3 Predictive Ways to Solve for CSAT
Embark on an insightful journey with Steve Trier, Chief of Operations and Product at Tethr, as he unveils the predictive mechanisms that are redefining Customer Satisfaction (CSAT). With a rich background spanning retail, B2C, B2B, and a 12-year stint at Best Buy, Steve brings a wealth of experience in aligning product strategy with evolving AI and machine learning technologies. This session will delve into how Tethr harnesses predictive analytics to stay ahead in meeting customer expectations. Attendees will gain a deeper understanding of how an adept incorporation of AI can significantly elevate customer satisfaction metrics, keeping businesses in sync with consumer needs and market trends. Through a blend of strategic insights and practical examples, discover how to transform reactive customer service approaches into proactive strategies that not only resolve issues but predict and prevent them.
(upbeat music)
- We're excited to get going here.
We got a great session.
We're gonna be talking about three predictive ways
to solve for CSAT and to do that.
We brought on Steve Trear.
Steve has done a lot in this space,
but he's currently the COA Chief Product Officer
at Tether.
Steve, so excited to have you joined
and how that got you.
- I'm doing great, Dave.
And thank you for having me here.
I'm looking forward to the conversation.
- Appreciate it.
You know, as I mentioned,
long background in kind of the CX space,
doing some really fun stuff,
do you mind just taking a minute
and tell us a little bit about some of the fun things
in that background and what you guys
have been doing over there at Tether?
- Yeah, so, you know, I started this a long time ago
in a direct-to-consumer catalog wheeler.
And you think about it,
that was sort of the beginning of moving the relationship
out of the retail store and kind of going more direct.
And that was a lot of fun.
And believe you engage the customers
and the way that you built that journey with them
was a little different.
Of course, we're on a really big learning curve
because we're kind of out.
And of course, not much later than that,
we saw the internet come along.
And that changed it all over again.
And I had the good fortune at that time
of being able to jump into best spot
and beyond the key and how to eat almost platform.
And again, just so much in learning
and how internet customers, how that channel is gonna work,
how it's gonna compliment your existing panels.
It's a lot of fun.
And then, you know,
as I kind of ended up there,
I thought about this thing called Tether.
And as we were looking at it
and kind of taking up the idea,
you were kind of talking about this idea
that you could listen to all of the phone calls
happening in your contact center
and uncovering ideas.
And I will tell you what I was doing
when I kept that's fine,
I'm not understanding it was
'cause it's so hard to get lots of really good data
in context.
It came to me that when I was trying to launch me
to the store, I would actually go to the contact center
and ask them to give me recording
to find them.
So, in the previous product launches, I got mentioned in calls,
because you couldn't get a better source of automation.
And so that's why I'm here at Tether.
And that's why we've been working at this for the last decade
because it's been so much fun
in the space of conversational intelligence.
It's a really kind of hone in on that customer experience,
getting their true voice for customers.
They're being able to provide that back to organizations
to allow them to produce their costs,
you know, improve their customer experience,
improve their conversion, reduce their appearance.
So, I mean, it's been a great journey a lot of time.
- Yeah, you guys really have been at the forefront.
I mean, you guys were talking about AI at Tether, you know,
I think five years ago or something, 40 years ago.
- Yeah.
So we, so we've been just recently now thinking about AI
as a real, well, let's just say that what's gonna happen
is it's gonna be transformative.
Like it's not gonna be just a cycle change.
Like we see this is actually completely transforming
the space brand.
And we're really looking forward to being on that journey
and helping lead the way.
So, you know, today we do a lot of sort of machine learning,
which is really helpful.
You can structure insight, you can pull that out,
and you can try a lot of that benefit.
But AI is going to allow you to really just change the way
that you think about the data that you get,
the way that you actually pull information from that data.
And probably the biggest thing that we see,
we hear it, we send a lot of the,
the things you'll be having with your summit,
is a lot of people do it go after the idea that you can
take as many numbers out of the contact as possible,
right, by building automated data to the API.
And I think that's a really good option.
And we, we've been providing data to have that journey.
But what that means is it's gonna leave the most complex
conversation left in your contact.
And so, your agents are gonna need to report a new help
and new ways to actually continue to improve
that customer experience,
even though it's really the more complex issue.
To get that, and then the only thing that you know
about the market is consumer's needs
and trends are gonna change.
And while you may have done one solution,
it's a never-ending journey.
So, you know, when we work with our customers,
it's funny the question is really pretty basic.
