247: Fireside with Austin Emser

Charlotte Ward: 0:12

Hello and welcome to episode two hundred and forty-seven of the Customer Support Leaders Podcast. I'm Charlotte Ward. Today, welcome Austin Emser for a fireside chat. I would like to welcome to the podcast. Today for the first time, Austin Emser. Austin, lovely to meet you. For the benefit of all of our listeners, would you like to introduce yourself?

Austin Emser: 0:44

Uh happy to and and thank you so much for having me. Uh I'm the co-founder and and CEO of Stylo.

Charlotte Ward: 0:54

Sure, but sweet. Thank you. Thank you so much. Okay.

Austin Emser: 0:57

Nobody hear me, Ramble on.

Charlotte Ward: 1:00

No, that says it all. That says it all. That's all we need to know. Right there, right there. Well, thanks for joining me. Um, and uh what what we had to talk about today?

Austin Emser: 1:10

Uh today we're gonna talk about prioritization. Um, I think especially with all the budget cuts right now, it's certainly a hot topic. And I I can I can already hear some of the eye rolls for the the future audience listening to this, but um I think it's a lot more complicated than people chalk it up to be.

Charlotte Ward: 1:30

And I I mean it's a super interesting topic because priority uh has um it it comes to bear how we prioritize our work in so many different ways in support. Like I we could talk, we could talk OKRs and we could talk talk tickets and anything in between, couldn't we? Sure. Uh well let's do that then. There you go. How's that for an opening? Let's talk it all.

Austin Emser: 1:53

Well, love it. Yeah, let's just jump right in. Um and I mean to uh to to back up a little bit, like what I was alluding to, some of the market changes and and some of the prior podcasts here. I I know there's been some great conversations on you know when to hire, how to hire. Um, and a lot of those rules are, you know, once we're getting to a certain number of bandwidth, then we need to start hiring. But like fact is today, especially in tech, you go to start asking for headcount, and that's that's no longer being approved. Like uh we're we're seeing a lot of budget cuts, right? Um, and fact is I think the general rule and support that you always hear, right, is like no ball can drop. We have to do every ticket in the SLAs that we lay out. And that is like I get the theory, but fact is that's not possible. Like having the same or increased number of tickets with less people, there's some ways that we could drive efficiency. And actually, I think in some ways this is forcing us to really think about that instead of just throwing headcount um at the problem. Uh, but at the same time, especially in this transitionary period, you have to prioritize the most important things. Um I'm curious, um, you know, Charlotte, in in your uh uh experience, like what are some of the ways that, you know, given all the tickets coming into the queue, like what are ways that you've prioritized um coming in?

Charlotte Ward: 3:13

Wow, asking me questions so early, that's bold. Um, okay. So oh wow. I like look, I mean, you said it all at the start, and I think this is a particularly, these are particularly straight, straightened times. That's the word, isn't it? Straightened times. But um it's I've been in support a long time, and I don't think this story in its broad strokes is new. Like the the thing that all people on the front line in support want to do is is get more people on the team, and every support lead knows that you have to do more with less people. So like it's just a constant battle. Uh, and headcount is you're right, being approved even less now than it has been historically. But let's face it, headcount is never abundance. So you've got to have, I think, I think you have to have a mind on this all the time. Um, and I think um the other thing that you said is interesting, which is that the work has to be done, you know. We we do have a constant flow of incoming tickets, we do have SLAs we have to meet. Um so I think that the the first thing, and like I absolutely tell every team who've ever led this, and and have to remind myself this occasionally when I was on the front line, which is that you actually can't do it all. Balls will be dropped, SLAs occasionally will be missed, and you've got to be okay with that. Um, you just have to find ways of managing that, you know. So I think it's super important that you have the follow-through when problems happen so that you can post mortem them and examine them for like really was this really simply a headcount issue, or is there something like I hate the phrase, like working smarter, not harder? But it is frankly, is there something we can be a bit cleverer about here? You know, I think that's really important. Um I'm giving you a really long answer here, but I I just want to round out on your actual question now. The the actual answer to your actual question, which was how do I prioritize on the front line? Um I think it ha you have to have a really clear understanding of what you mean by priority. And I and I think that that's what it comes down to for me, because priority means different things to different people. But if you understand what you mean by priority in your organization, then you can shape the way you operationalize around that. So so I think you know, priority can mean serving particular uh you know, high-value customers, it can mean serving more urgent tickets, it can mean serving customers at a particular stage in their life cycle. And and priority then I think is also nuanced. So there's definitely priority, which is the thing that is most important. But then I would also add two other flavours of priority. The the other thing I was talking about is severity and then urgency. And I really see them as three very different things. The priority is the order in which it's where you put it on your to-do list, effectively. Your uh urgent urgency is something that is kind of somewhat artificially created. It's usually driven by people who put deadlines on things, driven by timescales. Severity is all about impact. And and by that, as a support person, I mean impact to the customer. That was a really long answer, but that's everything unpacked. How about you? I'm gonna I'm gonna return the question. I mean, how how do you prioritize?

