Cookies

We use cookies to provide you with the best experience and deliver personalized content. By using Stonly, you agree to our Cookie Policy.

Browse other podcasts

How to Handle Unpredictable Spikes in Ticket Volume

Guest

James Baldwin, Director of User Success and Support at Change.org

Summary

James Baldwin, Director of User Success and Support at Change.org, talks about how he handles unpredictable spikes in ticket volume. James has a ton of experience with this topic because Change.org’s ticket volume is largely tied to world events, which means he can’t always predict when spikes will happen, how big they’ll be, or how long they’ll last.

In this episode, James shares how he plans ahead for these spikes, what’s included in the plan, and how he keeps his team motivated and energized when ticket volume skyrockets.

See the episode’s transcript

Meredith Metsker: Hey everyone. Welcome back to another episode of Beyond the Queue. Today, I am very excited to welcome James Baldwin. He is the director of user success and support at Change.org. James, thank you so much for taking the time to talk with me today.

James Baldwin: Of course. Thank you so much for having me. I'm excited.

Meredith Metsker: Yeah. Awesome. Me too. So our topic today is really cool. Basically we're going to talk about dealing with unpredictable spikes in ticket volume, which it sounds like you have some unique experience in at Change.org. So can you just give me an overview of what ticket volume generally looks like at Change.org? Sounds like it's definitely different than some other companies and organizations.

James Baldwin: Yeah. So averaged out over the course of a year I'd say we probably see anywhere between 1,800, 2,000 tickets a week coming in. But like you alluded to, we don't really see seasonality like you would say, for example, if you're a retailer and you come through like Christmas season or things like that, but we do have some levels of seasonality, like we will see drops in certain times of the year that we can predict. But we do have semi frequent spikes in our tickets based on what's happening around the world that's really difficult to predict and also really difficult to understand the size of the spike, how long it's going to last, things like that.

Meredith Metsker: Okay. Yeah. So you mentioned that some of these unpredictable spikes are based on world events rather than seasonality. So can you give me some examples?

James Baldwin: I can't think of a better example than COVID.

Meredith Metsker: Oh yeah.

James Baldwin: This craziness that we've lived through for the last, what? Two years now, almost at this stage. It was really interesting for us because when COVID first started hitting the world, we looked at it and we were trying to figure out, like I think many of us around the world were, what is this? How big is it? How long is it going to last? And naturally as the virus became more and more widespread and deadly, we started seeing more traffic on our website, more petitions coming through and things like that. If you recall, at the same time this was around the time that George Floyd was murdered and a lot of the Black Lives Matter movement started kicking into high gear and there's a lot of racial justice campaigns going on around the world. So simultaneously, we actually had both with those spikes coming together, verging on our platform at once. And actually right now, I think the George Floyd petition is the largest petition on our platform's history with just shy of 20 million signatures. So the end result of that is we saw an overnight spike of about five times our usual weekly volumes.

Yeah, it was really challenging because, like I said before, we took a look at this and we were like, "Okay, well, I don't want to do a knee-jerk reaction to this spike because maybe this 5x is only going to last a day. Maybe it'll last a week, maybe it'll last three months or a year. I have no idea. Maybe it's going to get bigger. Maybe it's just going to subside." There was so many questions, there was so much stuff coming at us at once that it was really difficult and in a way it kind of paralyzed us from being able to make prior... Or sorry, proper decision-making with that.

I can give you one other example that's not as severe, so you can kind of get a better sense of like outside of COVID, what does this usually look like. We'll see spikes around anything kind of happening in the world that's of importance. So this can be stuff like presidential elections, not just in the US, like the latest election in France had a lot of activity on our platform. Global events, like the protest that happened in Hong Kong recently, or even things like Kobe Bryant's death, which actually had a petition asking for Kobe Bryant to be the new logo of the NBA and that had over three million signatures on it. So it really can just to be anything that's happening in the world.

Meredith Metsker: Okay. Wow. So how do you and your team handle these spikes when they happen? I mean, obviously you can't predict exactly when they'll happen, but you know that they're going to happen at some point. So yeah, how do you handle the spikes? Can you kind of walk me through your process?

James Baldwin: So admittedly, it's been a very iterative process to try to figure out how to appropriately respond to these spikes and these increases in traffic, but what we've kind of learned over time and I'll get more into this here in a second is we can't just look at, "Oh, our incoming volume is increasing by X." Because we also have to look at our staffing levels. If our volumes only go up, say by 10% in any given day, that's fine. But if we have a few people out on vacation, maybe we have some attrition, maybe people are sick or lots of training or something like that, well, that 10% might feel like 80%, right? So we've had to learn that we have to look at both sides of the spectrum here when we understand how to properly address the demands on our team, if you will.

