Automating (some of) your CX

Release Date: February 11, 2020 • Episode #103

Customer experience professionals sometimes hit that wall of having to handle a high volume of tasks and customer interactions but there isn’t the peoplepower or resources to keep up. Strategically add a little automation to your processes and you could have a powerful CX tool that allows the customer to personalize their experience and relieve strained resources. Guest host Pat Gibbons welcome Ejieme Eromosele, Vice President of Customer Success at Snaps, to discuss the advantages of using automation and A.I. in your customer experience.

Transcript

CX Leader Podcast: "Automating (some of) Your CX" transcript powered by Sonix—the best audio to text transcription service

Download the “CX Leader Podcast: "Automating (some of) Your CX" audio file directly. This wav was automatically transcribed by Sonix (https://sonix.ai).

CX Leader Podcast: "Automating (some of) Your CX" was automatically transcribed by Sonix with the latest audio-to-text algorithms. This transcript may contain errors. Sonix is the best way to convert your audio to text in 2020.

Pat:
So many things in our lives today are automated, everything from paying bills, opening doors, even driving. So how many companies are implementing automation in their CX programs?

Ejieme:
Oftentimes we work with an individual functional area. So let's say it's a customer service team. Oftentimes they already have touchpoints where they're engaging with their customers already and they're looking for ways to drive efficiency, efficacy and to really continue to drive a great customer experience.

Pat:
How automation impacts the customer experience on this episode of the CX Leader podcast.

Announcer:
The CX Leader Podcast with Steve Walker is a production of Walker, an experience management firm that helps companies accelerate their CX success. Find out more at walkerinfo.com.

Pat:
Hello, everyone. I'm Pat Gibbons. I'm guest host of The CX Leader Podcast today. I'm filling in for Steve Walker and thanks for listening. On The CX Leader Podcast, we try to explore the most relevant topics for CX leaders and I'm excited about our our topic today. We live in a world of automation and most of the time we either take it for granted or we may not even notice it at all. Having tasks and business processes executed without the need for employee attention can create some cost savings efficiencies. So what about customer experience? What do our customers see? Well, strategically introducing automation into your CX program can potentially dial your customers experience up to eleven. But if it's implemented poorly, we all know it can hurt how customers perceive your company. My guest on this episode is Ejieme Eromosele. She's vice president of customer success at Snaps, a company that develops A.I. conversational solutions to help improve customer experience. She's going to help us understand the rewards, the pitfalls, everything we can fit into 15 or 20 minutes on automation and CX. Ejieme. Welcome to The CX Leader Podcast.

Ejieme:
Thanks, Pat. Thanks so much for having me. I'm really excited to speak to you about this topic and how some of the brands we get to work with these automation to really enhance their customer experiences.

Pat:
Oh, that's great. You know, we were talking ahead of time. This is a topic that I have… I've had a number of conversations with other CX professionals talking about how they make this transformation to provide a digital experience using automation techniques. And so I know it's highly relevant in CX today. So why don't we start by: give us a little bit of your background and how you got into CX.

Ejieme:
Yeah, that's a great question. I really started in the call center world. I started as a management consultant working at Accenture, working with, as you can imagine, large enterprises on their customer service transformation. This was back when outsourcing to places like the Philippines and India was just getting started. And so my role as part of the team was really in thinking about workflow and workforce management, helping to figure out what the best technological solutions in terms of ancient desktops could look like and help with rolling out those solutions across many different locations that these large companies had operations in. And so I did that for several years, then decided to leave consulting and to really be on the customer side. I got a great gig at The New York Times – the first role in customer experience for them. And I was really helping the Times think about how is their business model is evolving to be less advertising based and more individual subscriber base. What that meant for working across different teams, teams like marketing product and I.T. and also the newsroom. What that meant for really adopting a culture of subscriber first and putting a few mechanisms and a few capabilities in place for them to do that. Had a lot of fun doing that. The Times as a great organization, but I wanted to further challenge myself and sort of grow my skill set and my… my toolkit and future proof my skills in terms of what was happening in our space. And so I you know, I started exploring, started thinking about joining a technology company that was at the nexus of that and Snaps is… Is that is, in essence, that we're helping build platforms and experiences that top brands can use to meet their consumers where they are and to scale resources to provide faster, always on personalized service and automation for those users.

Pat:
Yeah, it seems like a really cool business. And I think sometimes when… when people think of automation in business, they may focus a little too narrowly on the contact center. But you're you are broader than that. It looks like you work with customers at all stages and… and it looks like you're primarily in the consumer space. Is that correct?