Why did my customers call, what was going on,
why were they dissatisfied,
what were the things I could have done about it?
Like these are not super complex questions that,
if questions haven't gotten any more complex,
it's the sophistication in which you can answer those
questions that change a lot.
And as far as AI can really kind of make it,
it's just you know, we always say,
we just had the phone call.
We have everything we need to know about that customer journey,
everything that's going on.
Why would we need this perp to them?
Why would we need any other source of data
that this giant source of data
would just a tremendous amount of us were insights in it.
And our goal is just to leverage all the goals
and technologies available to help our customers
take full advantage of that.
- Yeah, I love that.
Especially that point of this idea between,
you know, you have this data,
yet we go out and we try to find like more data,
you know, via surveys.
And I want to bring you back there for a minute
'cause you've been playing in this space for a long time.
CSAT's one of those words that, you know,
everybody knows it, everybody knows what's been going on
with it based on some of the stuff you were just talking about.
What's kind of the current thinking in the market?
Where are you guys kind of around this idea of CSAT?
Where is it?
Where does it need to go?
Let's start big pictures.
We dive into some of these transformational things
you guys are doing.
- Yeah, so, you know, as we look at CSAT,
you know, like, well, let's just talk about surveys
for a minute because, you know,
you kind of have to have the ground set the ground
for like, what's the current state?
And don't be like, I'm going to pick on surveys,
but they had a, there's very few ways
a lot of information back from the cost actually,
their customers, their surveys are it.
It's an amp better tool.
So what, you know, what we can do now though is
that we can say, hey, I don't need that survey anymore,
right, and first of all, like, people aren't taking surveys
anymore, they're getting colder end,
the bias responses to those,
because only we're really happy and really at the people
who are on, they're not very timely, they're really costly.
And so, you know, we see that with all of our,
like they're spending all this time and energy
on the survey and in fact, they're not getting any really
meaningful, actionable data back to help them move the,
I move the ball and, you know, again, your comfort
sits right on the nose because if you're not leaving,
you're not taking the customer experience,
you're gonna get left behind.
Like it is a different career and it's gonna stay that way
for some period of time.
So we look at that, what we're trying to do is,
we're trying to think, you don't need the survey
and somebody asked me the other day,
are we trying to fill this E-stat survey?
I say, you know what, it's been killing itself for years.
You don't have to join the authority on its way out,
we just have to find a fitting replacement.
And that's what we're trying to do.
So we're trying to find that fitting replacement
for our customers where you can predict
the E-stat score on every single interactive.
So right at the end of that interactive,
you're gonna get a score and you're gonna know
that I have a dissatisfied customer.
And so you're gonna have that, now think about that,
100% of your interactions are the result on it,
not just 2%.
And you have the, you get the middle as well,
so you kind of get the really good balance
of everything that you're hearing from your customers.
And by the way, you have everything that they shared
about why they're dissatisfied.
You know if it's a product issue, you know if it's an 8 issue,
if it's a location issue, if it's a service issue,
a technician, you know, if you're always doing like,
you'll be able to really one uncover
what's dissatisfying your customers,
what is happening in that customer journey
that's causing that, and then give you really actual data
to then don't make business business
about where you wanna invest,
to actually improve that customer journey
and remove that dissatisfaction,
which of course we all care about
because we want our customers to stay around for a long time.
- Yeah, yeah.
I mean, I love your, I think most people would agree
that CSAT is broke, I love your point, it's killing itself.
It's been difficult, it's been difficult
to measure, manage, survey.
So I love the concept of trying to get away from that,
I guess maybe is the word I'd use on that.
How do you start, double click in this though for me,
I mean, I'm so old school,
I'm so used to seeing people,
I get a survey, I click a number,
I put in some comments like, how do you get around that?
I mean, how do you start to analyze interactions
in a way that still delivers that same type of information?
People are happy, you're not happy, how's it done?
- Yeah, that's the beauty of the concept of the fear
of, you know, artificial intelligence,
which is, you know, we now use the language based model
to actually predict based on the language interaction
on what they would have given you,
like a survey.
You don't have to send that survey,
you don't have to refer to it to come back
and you don't have to hope that it gave you any good verbatim.
You don't, it's all right there enough
where you can interact along with everything you shared
and one of the concepts we're really writing about
is like you said, what are public contact?
Like, you know, like I've been doing my post-care,
I've been super helpful, but at the end of the day,
there's this idea of like, are we all over 800?