Austin Emser: 6:59

I I like like that framework a lot. Um, and you know, really, you know, with that concept of let's let's roll with that concept of priority being, you know, what in what order things happened. Um, to your point, there's urgency and severity. But then there's a lot of times these objective and subjective components. So let's break that out for a quick moment, right? So let's start with objective. So the most common thing that we see, unfortunately, is first in, first out with some sort of um, I I've heard different support leaders call it like grooming, where maybe at the top of the hour, um certain number of support reps are dedicated to reading through, trying to figure out, hey, do we see anything that's more important that needs to be bumped up or down?

Charlotte Ward: 7:41

A triage system, I guess.

Austin Emser: 7:43

Exactly. But like, are all customers created equal? Are all their problems created equal? It's, you know, uh I I've seen uh some people make it so they're freemium users with a password reset request. They're you know, three tickets ahead of one of their most critical customers um spending a ton of money on an enterprise contract that has an outage because they're just doing first in, first out. And it's a systematic manner um to do that. Um so first in, first out, um that that's certainly the most common, but some other ones we see um occasionally something like uh service tiers. So we see something that you know maybe customers buy into platinum, gold, or diamond or whatever gemstone we can think of that uh we think we could get more money for. Um uh then you have uh customer revenue. Don't see that a ton, but that that is an important component here.

Charlotte Ward: 8:35

Sure. Yeah, yeah.

Austin Emser: 8:36

Um, but then, you know, I I think one of the best, uh, but it is hard to implement in practice just due to the amount of data is you know that the different ticket reasons, and that goes into both the severity and the urgency factors that you're talking about. So um, you know, let's take, for example, uh there's let's go really extreme, something like an outage, right, where multiple people aren't able to use the system. Maybe it's something business critical. That's going to be very high urgency and very high severity. Um, so being able to understand, you know, what that ticket is about um in order to figure out how to prioritize that in the queue. Now, uh, you know, the tricky part there, as I mentioned, is you either have to have, if you have a very high volume, maybe you could afford a triage uh team member who sits in front of, you know, tier one or or whatever that that happens to be called in the nomenclature. Literally their whole job just going through tagging things, trying to figure out priority, and where does it go? Most of us, unfortunately, don't have the sort of resources in order to hire somebody who only does that triaging. So then it kind of comes back to some of that uh that cue combing that we talked about earlier, where comes a shared responsibility and everybody has different different questions.

Charlotte Ward: 9:45

I I think I think I would also add that my gut feel is that's a deeply unfulfilling role as well. So I would want to I would want to stray away. I wouldn't want to do that, not full time. So I mean, I think I think like being creative, like if that's the direction you want to go, if you want a triage mechanism. Um, and and actually I've implemented this here and in in the in my current place and in other places, is thinking about ways you share that out, you know, whether it's time slicing or whether it's some clever automation AR, like what whatever you do, it's like how you how you just take the that soul-destroying human load off one person and and distribute it somehow.

