform, emphasizing the importance of customer support for retention and satisfaction. The platform integrates multiple communication channels, such as email, chat, and phone, allowing for a unified support experience. The demo showcases an inbox view with ongoing conversations and highlights the platform's ability to meet SLA (Service Level Agreement) targets based on customer tier and issue severity. The speaker also switches perspectives to demonstrate the end-user experience, including the use of a chat widget for initiating support requests.

05:03

🔄 AI-Powered Ticket Escalation and Issue Tracking

The script continues with a support engineer's perspective, escalating a customer's issue into a ticket and linking it to a similar existing issue. The platform's AI, named devrev Turing, is utilized to automatically deflect common queries and provide recommendations based on user context and external information. The speaker then discusses the process of identifying and prioritizing widespread issues by searching and linking tickets to ongoing development work, illustrating a collaborative approach to resolving customer problems.

10:04

🤖 Streamlining Support Workflows with AI and Huddles

The speaker demonstrates the use of the devrev app for initiating a 'huddle'—a quick meeting with engineers—to prioritize and address a customer's issue. The platform's integration allows for seamless communication and collaboration without the need for external tools. The AI capability is highlighted again for its role in updating ticket statuses and linking issues to tickets automatically, providing a clear view of the customer's journey from inquiry to resolution.

15:12

📊 Leveraging AI for Ticket Clustering and Summarization

The final paragraph focuses on the use of AI for ticket management, including automatic clustering of tickets based on similarity to identify common issues and prioritize them effectively. The speaker also introduces the 'summarize' command, which uses devrev Turing to generate a synopsis of a ticket's history and resolution, enhancing efficiency for support engineers. The demo concludes with a summary of the platform's capabilities and a tease for future features.

Keywords

💡Support Capability

Support Capability refers to the features and tools that enable efficient customer service. In the context of the video, it is a critical aspect of the DevRev platform, aimed at ensuring customer satisfaction and retention. The script mentions the importance of supporting customers across various mediums like email, phone, and chat, highlighting the platform's ability to handle support requests from multiple sources.

💡Customer Retention

Customer Retention is the ability of a company to keep its customers over time. It is central to the video's theme as it discusses how a happy customer is more likely to continue using a product and recommend it to others. The script emphasizes the role of effective support in enhancing customer happiness and, by extension, retention.

💡Persona

In the script, 'Persona' represents a specific role or user archetype within the platform. The speaker adopts the 'support engineer' persona to demonstrate the platform's functionality from the perspective of someone handling customer requests. This concept helps illustrate the user experience and the platform's capabilities in a real-world scenario.

💡Inbox View

Inbox View is the interface where support engineers can see and manage customer conversations. The script describes it as a place to triage and manage incoming support requests, showcasing the platform's organization and visibility features for handling multiple customer interactions.

💡SLA (Service Level Agreement)

SLA, or Service Level Agreement, is a contractually agreed-upon level of service between a service provider and a customer. The script mentions SLA Targets, which are specific metrics that must be met in terms of response times and resolution, emphasizing the platform's focus on meeting contractual obligations for customer support.

💡Plug Widget

The Plug Widget is a chat interface that can be embedded within applications to facilitate customer support. The script describes its portability and plugability, allowing it to be used in various environments like mobile or web apps, enhancing the accessibility of support services for end-users.

💡DevRev Turing

DevRev Turing is an AI capability within the platform that aids in understanding user queries and providing context-based recommendations. The script illustrates its use in deflection, where the AI assesses customer inquiries and offers solutions or knowledge base articles, reducing the need for human intervention in certain cases.

💡Ticket

A Ticket in the script represents a formal record of a customer's issue that has been escalated from a conversation. It is used to track and manage the resolution process. The speaker demonstrates creating a new ticket and linking it to an issue, showing how the platform organizes and prioritizes customer problems.

💡Huddle

Huddle, as used in the script, refers to an impromptu meeting or discussion within the platform, often involving multiple team members. The speaker uses it to quickly engage with an engineer to discuss and prioritize a customer's issue, demonstrating the platform's collaborative features.

💡CICD (Continuous Integration/Continuous Deployment)

CICD is a software development practice where code changes are automatically built, tested, and prepared for a release. In the script, it is mentioned in the context of an engineer working on a problem, indicating the platform's integration with development workflows to streamline issue resolution.