But initially and historically, we've actually run our team at a slightly larger headcount than our team would be on the week to week basis to help account for those sort of factors. And the end benefit that we see with this is that when spikes are coming in we're able to deal with the immediate impacts more fluidly and really not burnout our staff, which is a huge other factor as well. You don't just want to just start saying, "Okay, everyone's going to work 80 hours a week." Or something ridiculous, right?

Meredith Metsker: Right.

James Baldwin: That's not a good working atmosphere. And the benefits of overstaffing the team is that when we're not in a spike and we're not trying to play catch-up, it gives the team a more relaxed atmosphere and allows them to expand their skills into other projects that we have available for us at any given time. So people can start working through different types of career path and training, things like that. So there's a lot of benefits to going through things that way and overstaffing your team. In hindsight though, it's really not the best solution. It feels more like a bandaid solution, admittedly. And there's kind of a few reasons why. So first if we have reduced staff capacity on vacation, attrition and things like that, like I mentioned, we're kind of back at square one anyways, unless we go crazily overstaffed, which we won't do.

And honestly, even if we do overstaff our team by say 10 or 20% with relation to the size of the spikes we get, it's almost not adequate to handle those anyways. So like I said, you would have to overstaff the team by probably a hundred percent to be able to say, "Okay, well we can handle any spike, no problem." And while skill development is always great for your team running your team well over capacity is really not a great business practice overall and there's probably better ways to use your money.

So taking a quick step back, actually, because I realized maybe I didn't answer your question fully. So when we do see a spike coming into our queue, there's a few things that we'll do. So first is we'll take a look at the numbers that we have at the end of the first day to kind of get a sense of the queue and the team health. Usually we're not changing much in day one. Second day, we'll keep a closer eye on things and if the tickets continue to surpass the team's capacity to handle them, we'll start engaging a workforce management team to understand various scenarios, including the nature of the situation or team's upcoming availability and start doing a few light models around the growth of the spike. Now, in most cases, our simplest measure is just to open overtime to the team, usually at a capped amount first and then eventually unlimited if the spike ends up large enough.

And we'll also start scaling back or fully eliminating all non-essential meetings. And in some cases we'll go into a full all hands on deck, bring management to the cues, things like that. And these are kind of last ditch efforts. But really we only start looking at these when our SLAs start getting beyond any sort of reasonable expectation. And in those cases, we'll put notices on our help center saying, "Hey, we have higher volumes." We'll start updating our responder emails, things like that. But it is very rare that we'll actually hire staff as a result of spike because again, they usually are short-lived and then by short lived I'm talking a few weeks to maybe a month at most, in most cases.

But like I said, to me this approach isn't ideal. There's a lot of things that can go wrong with that sort of approach. You could either over overestimate or underestimate the impact of those steps will have on your team's capacity to do work. So for example, if we saw a large spike and we went in the full all hands on deck and we stopped project work, we stopped training, we stopped one-on-ones or whatever we did, and the spike went away in a day, then that would seriously disrupt the flow of things and it just wouldn't be appropriate. So what we're doing right now, as we speak is we're developing a new document called the business contingency plan. And the idea behind this plan is to create a lot of automation and clear process guidelines that anybody in the organization can execute.

So the idea is that we will say, okay, well, the system will automatically look at hourly, daily, weekly ticket trends and contrast that against our workforce management system, which will look at staffing schedules, vacations, things like that, and come to a conclusion as to, okay, well, how well are we able to adhere to our SLA goals? Because ultimately that's what we use to kind of guide us in our decision making. And if the system says, "Hey, there's no spike, but for some reason, we're at half capacity in our team." Then we'll start kicking into gear and going through a series of escalation steps that we will measure over time to say, "Okay, is this too severe? Maybe we do this one if the SLA goals are X, Y, or whatever it may be."

Because the goal is to, like I said, appropriately respond to the situation and not go overkill, but also not have people waiting longer than they probably should be. So yeah, it's ambitious, but we're really confident and we're very close to being able to get to a stage where we can deploy this out and start iterating on it and I really can't wait to see the results of this, because this really should alleviate a lot of our pain. Like you said earlier, we know spikes are going to happen. Stuff's going to keep happening around the world for the better and for the worst. We don't know when it is, but if we have a plan in place to say, "Okay, here's how we'll respond to the impacts of our team as a result of these things." Then we'll be in a much better place to be able to do this consistently and not also have to worry about a single point of failure at any stage the process.

Meredith Metsker: Right. Yeah. That's awesome. I love that it's kind of a mixture of your human help, your agents on your team, but also deploying some automation to help make things a little bit easier there.