Austin Emser: 10:30

Yeah, exactly. And and you know, I I think that's an example where some of this is great, is a stepping stone for this is a fantastic automation use case, right? Most jobs where we could say, oh my gosh, that would be the worst thing I've ever heard of. Um, I don't want to do that. Um, it's very tactical. That's probably something good for automation here, right? So starting to create formulas, and and again, this depends on every business. So there's not a one size fits all, but saying, hey, you know, for for one, um, you know, if if we're using Zendesk in Salesforce, let's make sure that we start getting our revenue information put into uh uh at account basis into Zendesk. Let's use revenue is one component or tiers of customers. Some people have, you know, enterprise or SMB, whatever it happens to be called. So that's kind of our baseline. But then with that, um, let's also include things like uh some of those ticket reasons. So um Zendesk, also Salesforce has their version of this. Um, you know, well, there are lots of tools out there to cover all of them, but everybody has some sort of concept of tagging or understanding what's in the ticket, right? Um, granted, I understand it can be really hard. It's like, how do we think of a million different synonyms for the same exact thing? Because customers will will really push that to the edges. But like, I think a lot of times that gets overwhelming, and people say, Oh, I could never think about all the different ways that that people are able to say things. Um, so I'm just not going to do it at all. But the fact is, even if you get 60 or 70% of the way there, that's gonna be a lot better than what you have today. So just start somewhere and start having those systems automatically, even if it's simple, tagging some of those most critical tickets. At least be looking for something like outages or the most critical issues. Um, tie into some of those revenue numbers, and then on at least on the objective side of this component, we're already starting to build a much more effective prioritization engine than just that first and first app that we started with.

Charlotte Ward: 12:25

You really start to get a sense you only need a couple of data points as you've described. It's like maybe it's maybe it's uh problem type, maybe it's ARR. You you could literally draw uh you know an X and Y axis of almost any two things that you think might be important to your business, but they're definitely like really common ones. And and in that X and Y axis, you know, you've got the top right quadrant is where you want to spend more time, more human effort, it's higher value, and the bottom left is where you want to be automating. And I think that's just like identifying what are the value components and like how you want to respond to them and just getting rid of the stuff that's in the bottom left of that kind of crosshairs graph is kind of really simple. And and you don't have to hit everything in one go, you know. As you said, if you can identify a chunk of tickets or a chunk of work, a bit a part of the process, you just take a bit of it away. It frees up everyone's mind very quickly to think about what's what's phase B of this, what's the next chunk, what's the next thing. Um I I have a real like something that I will ask my team to do quite often is think about what they would like to do less of in a day. And it's just a reason. I like that, you know, it's like what is just the thing that you just want to do, the thing that you say a hundred times, the thing that you do 20 times an hour, the you know, the thing that is uh uh just a click that you just don't want to do anymore. Let's find a way to make it not happen. Because that's time as well as as like, you know, uh engagement and and and you know, exciting work, isn't it? It's time, which is money.

Austin Emser: 14:03

Yeah, it's time and frustration. And and earlier with the example of triage, you mentioned that's not a job somebody wants to do long term. Like that, I mean, you know, this is an entirely other like three-hour podcast we could have, but like um uh unfortunately support is a role that so many people want to get out of because you know, we structure it in this way where you know we don't automate the things that we necessarily could. And it's those highly monotonous tasks that are just really frustrating.

Charlotte Ward: 14:29

Yeah, yeah. And you know, in an ideal world, everything would be fixed upstream, the product would be perfect, the documentation would be perfect. It all works, but then we wouldn't probably have a job, but you know, we'd have a knowledge base that was all self-served, like you'd sit there all day waiting for the one perfect consultant consultative level ticket, you know, that's gonna take your entire day and it's really interesting work. But you know, all you can do is like tend towards that is aim for that, but you're probably never gonna get there.