Which is really the best part of the thing,
you just can't read them all.
I mean, like it's just like,
when you do get the data, it's hard to actually use it all.
But what we're trying to do is we're trying to say,
hey, what were the key moments you need to interact
in a really causal of the customer satisfaction?
Because if we can serve the code and then use AI
to actually bring together common themes,
you actually effectively have taken two steps.
You predicted that there was going to be dissatisfaction
and you've actually rolled up common themes
that are causing that dissatisfaction
and effect allowing you to get active against them
a lot faster, a lot more effectively
and on a much grander scale against your target audience.
(sighs)
- Yeah, yeah, I do feel like,
Steve, you've hit a really important point there,
that if you can mine the current data
and be able to pull out that juice, that goodness,
you know, you really are bypassed
and maybe even getting something that's far more accurate
because like you said, surveys are a little biased
and they run into some of these challenges
and et cetera, et cetera.
How do you feel like people can start to utilize
this type of data?
You know, I think sometimes the challenge we've had
with CSAT in the past is one dear point,
we can't really get through the verbatim, I love that.
But also it just becomes a score
and we can't really train or do much on it.
How have you been able to find and help organizations
think through that next step?
Okay, I'm looking at customer satisfaction
to dissatisfaction differently.
But what do I do with it?
Or how do I implement it?
Thoughts on that one?
Yeah, so it's been a lot of fun.
And so we've obviously had to test this with us
for quite a while and there's a few
that really love pushing yet.
You know, they're like giving this thing,
giving this thing or the thing.
And they just are a single thing.
But a couple of areas that we've seen,
one, you know, when you think about the contact center,
when you think about CSAT,
their moment of truth is when they're on that interact.
Well, of course, there's CSAT leaders,
product leaders and other folks inside the organization
that could use that data to go back and analyze
what they've heard about the customer journey and help.
But when that agent starts an interaction,
learn a moment of truth, they have to do something.
And so what we've seen is we've seen operational leaders
in the contact center actually start to focus in
on the agents themselves.
So in other words, did I have agents that actually produced
more dissatisfaction than other agents?
And how do I actually understand that?
And, you know, again, if you went back to,
I want to understand which interactions
are causing the satisfaction or resulting dissatisfaction.
Think of it, even if I knew the 2%
and I could attack the interaction,
it's a tiny percent of the interactions
and doesn't really give me a full picture.
So now on 100% of the conversations happening
inside your contact center,
you can look at it by team,
you can look at it by line of business,
you can look at it by,
so you can begin to understand,
are there organizational things that I've done?
Is there training that I have or have them provided?
Are there tools I have or have on the guild
that are starting the purpose of ways
to help that agent in the moment of truth
as they're trying to help a customer through a complex problem
or a difficult question on their journey?
And so one, obviously, is just how do we help the agents?
How do we help the agent win?
And as we see this, where we see it going,
is right now the company went in
and they looked at it and they made some suggestions.
What we really see this going to is that AI
will help surface that opportunity.
Think about it, like, hey,
well, based on this interaction,
there was a lot of satisfaction
and here's something vision could have done differently.
I could have changed its texture.
And that's a lot of fun.
The next thing that's really exciting
is now you're really automating the process
doing high quality input, helping your agent
decrease in your cost to do it all.
(laughs)
- Thank you.
- I'm not really getting a better agent
in your lowering cost.
That's gonna one great example.
The other great example here is
this idea of being able to track the customer journey.
So some applications all get wrapped up
through one of those stations.
Some are a service journey, right?
Do this, call me back, you do that,
make sure you go to the line and you're going to track
maybe a quarter of something new.
I got a continue email,
work ships, I always find out what's going on.
So what we've seen customers do is
obviously there's people that'll use this data
inside each other to do all the things we talked about.
But you can ship this data into other systems.
So you can ship it into a smart ramp.
You can ship it into a ticketing platform.
And you can set up conditions and say,
if I just had a conversation that was in dissatisfaction
about this product based on a shipping,
I want to put that in a few minutes review.
I'm gonna take a take.
What they do is they automate the case
off of that ESAP data and the others brought your data
and they determined to reach out a necessary
to go correct or wrong in a customer journey.
Really cool.
The other one is while we pick on their case,
what they're doing is they're a product company,
they sell through a lot of big distribution channels.
They have a lot of different brands
and they're trying to come back to their engineers,
the corporate product, to their packages,
to their distribution panel partners.
And that's one of the things they've been using
conversational functions for.