Ejieme:
That's right. So we primarily work with brands that have an end customer that's an individual consumer. And so those are companies that are in retail, ecommerce, consumer packaged goods. Travel and hospitality and media and entertainment. We do a little bit of financial services as well. But I think you hit on the key that, you know, experiences are not discrete and siloed anymore. And so we wanted to have brands have the ability to market and gain awareness to a subset of consumers, really prime them pre-purchased, help them with that purchase decision by providing personalized recommendations and guided shopping, but then also being there when they needed help post-purchase and needed a bit of support. So our platform can really help brands enable automation across many different channels, across many of those customer touchpoints.

Pat:
Yeah, I love that. I think the idea of the seamless customer experience. You know, in most businesses we you know, we compartmentalize and we do that because it's… it's an efficient way to get things done, but the customer doesn't see it that way. So maybe paint a little picture for us: how… how does automation work when you're trying to kind of look at the entire customer experience and how you… you kind of digitize it and use automation to make that happen?

Ejieme:
Yeah, that's a great question. When we work with brands, when we engage early on, it's really about understanding, you know, what the brand's strategic priorities are and which team we're initially engaged with. I think now we can be very fortunate to engage with a team that has ownership of the customer experience or the digital experience, because that way we really have the green light to plug in to many different touchpoints. But oftentimes we work with an individual functional area. So let's say it's a customer service team. Oftentimes they already have touchpoints where they're engaging with their customers already and they're looking for ways to drive efficiency, efficacy and to really continue to drive a great customer experience. So we really start with understanding what their top use cases or top contact drivers already are. What are some of those challenges? What are their customers looking to do with them? And once we have that understanding, we try to take a guess at what types of those use cases are potentially highly automateable. So these are things that are more transaction base. They're also use cases that we can tap into a preexisting data source to feed the automation with an initial set of data and then over time to get the automation smarter. And so we really like to start there because that's sort of the low hanging fruit, if you will, of, you know, let's start with something simple and then scale out to more of the complex use cases. We also advise our customers to… we call it throttling the experience. So we work with them to sort of ramp up how much exposure their customers have with this new… new tool and new technology. Oftentimes you start with just 5 percent of users see the automated chat experience and then went back up to 20 and 50 and so on till we get to 100 percent. That gives us a good testing ground to be able to use real life customer data, but not have it be a risky endeavor from the brand as we as we fine tune and tailor that automation to work better.

Pat:
Yeah. I mean it does really. You know, when when a company goes through this kind of transformation, you really have to be sensitive to how it's going to affect their brand, right?

Ejieme:
Absolutely. And we work with very large enterprise brands. And so there is a sense of wanting to do right by their customers and not wanting to put something out in market and put something in front of them. That may not be the best work.

Pat:
Yeah. So can you give us an example? Name names if you can, but if you can't, we we understand maybe of ones that, you know, took that initial step and kind of grew into it.

Ejieme:
Yeah. So one of our customers is a plus sized segment retailer. They've got physical retail locations and a strong digital presence as well. And we work with them to deploy automation across the customer journey. So from marketing through purchase and care and across two different touchpoints for them as well. And the way we started with that was really understanding what was their key challenge. And they had two. One was really being able to increase conversion on their website, and the other was to reduce the chat volume on an omni channel that they had just opened. And so we wanted to test both of those use cases in the same digital experience. And we wanted to start with a lower risk area for them. That low risk area was on social. So we first deployed a Facebook messenger experience that had a guided shopping flow that had personalized recommendations to really address the first challenge of increasing conversion. And the second challenge we incorporated a where my order flow, "WIsMO" is the shorthand acronym, to help their… their users get proactive notifications. So hopefully deflect volume. And then also, if they came back to the experience and had a question about where their order was, we tapped into their order management system to be able to serve them back some details about the status of that order. So we started on Facebook Messenger because we know that, you know, it's not an open channel. They saw a lot less traffic there and we deploy that over time for a few months. Once we felt really good about that, then we took that automation and deployed on their website, which as you can imagine for a retailer and a dot… with a dot com presence, that's that's the Holy Grail. And even when we deployed on the website, we throttled with 5 percent of traffic of users seeing that the chat bot and seeing the automation first. And then again, we knew that there may be a difference in the users that are going to Facebook and what they're asking and how they're using the bot versus those who are coming to the website. So we wanted to see how are some of our initial hypothesis changed. And so we started with 5 percent, fine-tuned from there, and now I'm happy to say that we're at 100 percent of traffic on their website. And it was a journey over a several months journey. But I think we took that approach to test and learn and to de-risk this, you know, really sophisticated technology for them and also took the opportunity to get by-in and support from other teams and make sure that we got the resources required to really make this worthwhile.