Austin Emser: 15:00

Exactly. You you can do with what you have, but you know, uh this granted, we're we're starting to get a little bit out of prioritization, but like I think one great example here is so actually I had um I took an Uber uh a few few weeks ago now, and it decided that the best point to let me off of was in the middle of a river so that I could walk through swim over the river to get to where I wanted to go. Because clearly that saved the driver some time. And so, you know, go to Uber support, but Uber support doesn't give you a person, it gives you an automated chatbot, right? And you sit there trying to explain what a river is to a chatbot, and it's just so frustrating. You know, you we talk about automating some of these uh you know, support as much as we can, but like fact is like granted, AI chatbots work really good um for a variety of use cases, um, but like it doesn't solve every problem, and and especially at these more complex, um, you know, especially B2B environments where maybe maybe there's all sorts of engineering constraints and questions in here as well. It's like we do need reps and we can't automate everything away. Um and yet things like this prioritization are things that we can automate away, whether it's for highly complex B2B or the most simplistic B2C solution.

Charlotte Ward: 16:17

Yeah, I I couldn't agree more. Like I think that the that making the work interesting is the the single thing we should be trying to do for our teams. Because if we make the work interesting, we are we are building careers for them, but we're definitely solving customers' problems better as well. And and I and and and we're doing that in multiple ways. We're we're doing it because we're providing more nuanced, complex human answers to things like the Uber River problem or automate, like in my world, it's all B2B deep engineering like support. Um so both of those require quite a human level of interaction. Um but it but in doing that, like we're we're also responding much more appropriately to our customers in in terms of their business value to us. I think that it just goes hand in hand, right?

Austin Emser: 17:13

I think um your point on that that human element is a perfect segue into the subjective parts of prioritization, right?

Charlotte Ward: 17:21

Oh, detail. Yeah.

Austin Emser: 17:24

So the challenge, uh right, it can't all just be objective, right? Because uh there's so many different things that that could happen here. Um, you know, maybe a customer starts out really chill, one of your favorite best customers, they're just getting progressively more frustrated, they're not getting the answers that they need. Um, you know, things are really starting to catch on fire, and and we need to identify the situations, but it is so hard as a human to objectively rate that nature, that subjective element of what's in front of you. Because I know whenever I'm involved with the ticket, it feels like the most important ticket. And then if you ask customers how important their ticket is, well, we all know that's going to be a P1, regardless of what it's about. Um, so every rep feels they have the most important ticket, every customer feels that they have the most important ticket. How do we start to understand some of these subjective elements? And and that's where it's, you know, that some of this gets gets really difficult. Um, you know, there's different solutions out on the market for it. But I think being able to understand, you know, some of those emotional responses with um you know technology is super critical because again, without that, we're trying to build a sub objective, subjective score, and then everybody has their own opinions and it just turns into the spiraling mess. But yet I think we can all acknowledge, even though it's very difficult to do with humans and reps, um, it's a really critical component of that prioritization.

Charlotte Ward: 18:51

Yeah, I I couldn't agree more. Um, I have a slightly um, I have a related funny story, actually. Uh going back uh 15 plus years, um into the depths of my support career. When I was on when I was on the front line, um we had a customer, a B2B support, you tend to develop longer running relationships with your customers, tickets run for a long time, you get to know them pretty well. We had one customer for whom very subjective opinion. Every single ticket he raised was a severity one. Everything that attracts that that's that's a subjective view, if ever there was one. And and what do you do when you've got 10 tickets from the same customer and they're all severity ones? Uh, well, we had a tried and tested uh human triage mechanism, which was that we looked at his opening uh problem statement, and you could tell how important it was actually to him in the nuanced levels of severity one that he was actually playing in, by how many exclamation marks there were. In his opening communication. And so I'm I'm not even kidding. Like we we treated his tickets largely by how much urgency he was putting in the language, effectively. But it it was there was a high correlation to exclamation marks. I can tell you that.

Austin Emser: 20:19

Yeah, no doubt. And I mean, that's another thing where, you know, fortunately, I'm seeing some advances in technology, whereas, you know, five, ten years ago, everything was kind of treated as a one-off. But now, um, you know, being able to take into account what is this user or company's baseline? Because some people are just angry people, but yet on the flip side, some people are really happy people. And when they come in and they have a really critical request, that I mean, that urgency is even that much higher than it would be for anybody else.