James Baldwin: It's honestly a bigger challenge than I thought it would be from the automation standpoint, because there's honestly not any tools that exist today that do this. So, hey, people listening, I know you're creating all sorts of AI tools. Here's an idea for you. So there was a lot more work internally for us to kind of set the sort of thing up and again, I think this would be a part of the way that we iterate is understanding how do we improve on the elevations and the flags, so things like that, that we have in place. And the rest of it, we're pretty confident being able to say, "Okay, well, it's fairly easy to tweak more manual side of things."

Meredith Metsker: Okay. Would you mind just to kind of clarify for me again, what all is going into this business contingency plan?

James Baldwin: So I'm breaking it up in kind of two halves, if you will. So the automation and the manual side of things. So the automation side of things will, like I said, take a look at our staffing levels, our upcoming vacation, our team size training, the age of the hire, as in how long are they 10 years with us, as well as looking at incoming volumes and looking at those two pieces of data and saying, "Okay, well, do these things match, does the capacity to handle work match the incoming work? And once we start seeing a spike, it should immediately start saying, "Okay, well, the capacity doesn't match the actual ability of our team to handle this thing in our SLA and so therefore we need to alert staff that we need to start implementing the manual side of things."

And the manual side of things will then say, "Okay, well, we're going to go into like a stage one of this plan." And stage one might be something like, we're going to immediately open over time. We're going to immediately start looking at light touch things that we can do that will vastly interfere with our day-to-day processes and handle this. But if the system, the automation system says, no, no, no, that's not enough. This spike is getting bigger and bigger and bigger, more stuff is starting to come in, then it'll automatically kick into like our stage two or stage three, which will gradually become more impactful of not only our processes, but also impacting the volume of work coming in. So this is when we start looking at all hands on deck, this is might when start suspending certain processes to be more efficient in handling large volumes of work at once, this is when we might suspend QA. Like the ideas to become more gradual, more severe as the impacts of the spikes become more and more severe as well.

Meredith Metsker: Okay.

James Baldwin: Does that make sense?

Meredith Metsker: Yes. Thank you for clarifying. That's awesome. So you mentioned that there aren't many AI tools for the more automated side of this plan. So if you're able to say, how are you doing that? Like if other support leaders wanted to do something similar, where should they start?

James Baldwin: I don't have a good answer for that, because like I said, not only is this still a kind of a first iteration, but we're building it all in-house and we're doing it in a way that we know are going to have to really improve because the system integrations are not natural with us. We outsource most of our support work, so we have a third-party vendor and their workforce management team is separate than our internal tools. So we're trying to find ways of linking this together. What I would say to any support leaders out there that are interested in trying a similar sort of solution is to do this more manually upfront and run regular reporting. It doesn't have to be hourly or anything crazy like that, but if you maybe run it once or twice a day, take a look at your volumes.

It should be pretty easy to create dashboards that look at day over day trends, week over week trends, things like that, and look at your staffing capacity and understand, "Okay, well, how much can we handle in a given day?" You should be able to say, "Okay, we'll have a pretty good sense of what we're going to need to do, and whether we need to actually take any action on these spikes or not."

Meredith Metsker: Okay. Gotcha. That makes sense. I know you just kind of started this, building this plan out, but in the past can you just walk me through like some real-world stories of when an unpredictable spike happened and then step-by-step how you dealt with it?

James Baldwin: Sure. COVID is probably the best example because of the size, but I don't think it's the most consistent example, so let's go back to the petition I mentioned earlier with the Kobe Bryant's death. Whenever we see large news items, obviously we all take attention just because of the nature of the situation, but then we start thinking, "Okay, well, is this going to end up on our platform? And if so, it's going to end up hitting our queue as well." And sure enough, very quickly, we started seeing a bunch of petitions about Kobe Bryant both in kind of memorial and all sorts of things, both serious and not serious. But as the petition I mentioned picked up where people were asking for Kobe Bryant to be the new logo of the NBA, and it started gaining steam we started seeing more traction coming and saying, "Okay, well, when will you start taking this more seriously? This is likely going to be a pretty severe spike in our volumes."

And so we immediately started taking a look at, "Okay, how quickly is this actually coming out? And are there any things that our marketing teams or our campaign teams are doing to help with this?" So digging one step back for those that don't know we have campaigners on our team that actually help petition starters with their petitions free of charge and we say, "Hey, here's how you can reach out to the media. Here's how you can optimize your condition. Here's all these things that will help you amplify your petition, be more successful." Things like that. So while we have thousands of petitions on our platform, we can't do it for every one. So we had to check, "Okay, is this petition being supported and sure not the fuss?" That's an indication to us that this is likely going to grow much faster than it might have otherwise.