Pat:
So I wanted to ask a couple of questions about buy in, because I would imagine that this can be a rather intimidating process for the company and that even if you have an enthusiastic person leading the charge that there's going to be others in the company is saying, I don't know about this then and all that. What are some of the biggest hesitations that people have when when they're entering into this type of process?

Ejieme:
That's a good question, I think we often hear questions around, one, you know what data is required, how much data do you need to get something up and running? How clean and accurate should that data be? And so there's a lot of questions around just, you know, what is the starting point to feed the automation? I think a lot of people now are a little bit more sophisticated about how some of these technologies work. So they realize that the bot is only going to be as good as what what we feed it. So I think there's a lot of questions around kind of just data integrity and accuracy. There's also a sense of any questions around data, there's always a security question, as well. You know, how are you using that data? Where does that data live? How are we making sure that if there is an E-component. Are we GDPR compliant? How are we making sure we're GDPR compliant? Now we have CCPA in California. So anytime we talk about data inevitably we also need to talk about security and what we're doing to make sure that the data that we're using is specific to the use cases that we're using them for and that if users need access to that data, or need, that data erased that we've got the right mechanisms and protocols to work with the brand to make sure that that happens. The other area that we get a lot of questions on are, you know, who owns this going forward and what's the amount of work required to keep something like this going? and to keep something like this improved. And so the short answer is that you do need someone, you do need a team, you do need a champion to own this and to make sure that the automation needs to be trained and needs to be optimized and it needs to learn. But that learning doesn't happen organically, that learning happens via human intervention. And so that's one of the areas that we consult our customers on and we've actually built some flexibility in our model where, you know, the team that I run actually have a team of conversational designers and it's their job essentially to do that. I do think that over time in the next 10 years, more companies are going to have conversational designers. They're going to have automation experts because it's going to become more and part and parcel to how they enhance their digital experiences, because over time, I think that this will end up being a core capability that more companies have internally.

Pat:
Yeah, I mean, without question, I think it makes sense. I think… you know, I'm assuming that a lot of companies… I guess I'm just curious from your perspective, you know, I feel like there's kind of an adage if you do something regularly and it's got like a three step process to it, and you're doing it all the time, you're like, okay, how can I automate this?

Ejieme:
Yes, exactly.

Pat:
And if that comes, I would think a lot of companies initially answer this, saying, hey, we can make ourselves more efficient and there may not be thinking as closely about the customer. So I guess my question is maybe a two parter. One, when people come to you, what what is it that's on their mind? What are some of their objectives and do they have them, you know, sort it out? And then second, are there times where there are certain experiences that you guide them either to stay away from or to say, you know, this is risky if you are going to mess with this part of the customer experience?

Ejieme:
That's right. Yeah, so to address the first part, one of the typical objectives that we see when a customer, a person that's about automation, if I can be specific and say let's… let's focus on let's say the customer service or customer experience leader, I think they're really looking to drive efficiency. Their businesses are growing, their businesses are scaling, and they can't scale their… that cost structure accordingly. They can't hire more people as fast as their business is growing. And so they're looking for a way to scale and to be efficient in scaling. And so what they're looking to do is reduce over time the volume of chats or calls or increase that are going to their human agent pool. I do find that more of the customer service leaders we're talking to there, they're being more sophisticated in their thinking. So it's less about reducing headcount. It's more about reallocating head counts, more value added work. So how do you use automation to take care of the transactional wrote questions that a bot is well suited to answer and customers honestly are happy to… to self service on. And so after you do that, then what are those types of questions that are more complex, more complicated, or potentially a revenue-generating increase that can be handed to a human agent to complete? So it's less about a reduction and it's more about reallocation of those resources I think is smart. They're also looking, though, to maintain a great customer experience. I think, you know, when automation was rolled out, whether you look at automation through things like IVR as in the 90s and even some of that first generation chat bots, they weren't very good.

Yeah.

Pat:
They were pretty static. They're terrible experiences. And I think we always though, probably a little bit traumatized from the experiences personally and then also as leaders in the states. And so I think the anchoring mechanism, the anchoring metric time and time again I am hearing from our partners is customer satisfaction, NPS, customer effort score. How do we make sure that this thing that… that, you know, we want to use to drive efficiency in our business also isn't pissing our customers off. And so it is a metric that we're also tracking. And the great thing is that in many of our experiences, we find that when automation and a… and a human agent work together, there's not a lot of difference in NPS or customer satisfaction. In… in fact, because automation is part of that process, an agent can get to a customer faster. So it has this downstream effect on the customer's experience because, guess what, they have to wait less. When an agent gets their case, they don't have to repeat themselves. So we find that augmenting that workflow with automation can actually enhance the customer experience if done well.

Pat:
Yeah, that makes sense. So if somebody is trying to get started, what are some of the first steps that you recommend?