Charlotte Ward: 20:52

Yeah, yeah. Um, this just brings to mind one of my very, very, very first podcasts where um I was talking to Ash Rhodes. It was the first week, the opening week of the podcast. I did five, five episodes on empathy and support. And um, Ash Rhodes said something like, you know, you have to kind of unbegin to understand the language and how one person might be shouting and screaming in, you know, in capital letters at you, and another person in a different ticket is saying, Well, that's not ideal, you know, and they could be equally angry, but like you have to understand, yeah, like understanding your customer's behavior is really, really important if you're looking at the subjective nature of it.

Austin Emser: 21:37

No, no doubt. It it really is. But um, I mean, especially as the support team grows, that's reasonable to do with five reps, 10 reps, maybe even 15 reps, 20 reps. Once you get to 50, 100 plus, it like you're just not staying on top of that. Um but it's like, you know, whereas the objective, um, you know, some of this objective prioritization is a little bit easier to do more with with uh some built-in tooling, this subjective nature. Um, granted, that that's something where we're a lot of times you kind of have to look look outside of the box, but yet it's such a critical component of that. So let's, you know, start to put that formula together. So we have the customer revenue, we're kind of their tier. We're understanding, you know, what are some of those those reasons that they're reaching out that we do know are really important escalations. Maybe we don't focus on on automating, um understanding every single problem that comes in, but we do have the most important ones. We're not gonna boil the ocean. Then on top of that, we could start to add in some additional things around sentiment on uh scores to figure out, you know, given this customer revenue, maybe it isn't as high as others and their issue wasn't as critical, yet they're really, really upset is gonna bump them into that next year of prioritization. So it's just, you know, with this, it's starting small, um, focusing on the things that we know are impactful, and then slowly building it up from there. Because, you know, we see a lot of people who want to, you know, just do everything at once. And granted, we all want to do that, but you know, that's unfortunately not the nature of this piece. And you just get overwhelmed and then you give up at some point, and then you go back to rep coming in first in, first out, just just like we were before.

Charlotte Ward: 23:16

Yeah, yeah. I I think you're absolutely right. I think like just um taking the simplest thing, like really understand that you're gonna have to sequence this, sequence it and and validate each stage, because actually what you're also describing is a set of assumptions, you know. And I I think that you know, um you're making assumptions about like you know what's important to your own business, but making assumptions about how that reflects and manifests for your customers and what they are um what they're feeling about the service you provide, like that those validation points are really critical as you work through each stage of this. So you start with your broad, broad that that crosshairs graph, you know, like what's the big chunk we can do? Well, let's work that and then let's see the effect of that before we move forward, and then you can get increasingly nuanced. And and and the fact is you probably don't need to boil the ocean, even if you are considered and and roll it out in a staged manner and take your time, take years to develop it even. But but the um I think the fact is that there'll come a point where you've done enough. And the ROI, the ROI on any of this just starts to be minimal eventually.

Austin Emser: 24:33

Yeah. Well, and it's it's all about that top of pyramid for the triage. Like we gotta focus on the really, really critical stuff. And granted, you know, maybe we messed up some things really low down on that pyramid, and it's not great, but it's also not the end of the world. What we need to avoid is accidentally putting that P1 at a P3 because we missed it, or a rep didn't understand the importance, or you know, a multitude of reasons for that. And really, I think that's where you know that the next most important thing is really around communication. And I mean that with both the business and the customer. It's like again, you know, oh, we can't drop any balls, everything has to be done within the SLAs. That's not going to happen. We have to accept that. Um, and given, you know, when we really start thinking about what we are prioritizing and starting to put in some of these automations, we need to communicate both with the business on what we're prioritizing to get buy-in, saying, hey, this is what we're moving up to the top. I want to get to all these tickets. That might not happen because we have to focus on this. Um, if you don't like that, by the way, we could have more headcount. Um, but if you're okay with that, that's what we're moving forward on. And on the customer end, you know, I I think uh most of us can identify it's like there's nothing more frustrating than not knowing how long a wait's gonna take or being told, like, oh yeah, the SLA is 24 hours, but yet, you know, we're we're 36, 48 hours plus deep in a key right now. That has to be realistically communicated because yeah, I don't want to wait 48 hours, but it's better than being told that it's gonna be answered in 12 hours and then me waiting the 48 hours.