💡Automation

Automation in the script refers to the platform's ability to perform tasks without human intervention, such as updating ticket statuses or summarizing conversations. This concept is highlighted as a way to increase efficiency and reduce the workload on support engineers, allowing them to focus on more complex issues.

💡Visibility

Visibility in the context of the video pertains to the clear and immediate access to information about customer issues, tickets, and their statuses. The script emphasizes the importance of visibility for support engineers to make informed decisions and provide better service, facilitated by the platform's automated updates and summaries.

💡Clustering

Clustering, as mentioned in the script, is the AI-powered process of grouping similar tickets together. This feature helps in reducing noise and providing a clearer view of common problems, enabling support teams to prioritize and address widespread issues more effectively.

Highlights

The importance of customer support in ensuring customer satisfaction and retention.

Support capabilities are not limited to email, phone, or chat, but span across multiple mediums.

The platform's ability to provide support where customers are, such as in-app chat or email.

The significance of Service Level Agreements (SLAs) in customer support and their tier-based variations.

Demonstration of the support engineer's perspective, handling customer requests and triaging issues.

Introduction of the AI capability called 'devrev Turing' for context-based support recommendations.

The system's automatic deflection using AI to provide immediate responses to customer queries.

How the platform leverages AI to offload the burden from users and make support more efficient.

The end-user perspective with the plug widget for in-app chat support.

The ability to escalate customer inquiries into tickets within the support system.

Linking similar work items and tickets to streamline issue tracking and management.

The use of 'huddle' feature to quickly collaborate with engineers working on related issues.

Automatic status updates and notifications propagated by the system, reducing manual effort.

Visibility into customer accounts, active conversations, and associated tickets for better customer management.

The introduction of AI-based ticket clustering to reduce noise and highlight common issues.

Automatic summarization of conversations and tickets using AI for internal clarity and customer communication.

The overall efficiency and visibility provided by the platform's AI capabilities for support engineers.

A sneak peek at upcoming features and the potential impact on customer support workflows.