Once we understood that we started looking at saying, "Okay, do we have anything in training? Do we have any upcoming vacations? Is there anything that we can do in the short term like I said a few weeks to about a month that we can use as a mitigation staff to help make sure that we're at the maximum capacity that we can?" In this case we actually had, I think, two people with planned vacation over the next month. By no means, were we going to force them to change it, but we said, "Hey, look, this thing is happening. Are you able to move this? Are you able to change this around?" And luckily enough for us, they were actually more than happy to change it to kind of about a month out. Worked out for everybody in that sense. Within about, I would say about a week or so we started getting a lift of about 60%, if my memory serves me right. 60% increase in volume around this particular petition.

And that is about average for most of the spikes that we see, but what made this one a little bit more challenging is that it was longer lasting. It lasted for about two or three weeks at that sort of volume.

Meredith Metsker: Wow. Okay.

James Baldwin: So as a result of that, like I said, we were kind of in our older process. So we continued to make sure that any meetings that weren't absent necessarily weren't happening, managements are jumping in the queue, taking a look at how we can help. We started helping people work through and bulk updating... Not bulk updating, but bulk handling similar type tickets whereas normally we would kind of work first in, first out. So we would work through various kind of strategies as we continue to see the health of the queue change and as things started returning back to normal and becoming a little bit more healthy, then we started saying, "Okay, we'll start capping over time again, we'll start having more regular meetings again, we'll start getting management out of the queue and things like that." So it was... Yeah, I think that's probably a pretty solid example of how that sort of spike has happened or has been dealt with historically internally.

Meredith Metsker: Okay, perfect. Thank you. So we've touched on this a little bit, but I kind of want to dig a little bit deeper since you can't predict exactly when the spikes will happen, but you know they'll happen and maybe sometimes you see a story on the news and you're thinking, "Okay, this might soon show up in our queue." How do you, as the support leader there, how do you prepare your team, prepare your processes? Like how do you prepare to handle that spike as efficiently as possible?

James Baldwin: Yeah. I kind of like I touched on earlier. Honestly, we didn't in the past. We had this kind of knee-jerk instinct like, "Oh, this again, okay. Here..." We know kind of inherently what to do, but besides from that, we really didn't have any way to prepare for these things. And again, that's why that plan that I mentioned earlier, business contingency plan is going to be in place to guide the team as to, "Okay, here's how we look at this. Here's how the automations are going to look for signs that we're going to start need to ready for various things overtime, et cetera, et cetera, et cetera." And so that will be kind of our main way of dealing with this moving forward.

The other element of this as well as communication. Communication is so important, not only within our team, but also within the rest of the company. We're not the only ones who see when something happens in the real world and we're definitely not the only ones that see on a petition start going viral on our platform. Our product team, our campaigns team, everybody in the company looks at these sorts of campaigns as they're happening and understands, "Okay, well, here's the things that we can do to best support them. Here's the things that we can do to help make sure that they're successful and make sure our users are supported at the same time." So opening up more channels of communication with those teams to make sure that we're all working together and saying, "Okay, well, what are you guys doing? How is this going to impact us? Here's the things that we're seeing from support." And kind of get things more oriented around that.

When there's large groups of petitions around a singular topic, what we'll sometimes do internally is create kind of tiger teams, which will have representatives from various people around the company, working together regularly to, again, share that data, maybe more frequently, talking about strategies, talking about kind of everything that we want to do as a company to work with and help this petition become successful. So making sure that those lines of communication as a summary is open and that we're actively building those relationships, those internally is so, so critical to making sure that we can prepare adequately for it.

Meredith Metsker: Okay. Speaking more internally with your team, how do you as a support leader keep your team engaged, keep them motivated when maybe it gets to an all hands on deck situation?

James Baldwin: Yeah, it's challenging, especially since what I mentioned earlier is our team is outsourced. So it's a little bit different now with everybody being fully remote, but essentially what our leaders do is making sure that there's a lot of visibility from within the team and transparency in the team around what we're doing and why we're doing it. And so we actually would find that there's a morale boost once the managers and supervisors jumping into the queue and the team can start seeing us all working together and saying, "Okay, let's figure out how to handle this together." We see a nice morale boost. And usually once we come out the other side of that, we'll kind of take this breather. We'll step back. We'll make sure that any sort of culture initiatives that we had planned are kind of put in the full throttle. We'll make sure that...

When we were in office, we might do like an ice cream day or something like that and just kind of help the team just not think about work as intently as maybe they had been before and making sure, again, looking at signs of burnout and making sure that the team feels healthy during these situations. In a past life, I used to work in the video game industry and something that is historically damaging in the industry and still happens today is what they call "crunch." So historically what happens in the video game industry as they're developing games, is they go through a period of crunch. And crunch, without much exaggeration, is sometimes people sleeping at work and doing extreme hours for extremely long periods of time to try to meet production deadlines. And I've seen the impact that crunch has in the video game industry on staff and on development and there's a reason why so many people burnout of that industry and leave because it's unhealthy.