Ejieme:
Great question. So… I would say the first step is just, reanchor your thinking? I think some folks approach this as a tactic. I think this… using automation is actually a strategy. And with that, with strategies come, you know, being really clear on what outcomes are trying to achieve, what use cases you want to start with initially, what set of metrics we want to use to measure success and how do you get other teams that are adjacent and that touch part of this experience, how do you get them involved in bot in? So I think a leader really needs to think of this as a strategy and as such also have an owner and a champion internally, whether that's them or whether that's someone on their team. But you're going to need someone to kind of own this journey because it will be a journey in your organization. I think the next area is really to do a little bit of consumer insights and ingathering. So it's part of any strategic effort, you want to understand where your customers are already going to reach out to you. Are you already on those channels? If not, why are you not on those channels? Then what are the things that you need to do to be where they are? And then when they are reaching out to you on those channels, what types of questions are they asking? What are they looking to do? So really understanding your customer base and how they're navigating with you or honestly without you in the digital ecosystem is a great starting point. And then lastly, what data do you have internally to help either feed some of that consumer insights or if you do go down an automation path, start to feed whatever tool and whatever process or workflows you'll be, you'll be standing up over time.

Pat:
That's a great list, great list. So let's… let's say for a moment that a company goes through the journey and they implement a strategy like this flawlessly. What are the biggest benefits that they're going to going to get out of it?

Ejieme:
Yeah. And so I think we hit on some earlier. But what we're seeing in terms of the success of some of our brands are… are finding, you know, there's definitely an efficiency aspect. So the amount of tasks that the agents are seeing are less. So we're being able to contain the automation experience, many of the more rote and transactional chats that that team typically sees. We also see that we can increase the effectiveness of those agents. So being able to have an agency, a transcript of what a user was trying to do and what questions the user was trying to answer before they're handed that user helps the agent get started faster. They have more context and they can get started before having to respond to that customer. We're also seeing that our brands – and this is something that we're learning over time as well – they're also getting a richer set of data and consumer insights on what their customers are looking for. We find that when you deploy an automation solution, let's say on your website, the way that your consumers are interacting with that experience, looks very different than the way that they would be interacting with a search bar on that website. They're asking questions in a more conversational way. They're asking questions in a more nuanced way. And they're asking for personalization. And so we're finding that we can actually enrichen the source of data and the consumer insights that a brand can have. And honestly, sometimes even act as an early warning trigger for things that the call center thinks of customer experience team may not know it's happening yet because if you open up these channels, consumers can type and ask questions and alert, alert these teams to issues and other parts of their business.

Pat:
So more efficiency, better skills, better data, ultimately resulting in a better experience and a better brand for the company. That sounds like a pretty good formula. So we're… we're at the point of our program where we ask you kind of our hallmark question, and that is we call it our take home value question. You've given some great advice. If you had to pick one of those things or maybe something that you haven't mentioned, if you had one piece of advice to somebody that was looking at this topic, customer experience professional, that they wanted to take the first step either within their company or to embark on this journey. What would be your first piece of advice?

Ejieme:
Yeah, I would say don't be afraid to get started. Consumers want more sophisticated ways to help themselves. I don't know about you, but the last thing I want to do is call a brand and and be be on hold trying to get help. So don't be afraid to get started. Consumers want this. The technology has really come a long way. And instead of rules-based automation, now there's A.I. based automation so there is context. There's personalization. And these solutions are yielding a lot better results and a lot better end-user experiences.

Pat:
That's good advice. You know, and I think about it and your comments earlier about where all this is headed, if you're not getting started, you might feel like you're already behind. So…

Ejieme:
That's right.

Pat:
Ejieme Eromosele is vice president of customer success at Snaps. Ejieme, thank you for being on The CX Leader Podcast today.

Ejieme:
Thanks for having me Pat.

Pat:
And I would also recommend, you know, just in learning a little more about Snaps, if you are taking that first step and you're getting over your fear of getting started, you should check them out. They do some really cool stuff. So check out Snaps and learn more about this whole area. And naturally, if you want to talk about anything you've heard on this podcast or how Walker can help you with your customer experience, feel free to email me at pgibbons@walkerinfo.com. Or you can get a hold of Steve Walker at steve.walker@walkerinformation.com. And if you haven't already subscribed, please do that. The place to go is our new website which is cxleaderpodcast.com. You'll find our previous episodes, we've got lots of great stuff. And the other thing is we're always looking for feedback and interest in other things you'd like to hear about. So please check out the website, send us a message, let us know how we're doing. The CX Leader Podcast is a production of Walker, we're an experience management firm that helps companies accelerate their CX success. You can read more about us at walkerinfo.com. Thanks for listening and we'll see you next time.