Transcripts

foreign

is walking through a quick demo of the

support capability which is part of the

devrev platform

now obviously support is a very very

critical thing nowadays when we start

thinking about uh customer retention how

to ensure that customers are happy right

the happier the customer is the more

inclined they are to One support your

product purchase your product as well as

promote your product and so I think

you're seeing a lot of reinvigorated

focus on the support front

um so

right now for this demo I'm going to

start it out as the Persona of a support

engineer right so I'm on the battlefield

handling customer requests you know

triaging things what be in so this is my

inbox View

so if I zoom in here you can see a few

things one I can see that I have four

conversations taking place uh but more

importantly these are coming from a

multitude of sources and this is an

important thing you know support is not

only email or phone or you know chat

it's really across all of those various

mediums and that's one of the nice

things with this platform is

irrespective of whether I you know send

an email to get support I could you know

in that same thread

get support via chat right and so you

know it kind of comes back to one of the

scenes which is bring your support to

where your customers actually are if

they're in your app bring them chat

support if they're in their inbox bring

them email based support and that

plugability and portability is very

important

one of the other things to note here is

for this customer conversation we have

an SLA Target now SLA is obviously when

it comes to support or very very

important just because these are you

know things that we need to meet

contractually and so depending on the

tier of support who the customer is as

well as the problem that they're facing

you could have various or different SLA

targets here

so that's a brief overview of the inbox

one of the things I'm going to do now is

actually shift context or personas and

show you what it looks like

to be an end user

so here we have our Maple software

quote-unquote company which is a demo

company that provides SAS services for

you know things like billing payment

Services Etc

now one of the keys to highlight here is

in this lower left hand corner

I can see our plug widget

the plug widget is essentially a chat

widget which you can deploy in your

application

whether it's a mobile app a web app

would be it so very plugable and

portable

so here you can see that I have a few

previous conversations as Elon Tusk and

to demonstrate how to initialize another

conversation I'm going to go ahead and

show that to you now

so I'm going to go ahead and send a

message I'm going to say

so one of the first things to highlight

here is obviously I sent my message to

the to the system to the support inbox

but before a user or support engineer

actually responded the system

automatically deflected that now if you

look at traditionally traditionally you

know deflection would be you know rule

based and look for keywords and be very

very rigid

however one of the things we built into

the platform is an AI capability called

devrev Turing

so for this functionality it's actually

using context about the user's query

context and detail about the user as

well as potentially you know external

information like logs and things that

the the application is generating and

it's feeding that into an AI service

very similar to what people see with

chat GPT and then based upon that

context providing a recommendation so in

this scenario I sent a message about a

payment

dispute

and it actually gave me a reference to a

knowledge based article as well as a

link including some summary information

now in certain scenarios this may solve

that problem and that's awesome you know

because me as a user or a support

engineer I didn't have to do anything

and that's one of the key themes of the

system is you know offload the burden or

onus from the user and push it onto the

system leverage an intelligent system to

make your life as a customer as well as

user much more efficient

and so this is one example now just for

the sake of this demo I'm going to say

that this did not solve my problem in

some cases it actually will

and I'm going to

input my stack trays that I'm seeing in

this scenario

so here I went ahead and set that Strack

Trace

now what we're going to do is we're

going to switch back to the support

Engineers persona

now in my inbox I can see that

conversation that was

started by quote unquote Elon Tusk so

now let's go ahead and click on that

conversation

now as a support engineer you know I've

may or may not have seen this stack

Trace before but this looks like

something that is outside of my control

so I'm actually going to escalate this

up into a ticket

so I go here to link tickets and then

I'm going to go ahead and create a new

ticket

I'll input the title

and one of the keys here is you can see

that there are some similar work items

and so you know the system has that

intelligence to see things that are very

similar and may actually make more sense

to link as compared to creating a new

item in this scenario we're just going

to go ahead and create a new ticket

and because this has to do with our pay

service I'm going to select our pay

feature to create this ticket under

so here what you can see is I have all

the history in regards to the initial

customer inquiry and then I escalated

that up into a ticket

now I want to see if this is actually a

common

problem or if anyone else is having the

problem so what I'm going to do here is

I'm actually going to

search

so here we can see that there are a few

tickets associated with this as well as

an issue now issues are things that

developers are actually working on

so if there's a bug or a defect or you

know a new functionality they would be

working on these issues no different

than you would in a traditional

engineering Work Management platform so

I'm going to go ahead and click on this

and here

something right off the bat looks very

very familiar so if I look at this

description

for this particular bug or issue I can

see that this stack Trace

is the exact same stack Trace that my

customer seen

now if I go up here I can see that you

know it looks like it's a P3 so it

hasn't been prioritized

and it's currently in backlog now I saw

my customer had this problem and I saw

that there were a few other uh tickets

so I'm going to go ahead and talk to

this engineer and see what's actually

going on here because this may be a more

widespread than we think

so here I'm actually going to go ahead

and start a huddle and then what we're

going to do is we're actually going to

invite a few people here so in this

scenario looks like there are a few

users which are actually actively

involved in working on this so I'm going

to go ahead and invite these users here

hey what's going on in India

hey Steve

hey how's it going man

um so I'm gonna go ahead and share my

screen out here

um it looks like there's an issue

assigned to you that my customer is

actually facing the same problem so let

me go ahead and share this with you

it looks like they're getting a a stack