And the last thing that I want to do is to create a similar sort of atmosphere when we have large volume spikes. I would much rather have us kill all of our SLA, not be able to meet any targets and just go to my management team and say, "Look, I can't burn my guys out. I can't have them working 80 hours a week. We're just going to have to work through this the best that we can for as long as it takes and come out the other side." And we've done that a few times and my management team has always been very supportive with me and fully understood that we can't treat staff that way. We need to make sure that everybody is healthy making sure that... people want to come out on the other side of this still happy with their jobs, they still want to come into work and still value the work that they do.

One of the things that we help instill with everybody who joins our team is understanding that the people in our platform are relatively unique in that there's two groups of people on our platform. There's people who are wanting to create some sort of change they want to see in the world and then there's people that want to support that. But ultimately that same sort of person is on our platform. Somebody is saying, there's something not right. I want to create this change or want to support this change. And that sort of empathy is really required in this role and so we really need to make sure that's in the forefront of... If we treat our team as just a machine that people just turn through and they check in and they check out, we're really not going to be doing our users the right level of service and we're not really going to build the relationships that we want to have with our users and our petition starters, our decision makers and everyone else who's on our platform. So it all comes full circle to making sure that we're really treating our team well.

Meredith Metsker: Yeah. I hadn't thought of that, but that's a really great point about approaching your users with empathy. I mean, if a user's on your platform, it means they're passionate about something or there's a very emotionally charged situation happening and yeah, you can't have burned out support folks trying to help people who are going through something.

James Baldwin: And that's exactly it, it's... Like I said, in some elements it's very unique because you really want to put yourself in the person's shoes who's signing this petition around the Yulin Dog Meat Festival in China, or whatever it is that...Whether they're programs, anti-guns, pro-life, whatever somebody is interested into, they're usually pretty passionate about it if they're willing to come to our platform and put their name next to it and say, "Yes, this is a topic that I personally believe in." And so we really need to make sure that if somebody is then writing to us and saying, "Hey, I need help with this, or I need advice with this or can you help me think through how do I be more successful with my petition?" We really want to put ourselves in their shoes and say, "Okay, well, how can we best help you through this situation? How can we help think through things in the future?"

And again, if we just think things through, as you're just a number, this is just a ticket, this is just whatever else, we're really not going to accomplish the mission that we have as an organization.

Meredith Metsker: That makes sense. So this is a little off topic, but since you outsource some of your support, how do you guarantee or ensure that same level of empathy with your internal part of your team, and then the external part?

James Baldwin: Lots of QA. No, well, that is true, but lots of training and ongoing training. Empathy is one of those weird things with QA that is subjective and so a lot of the times that we do training around empathy, it really is reinforcing that, put yourself in the shoes of this person and their situation, understand where they are on their journey in our platform. Are they somebody who's just getting started with a petition? And understanding their world and then saying, "Okay, well, here's how we can not only answer this question, but also maybe three questions down the road of where you're going from here." Really kind of reinforcing that. And so whenever we're going through quality with our team and we're looking at things we're saying, "Okay, well, here's different ways that we can handle this both better and both worse. Here's how you've done it." And really continue to think through the lens of the user, think through the lens of the person who's on our platform, really think through, like I said, their journey, where they are in their life cycle on our platform.

Like I said, some people are here and saying, "Oh, I saw this on Facebook. I agree." And some people are like, "I'm about to go host a rally outside of Washington, DC." And everybody in between. And so we really need to kind of ensure that we have that level of empathy at all times. The other element to this as well, which is fairly unique is we're an international organization. We're not just based in the US. So we have, I think about 12 or 13 different languages of users who write to us on a weekly basis. And so as a team we need to understand the cultural impact of what's happening around the world. So we can look at something and say, "Oh, well this situation over here is kicking up some news, but it seems okay." But until we actually get the cultural context behind it, then we can say, "Oh, that's actually really big. That's really important and we're likely going to see a lot of impact from this." And also that's going to engage a level of empathy that we want to use when people are writing in about these situations.

And sometimes those cultural impacts are from people within the team who might know that area or have been there, or we have campaign staff around the world. So we might reach out to our campaigns team in Turkey or Japan or Russia and be like, "Hey, can you help me really understand what is happening in this petition? Why is this really important? Why is this taking off?" And that extra level of information really helps us do our job even more. So again, it also circles back to that kind of communication thing, right?

Meredith Metsker: Right. Wow. Yeah. I hadn't considered that either. So I imagine this all becomes even more relevant when there's an unpredictable spike. Like not only do you have the nuts and bolts of handling the spike and making sure tickets are resolved, but you've got this whole other kind of emotional side of the issue that you and your team have to deal with as well. Yeah, that's crazy.