Charlotte Ward: 26:11

Yeah, yeah. I I couldn't agree more. I mean, I mean, the the old adage of like uh um underpromise and overdeliver is is just it's so simple, but tell the customer 48 hours and take 36, they'll be much happier than if you said 24 and took 36, right? That's really that's just a no-brainer in my book. Um but then I think the the the the other thing that you talked about um was talking to the business about this. And I think that that is um a really interesting conversation to have because in the early days, particularly if you're the first support hire or the first like professional support lead in a business, probably the rest of the business doesn't know support as well as you. Uh and so they have probably quite a simplistic view of what a good support looks like and and you know, our ability to respond, like if at the very simplest level, if you've got a first response SLA and and you're you're hitting a certain service level because you don't have any of the the nuance in place that allows you to differentiate your services, um, that's that can be not only what your customers come to expect, which is a problem in itself, but also what the business comes to expect. And so I I've seen some interesting transitions in various roles I've had where I've put that first um that first differentiation in, right? We've got these low value customers and these high value customers, and we're gonna spend, we're gonna be more responsive and spend more time with our high value customers. Um, that means that actually when you look at the previous average, your lower value customers are getting effectively in little bunny ear quotes here, which my listeners won't be able to see, but you know, the in the kind of they're being underserved compared to where they were. And that's kind of an interesting narrative to tell a business because if you're still reporting in the same old way and you're saying, yes, our first response time has gone from an average of, you know, two hours 20 minutes to now 18 hours 40 minutes, um, it's because what you're really doing is serving appropriately. And I think it's really uh it's the job of the support leader to make that really clear that you're investing in the right place, even though the graph is just hockey sticked, you know.

Austin Emser: 28:30

Uh, a hundred percent. I mean, that that resonates so much. And who knows, maybe some of this would be a future conversation topic, perhaps. Um who knows?

Charlotte Ward: 28:42

Who knows?

Austin Emser: 28:42

Like the punchline is we gotta communicate to the business, particularly you talk about maybe you're one of the earlier sport hires. I actually, whether you're the first or the like, you know, hundredth support hire, um, the thing, what's the thing that all business understands? It's dollars. The end of the day, it's dollars. We could complain, oh, you know, our queues X many tickets long, our reps are working X many hours. We could talk about how, you know, all you know, these customers, these three customers are really frustrated. That's nice, but when we start putting that in context of this is how many dollars we spent on this specific issue, this is how many dollars we lost due to churn and that customer lifetime value, looking historically because these tickets took too long because we didn't have the headcount or the training or X in order to solve that. That's when we really start to get the business to listen.

Charlotte Ward: 29:35

Yeah, I couldn't agree more. I couldn't agree more. I think that's the perfect end point for this conversation before we do hit three or four others that I know are hidden in this conversation. I would love to have you back to unpack some of the things that we just skipped over as like skillful podcasters that we are. Like, nope, that's not this topic. We we we kind of kept coming back. Uh, but there's definitely there's definitely at least three or four other things I would love to unpack from this, and and I know for sure other things we can talk about. Will you come back another time, Austin, and and unpack some of that other material with me?

Austin Emser: 30:12

I would love to.

Charlotte Ward: 30:13

Awesome, cool. Well, I look forward to that. Thank you so much for joining me today. Uh, talk to you soon.

Austin Emser: 30:19

Thank you so much.

Charlotte Ward: 30:23

That's it for today. Go to customersupportleaders.com forward slash two four seven for the show notes, and I'll see you next time.