Trace

um for the payment API uh where there's

a you know it's throwing a 404 due to a

nil pointer reference

um is this something that you're

familiar with

yeah I actually saw this in one other uh

stack days for some other customer so

I'm pretty sure this is happening across

the board

yeah no definitely definitely

interesting so it looks like right now

it's just in backlog

um and a P3 is this something that you

could help out with and and hop on and

start handling

yeah this definitely looks like a bigger

problem in the blast radius is pretty

big so I'll start working on it I know

it's a P3 right now I'll bump up the

priority I'll let the team know that I'm

prioritizing this over the others and

I'll start working on it

perfect no thanks again yeah I got to

make sure this customer's happy

um so appreciate you hopping on this and

and starting to work on this uh uh on a

minute swim I may get in trouble for

showing this little hangout

functionality or feature just because

it's supposed to be hidden right now but

uh this is pretty awesome we didn't go

to slack did we where are we we're in

the devra v app right no yeah we are in

order yep

nice not good stuff man all right well

yeah thanks again in India and uh yeah

let me know if you have any questions on

from the customer obviously you can see

the associated uh conversation history

and logs but yeah let me know if you

need anything from me man and definitely

appreciate it

thanks for helping our customers Steve

appreciate it yeah okay that that's

that's my job

thank you

so there you can see all natively within

the devrov app I didn't have to switch

contacts I didn't have to go to slack I

didn't have to you know dial someone I

could literally just start a quick

huddle there with the engineer working

on the issue and

get some feedback from him to help

prioritize this issue

and so because this is related I'm

actually going to go ahead and Link this

issue

to this ticket so that's issue five six

one it's cool that looks good

now what are the keys to highlight here

is when I look at this View

I can see the initial customer inquiry

right and in that

I can see that the devrev bot is

actually automatically updating status

for me uh when tickets are linked when

issues are starting development

and that's a very important thing you

know I didn't have to send these status

updates the system is actually

propagating and sending those updates to

me automatically

so here I can see the customer inquiry

I can see my escalation which I'm

working on

I can see the linkage to the issue and I

can actually you know see that in India

has started working

on this problem and as he you know

progresses through that cicd flow the

status updates will propagate to my

ticket as well as to that conversation

without me having to do anything and

that visibility I mean we can't harp on

that enough that visibility that

automatic State transitioning as well as

notifications having the system do that

makes my life as a support engineer so

much easier now one of the things that

has kind of been visible here but you

know hasn't really been pointed out is

if I look at this I can see the

associated customer

so now I'm going to go ahead and view

this customer account

so in here I can see we have a customer

account for Umbrella Corporation

obviously a lot of people seem to be

fans of raccoon city or Resident Evil

but underneath here I have visibility in

regards to you know obviously some

internal conversation and discussion

here

but also you know currently active

conversations who the users are as well

as what tickets

they currently have with us

and so here let's go ahead and take a

look at all these tickets

so in this scenario we can see that this

customer in particular has about you

know 12 tickets

but this is just one of my customers so

let's go ahead and take a look at

tickets for all customers

here you can see there's a lot of

tickets right there's a lot of stuff

going on I may have 5 000 tickets and so

one of the key things that we introduced

was leveraging that same artificial

intelligence that we leveraged for

deflection as well as a few other things

we added the ability to group Things

based upon similarity

so now you can see instead of having you

know a list of 40 distinct tickets

the system automatically clusters those

into common themes so here we can see

very frequently I'm seeing a lot of

payment and API issues I can see there's

a payment dispute 404 error

you know there's some network problems I

need some slack payment Integrations

and so you know rather than you know

having all that noise I can see what the

actual real problems and priorities and

common problems that my customers are

facing and this is a very very important

thing it's all about cutting down noise

and providing visibility if we can

provide as a system visibility and

Clarity it makes everyone's lives easier

the ability to leverage devrev Turing to

automatically cluster your tickets into

chunks or common items which you can

then use to prioritize adequately

now I'm going to hop back over to my

ticket

and one of the things is you know I want

to to summarize this kind of ticket

experience right I may be obligated to

send this to my user or customers but

just for internal Clarity you know I

want to make sure that we have a summary

of the events that actually took place

here

and so that's where you know leveraging

that same

turning AI offload I can actually

automatically summarize the conversation

and so one of the things that we added

was this summarize command and

essentially what this does is this

leverages the exact same devrev touring

artificial intelligence that we

leveraged and highlighted throughout

multiple pieces of this demo but it will

look at the actual history the series of

events as well as provide a synopsis on

what actually happened and what the

resolution was so in this scenario it

actually you know shows that the user

was having a problem with the payment

dispute

um and then based upon the log message

the customer actually sent me it

extracted the exact method and object

that was having the problem right and

you know obviously there's a blurb that

it was traced back to you know various

components of the system obviously we

saw that uh pay feature was having some

problems that in India was working on

so by leveraging these artificial

intelligence offloads not only can we

provide much better visibility

but we can also provide extreme

efficiency right so in that deflection

scenario I may not have actually had to

do anything as a support engineer and

that makes me much more productive

you know rather than having to go

through my case history here or my

conversation history the system actually

automatically summarized that for me and

again a very very powerful thing about

this platform is it can do all that

stuff for you

so you know just a very short demo for

you today but I hope you enjoyed it we

obviously have a lot of very cool stuff

hopefully I don't get in trouble for

showing that hangout thing

but a lot of cool stuff in the pipeline

so definitely stay tuned and we'll show

you some sweet stuff soon

foreign