James Baldwin: Yeah. And the other thing I will say is whether... There's a lot of stuff on our platform that you will or will not agree with. And that's just the natural way of our platform. We don't want to be solely focused on left-leaning or right-leaning or anything like that. We're basically saying, "Look, as long as it's not violation of our guidelines, put a petition on our platform that you personally believe in, and people who agree with you we'll support you." And so our job is to make sure that we stay neutral, understanding, again, the person, their situation, their journey, and putting ourselves in their shoes and helping them accomplish their goals, whether or not we even personally agree or disagree with what's going on.

Meredith Metsker: Okay. So I imagine there in order to be more prepared for when unpredictable spikes happen, imagine you kind of really have to establish the quality of care you're giving customers, the amount of empathy, cultural understanding, and just in general, the way you approach these issues. I imagine you kind of have to have that very well laid out ahead of time so that when spikes do happen, everyone's just like, "Okay, we know how to do this. We're just doing more of it right now."

James Baldwin: Yeah, yeah. In theory, yes. That would be great. As I said admittedly, we haven't done this the best, but we're learning. The more that we go through this, the more that we understand the value of opening communication lines, the more that we start understanding the effectiveness of some things or other things. We learn more and more and more, and we're just getting better and more efficient as we go. It may not make sense that we have half of our team going and reaching out to somebody in Russia to understand cultural impacts of rotation. Maybe it makes sense that supervisor do that or maybe, like I said, we're continually getting better, but you're right, in an ideal world, everything would run that way.

And we're hopeful that as we continue to run through this plan that we're building, as we continue to get better at these communication lines internally as we continue to build these relationships will be that these things will become more fluid, they will become less chaotic and we'll be able to hit not only our internal goals, but also help our users with what they need in a reasonable amount of time.

Meredith Metsker: Gotcha. Kind of going back to that, this business contingency plan that you're building, I'm curious what the timeline has been on that. When did you start working on that? When do you expect it to be ready to implement? Just to give our audience an idea of how long and how much work something like this takes.

James Baldwin: So we started thinking about it at the tail end of our COVID spike, but we didn't really start putting pen to paper until just before the holiday season last year. Now, for those watching, you probably don't know, but I also went on parental leave in April and then just about to come back to work now. So we did pause it for quite a while as we worked on it. But I was something that we didn't want to rush through. It's not something that we just wanted to get out of the door. Like I said, the process that we had in place were fine. It did good job. We just knew that they weren't as efficient or as efficient as they probably could be. So we really want to think through like, are there tools out there that can help us with this? What do we have to do internally? How can we make sure that we're not just putting something to paper that's essentially what we're already doing? How do we make sure that we're actually effective?

So we intentionally took probably more time than many might at that sort of plan to really make sure that we try to get it right on the first try because like I said, when these things happen, we have to react quickly and we have to react appropriately. And I don't want to just put a plan out the door for the sake of saying we have a plan that's going to end up actually having our users in a worse place than if we didn't do it at all.

Meredith Metsker: Gotcha. Okay. That makes sense. I'm curious, I noticed that you've been at Change.org for about eight years now. In that time, have you seen any changes with these unpredictable spikes? Are there more of them, less of them, and so on? And then, how have you adapted your strategy for handling them over the course of your tenure there?

James Baldwin: Yes, and yes they're happening more frequently and because our platform is continuing growing in size, they are happening in much larger sizes as well. Eight years ago, seven years ago, around then we might see a few kind of petitions a year that would spike our volumes and now it could be as frequently as once a month, depending on what's happening. I would say about 40% of our volume is from English speaking users, primarily in the US but not definitely not all exclusively. And so a lot of things that are happening in the US especially over these last few years has caused a lot of controversy and as a result is continuing to drive traffic in.

What we've looked at is we can't just throw money or bodies at this. So we've tried to also take a step back on top of everything else that we've kind of spoken about today to understand self-service and what are the most effective and efficient ways that we can utilize self-service to help users when they contact us. Because honestly, a lot of times when somebody is contacting us, they don't actually need a human to resolve their issue. They could either do it themselves or using some sort of like knowledge article they can fix it. They just don't know how to. So we've invested a lot of time and money in our knowledge base, ticket deflection tools, things like that. And this is an area that we have a lot of focus in as we move forward into the end of this year, as well as 2022, to further understand, like how can we utilize ticket deflection techniques, knowledge-base improvements, things like that to really help our users with the problems that they have, and also make sure that those who really do need humans help can actually get that as quickly as possible.