But with AI, again, it allows them to target interactions
to give them the best information.
But then they view it, they strip it into a CRN system.
And if it meets a certain criteria again,
they actually ship a follow up survey that's super targeted.
It's like, hey, we know we get that conversation with you.
We know that people go well
and we know that you were talking about this kind of kind.
Could you answer these three?
There's no question for us.
And now you're getting feedback that's just like,
everyone's gonna read those verbatim, right?
Because usually I just asked the customer about a problem.
They told me they had about a product
that I really care about.
And I'm gonna ship that to the engineers
so that they can do something about it.
So those are just some really, you know,
early examples of ways that we're seeing customers
kind of dive in.
- I love that.
That's super tangible, very actionable.
And I love that point just on actionability.
I think, again, to be able to surface that type of insight
so that agents can take potential to that type of action.
That, I think, is sometimes lacking
in these survey-based feedback mechanisms where it's like,
okay, what do I do with this?
Now I got a verbatim and a score.
I don't really know what I'm doing.
So love that you're trying to surface more actionability.
As we kind of look to wrap here, Steve, you know,
one of the challenges I think people are all trying to garner
or grapple with at the moment is, you know, is AI and CX?
You know, it's like, where do I start?
How do I do it?
What's the best place to go?
For people who are kind of starting this journey,
you know, what advice would you give them about AI and CX
so that they could successfully maybe take those first
couple steps in that direction?
Yeah, so, you know, I guess I've been doing this long time.
And by the way, what's really interesting about this time
is this is like one of the most areas of opportunity we've had
as business leaders in decades.
I mean, to think about, with the internet came along,
when Doccom came along, it just opened all these new
opportunities to work with our customers,
give them different experiences.
And this is one of those times, really exciting times.
Well, you know, based on that premise there,
as, you know, as the, as I've built a lot of programs
over the years, it's a customer-led opportunity,
which is where are your customer point points, right?
So let's just focus there.
But one of the things that we've really been focused on
over this past couple of years,
really just more the customer opinion,
but where's the money?
I think it was a caution I would give people.
'Cause I hear a lot of things coming in market,
but I don't see the ROI.
It's cool.
They might be very interesting.
And I'm a curious person too,
if my curiosity doesn't result in a business outcome
for my customers, we're probably not on the right track.
And so, you know, as we kind of scored through these things,
we look at our customer,
and we also look at the way that they,
where we can impact our business.
So in other words, if you can use AI
to reduce their cost of operations,
you have something that's meaningful.
And so we try to channel it more,
where I can specifically do things that,
you know, create better lifetime value or decrease churn.
Those are the things that we're focused on.
So early on in the stages here,
I'm just gonna follow the money kind of guy.
(laughs)
I care about customers, I care about my customers,
and we want them both to be really happy,
and that means you have to focus on customer pain points
and the ROI that's always talking about it.
It's just kind of an initial framework.
But, you know, the other thing that we've been doing
is we've been doing more rapid trumpet plaything.
So, you know, in the past,
we used to kind of like,
you built a new QA feature,
and it looked a lot like other QA features in the market,
and you didn't really need a lot of feedback on it,
and you could just kind of launch it and tweak it.
And, you know, we've done a much more iterative approach,
where we're actually putting the tools
in the hands of our customers very early.
We create what we call early adoption programs,
and we work together with them,
and they actually, right there on the journey with us,
kind of leading the way.
So, it's really just kind of, you know,
go in and stay focused on what matters.
- Yeah, yeah.
I think those are really wise ones,
especially the ROI.
There's so many fun little tools out there,
and I like your bright, they're cool,
but boy, at some point, you're gonna have to justify
those costs, so you might want to start now.
So, Steve, really appreciate the talk track.
If somebody wanted to, you know,
learn a little bit more about Tether,
or maybe get in contact with you guys,
what's the best way to do that?
- Well, it could reach me directly, Steve.Trier@tubble.com,
I'd be happy to chat with anybody
who's interested in having a conversation,
and talk more about this space.
Certainly, they could go to Tether.com,
and they could request a meeting,
and I'll certainly do it that way.
So, I'd be one of those things.
- Awesome.
I would definitely invite you to do that,
Steve, like you said, has been doing this for a little bit,
and have some great experiences,
both tactical and strategic,
I think, to share with the audience.
So, Steve, really appreciate you and Tether joining today,
and for the audience,
hope everyone has a fantastic day.
Take care.
- Thank you so much, King.
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)
(upbeat music)