Meredith Metsker: Okay. I'm curious, is some of that creation of that self-serve content, is that some of the projects that you were mentioning earlier, that when they're not in the queue, your team members might be working on outside projects?

James Baldwin: That's definitely one of them. Yeah. Understanding everything from style as well as accuracy of the articles, making sure that the articles that we have in our platform match the type of questions that we're receiving from our users, making sure that they're localized in all of our languages, which historically we honestly haven't done a good enough job with and so that's one of the main things that we're focused on right now. So yeah, that's definitely one of the projects that the team is focused on for sure.

Meredith Metsker: Okay. And as you've gotten started with building out your self-service program, I guess, how did you start? Did you just look at your most frequently asked questions and go from there? What was your process?

James Baldwin: So it's a mixture of looking at the types of tickets that are coming in and making sure that, "Hey, do we have articles on these? Are they accurate? Are they easily discovered on our self-service portal?" But we also look at the language that people search on our self-service to understand if that's matching as well. Like for example, maybe Spanish users are more frequently looking at a particular topic. And so, okay, maybe that's where we want to prioritize our time next. Things like that. So we kind of look at all available types of data that we have to understand our prioritization of work.

Meredith Metsker: Okay. Yeah. That makes sense. Very cool. Man, I am loving this topic because I'm discovering that so much more goes into dealing with unpredictable spikes than I ever could have predicted, pun intended. Like there's the human side of things and even within that side, you have the auto... some things that you're trying to automate, you have the manual response. But then you also have the self-service side and yeah, it seems like this is a very multifaceted issue in order to handle unpredictable spikes well.

James Baldwin: And you could definitely point to any of the things that we've spoken about today and say, "Well, of course, yeah. No, we do that all the time, even without spikes." While a lot of that is true we found that investing heavily into a lot of these areas allows us to better mitigate the effect of that these spikes would have, the negative effects that these spikes would have. Because again, ultimately we want to put ourselves back in the shoes of the somebody who's contacting us, right? If it's someone like, "Oh, I want to sign this petition. It's really important, but I forgot my password." That's not something that we want somebody to wait some ridiculous amount of time, four days for an answer or something like that. We want to say, "Oh, well, here's how you do it yourself." If it's somebody that's more, oh, I can't figure out how to get this petition on the ground, but it's so important. Okay. Well, how do we make sure we get you through to human as quickly as possible to really help you think through the intrinsic things?

And so understanding that, oh, well, we need to make sure that users are able to self-service so that those users can get through the humans faster to make sure that we have the proper staffing in place to make sure that we can do that in an appropriate amount of time, all those sorts of things tied together to make sure that we can do this well. So like I said, it's been a learning journey for me over these last eight years. We haven't done it well at times. We've definitely made a lot of impactful mistakes and we've had some hard lessons from it, but we're at a point now that we're very confident that we're on the right path to help make sure that we can do this while moving forward and that we're focused on the right things moving forward to really make sure that as our platform continues to grow, as the spikes continue to grow in size, that it won't actually impact our day to day operations as severely as it may have been eight years ago.

Meredith Metsker: Gotcha. So kind of on the note of all of the things you've learned and the improvements you've made over the last eight years, what advice would you give to other support leaders about how they can go about handling unpredictable spikes?

James Baldwin: So one thing I understand is what I've spoken about today is relatively unique to me in our team, in our company. Not only the platform of Change.org is relatively unique, but also we only deal with email only tickets. So if you're listening to this or watching this video, maybe you deal with phone or chat or all of the above or social media, like there's multiple different avenues that you might look and say, "Okay, well maybe this isn't hyper relative to me." But I think ultimately device that we have comes back to that notion of try to have a plan for things, try to think through, "Okay, well, if X were to happen, how would we properly want to react to that? What are the things that we could have in place to allow us to do this and ideally as quickly and with as much automation as possible upfront?"

I'm not a big fan of like automating full flows, but automating those differences in your team's capacity versus the incoming work and how that would change in certain situations, what if your team went on strike? What if all of a sudden a new product came out and it killed your phone lines? Whatever it may be what are the tools and systems and things that you would need to have in place to be able to address those? And you'll learn over time as you continue to think through those plans, how well or how poorly those plans actually ended up and you'll learn from that and to continue to iterate and say, "Okay, well maybe we went overkill here or no, this wasn't effective at all."

And I guess the second thing is I'm a really big fan of communication as I mentioned earlier. So really go out of your way to be as open as possible, both internally and externally especially in situations where you can't contain a spike and you're finding yourself a bit under water, make sure that you're continuing... or that you're communicating internally. I know in the past I've hesitated on telling people that, "Hey, we're struggling a bit." Because I want us to look good all the time. There's nothing wrong. It's under control, don't worry about it. But I learned really quickly that once I actually started saying, "Hey, actually this is kind of hurting us a bit." Leaders were more than happy to say, "Hey, how can I help you? What are the things that our team can do to help alleviate this? Are there..." People are really willing to come to you and it's not actually sign of weakness as I initially felt it might be.

And as a result, we've been able to get people like our product team to help us listen to features and all these sorts of things that have really helped us over the years. And externally as well, like I said, the last thing you want to be is... How many times have you called a credit card company or phone company or whatever else and you're just sitting, listening to music for hours on end sometimes? But sometimes I've called people and they said, "Oh, hey, yeah, we are expecting an eight minute wait." Well, great. Setting that expectation up front allows me as the end user to say, "Okay, well I'm willing to wait or I'm not willing to wait at that time. Maybe I'll try again later." But if you just keep people in this indefinite darkness of not knowing how long they're going to wait to speak to somebody, they're going to start feeling ignored and unappreciated. And that's the last thing that you as a company wants your customers to feel.

So even if you can't communicate specifics, like you have wait X amount of time, just saying, "Hey, look, we know that this is going to take longer than usual to get back to you, please be patient with us." And again, use some sort of empathy there and that goes a long way with people.

Meredith Metsker: I love that. So I kind of want to wrap us up a little bit, but... And I have one more question, but before I ask that last question, is there anything else on this topic dealing with unpredictable spikes that you would like to add that we haven't covered yet?

James Baldwin: I think you've done a very thorough job at helping think things through and like I said just here a few minutes ago, I know that our team is relatively unique in some ways, but I hope that those listening to this are able to pick through the kind of core themes that I've been trying to outline and really think through how can that apply to your work and your team and understanding you don't have to be as automated as, you don't have to have this hugely fleshed out plan and it doesn't have to be perfect. But ultimately think through these sorts of scenarios, have a plan, even if it's a rough plan and a rough idea of like, what are the things that you would be able to do in these sorts of situations to help make sure that not only your team doesn't burnout, but also your users, your customers are taking care of it appropriately?

And I think you'll find that you'll be less stressed. Things will be less chaotic that's you'll be able to work with your team much more fluidly than you may have otherwise. And so I think that's just like the ultimately overarching things I would just kind of say.

Meredith Metsker: Okay. Great. So last question, this is the big one. Just in general, what advice do you as a very experienced support leader, what advice do you have for up and coming support leaders?

James Baldwin: That's a good question. So I think the best piece of advice that I've learned over the years of kind of being a support leader is take time for yourself. If possible, take dedicated times, block it on your calendar once a week if possible, maybe once a month where you have no meetings and you can step back and take a look at a higher level of your work, your feeling. Are you stressed? Are you happy? Are you doing your job in the way that you want to or are you finding yourself always in sort of a reflex position? Are you finding yourself stressed that maybe you're not working at the way that you want? And understand those feelings and really try to digest them in a way that you can make sense and that you're not just constantly in a stress or reactive state and think through like, "Okay, well, is this just a temporary thing? Is this a reaction? Is this longer term? What can I do to help my own wellbeing in this sort of role?"

Because our job can be very rewarding, very fulfilling and exciting in many ways. I hope that those listening today really have a chance to experience those feelings on a regular basis. I've seen many people get burned out in this sort of role and most often than not, when I've seen that happen, it's because they are in this constant reactive fight or flight mode. And you don't have to be, take some time to yourself, step back, take it slow, make a plan. And yeah, you'll hopefully be able to be a little bit more, I don't want to say level headed, but you'll be able to react to things in a much more calm and calculated manner then you may have if it's just all reactionary.

Meredith Metsker: Oh. Perfect. I love that. Very well said.

James Baldwin: Yeah. Sorry. I know my topic and my overall theme today is planning and that's not everybody's strength and I know that and I appreciate that. And I guess the other thing that I could say on that topic is like, look, there's lots of tools out there. There's lots of articles, there's books, there's all sorts of things that... Like if you're not a natural planner, if you're not somebody who feels that you can do that well, there's resources out there available to you that will help you with those sorts of things, that will help you think through. And of course there's other people in the community, whether it's support driven, whether it's podcasts like these, there's lots of different places that you can connect with your peers that will help you think through these sorts of challenges and help you get through the other side.

Meredith Metsker: Love it. That was great, James. Yeah, with that, I think we'll go ahead and wrap up. James, thank you so much again for taking the time to talk with me today. I think our audience is going to get a ton of value from this interview. So thank you very much.

James Baldwin: Fantastic. I'm glad to hear it. And again, I appreciate you having me on. If anybody ever wants to kind of connect with me, feel free to. I think my LinkedIn will be attached to this podcast, but again, you can also find me on the Support Driven Slack if you're on there. I look forward to chatting with everybody.