Consensus Risk: How BMC Helix Aligned CS and Sales to Drive Retention

Well, good morning everyone. I want to start this off with a question. How many of you guys have been in a conversation, just generally speaking, with sales, where they're saying, "This account is on track," and you, as a CS leader or a CSM, for one reason or another are thinking, "Hmm, I don't know about that. I don't think so." How many of you guys?

Yeah, there we go. All right, so that's exactly what we're going to be talking about today. How do you get to a consensus on what you need to do about risk and your retention? So round of introductions here.

Again, my name is Trent Vincent. I'm a Customer Success Manager here at BMC Helix. To my left is Joel Skaggs. He's our VP of CS Operations who decided to do something about consensus risk.

Then to his left is our VP of Customer Success, Sarah Othmer, who had to trust his solution enough to walk into a boardroom with it. So Sarah, let's go ahead and start from the top. Before the model, before the cadence, before the joint reviews, what was the actual problem that you were trying to solve? Not the operational problem, but the business problem.

Thanks, Trent. So I want to give a little bit of context as to BMC Helix as a company and some of our history just so that you can relate to the problems that we're facing. I know that we're all kind of facing the same types of problems. We were an on-prem company.

We sell software. We are an on-prem company, transitioned to SaaS, going through that transformation. Many verticals focused on IT service management. And as a software company, of course, we have a really strong sales organization, actually led our pre-sales organization before moving over to customer success.

And we have disparate systems that we all like to work in. We've got sales working in Salesforce. We're tracking opportunities in Salesforce. We've got customer success managers in GainSight.

We've got services in another solution. We've got renewals in Salesforce. But everybody's got their own lens and their own eye and judgment call that they're making on all of the upcoming renewals. And so when you start to think about the spaghetti of all these different insights and all these different perspectives on what's really going on in an account, that's the problem that we're trying to solve.

How do we go bring all of those really rich data sources together and have honest conversations led by data, led by an AI-driven approach, which Joel's a great champion for, and have some meaningful discussions around the true risk in our accounts? And what do we actually go do about it? If we don't agree on what the risk is in the account, we're not going to drive the right types of actions. We're all going to be running, doing different things.

And of course, when you're looking at your top accounts, like everybody else, we've got segmentation. We've got strategic accounts. We've got customers that are going to be representing the bulk of our ARR, where we've got a lot of people that are in there day in and day out working those relationships. But then we have, as we talked about on the keynote, we've got the long tail.

And so how do you go and really have an honest conversation around, are they adopting? Are they actually getting value out of our solution? That's a really tough conversation to have when you don't have humans in the relationship with the customer. So we need to bring all of that together.

And of course, no surprise. This is important because we have the same goal. We've got to retain our customers. We're not going to be able to grow our business if our customers constantly turn.

And we're not going to have credibility as a customer success organization if we can't talk about the key reasons why we see customers leave. And so this has been a really-- it's been a labor of love. It's been a journey, quite honestly, over the last year, bringing together these insights and these conversations. And it's been a fantastic-- I think we've got some fantastic outcomes trying.

Awesome. So Joel, Sarah just described the problem, right? So you were the one that had to build the answer. So what is consensus risk?

In plain English, if someone in this room has never heard about it before, what are they looking at? Yeah, I mean, I don't think it's anything that's super magic. It's basically a common language of what is risk on an account. That's what we were missing.

As Sarah mentioned, when there was a talk about, hey, is this renewal going to happen? We're going to go talk to sales over the NULS team. If there's something related to an adoption or risk issue, what's CS's perspective on this? It ended up being very inefficient for our leaders to go around and try to figure out, who do I talk to when I care about a particular type of risk?

So we needed to bring those things together. So we had three independent inputs that kind of came into this. It's not like this incredible technology solution. But what it is is it's taking things that we have that are useful, that people are working on a daily basis, and bringing it together to a single unified kind of assessment.

And so Salesforce Renewal Team actually had their own assessment of risk. And that was usually within a three to six month time frame that they cared the most about that risk. Then we had what our CSMs, the high touch CSMs, were doing, which was we care about risk no matter when the renewal is going to happen. So we have this lifetime assessment of risk.

And we're always looking in and raising CTAs of different severity levels. So that was another piece. And then we had a machine learning statistical model that did a really good job of looking across metrics that are consistent across all of our companies, whether they had coverage or not. So that gave us a little bit of insight on some of those customers that maybe we don't have much touch, but we needed to have some sort of assessment of where they were.

And we had a model that was statistically relevant, but sometimes wasn't used very often for this. So we brought those things together. I think it ended up making it much simpler for the team to really get their head around. And then we actually started prioritizing things accordingly, which was probably the biggest thing that we did together.

The core design philosophy is whatever gave you the worst signal out of those three is what actually won at the product level. And then we had five risk bands all the way from very high to low. So how did you come up with the risk categories? Why are they meaningful at BMC Helix?

Yeah. So the risk bands are-- we needed to have enough of a differentiation because the different organizations that were putting in assessments of risk had different ways of doing it. Renewals kind of just had something at risk or on track. That's not very much bifurcation between that.

And then we had CSMs where we're using kind of the normal gain site risk severity. And then we had a concept of our machine learning model. So that was kind of the risk bands. But more importantly is how do we talk about the risk, and especially how do we make it simple so that we can talk to the board level, we can talk to the HLT level.

People that actually can solve problems and change decisions, they needed to net that out. So you can see on the slide here, we had all of these cancel reasons for our churn customers that we needed to look at the reasons and basically narrow that down. And how do we do that? Well, first of all, there's great information out there about industry best practices around what are the best reason code groupings that we should be using as a software-- SaaS software company.

So that was definitely one of those things. We also worked with our renewals team to look at the history of all of the commentary associated with all of those churns, and then also kind of balance that up against what the best practices were. And we aligned with these six different reason codes. Four, we would call it controllable, and two, that were called uncontrollable.

And so we'll get into a little more about the debates over that. But that was really important for us to kind of say, hey, if we're talking to our owners, we need to say, hey, this is how it nets out. It's product challenges, or it's customer experience, or another on the list. And it helped us basically really talk about what are the mitigation efforts that are going to be unique to those different problems.

Yeah, and if I think about the mitigation efforts, as you mentioned, Joel, we had to simplify. If you think about a renewals team that's overworked, they're tired, they're hitting all renewals right across the business, strategic down to long tail. And they've got about 30 plus different potential reasons for churn, for a renewal that doesn't go through. Those are not going to be accurate selections.

We were being asked to go and assess that data and to sum up, hey, what do we go do about this bulk of customers that ultimately decide not to renew? What do we go do about it? Well, you want to go make that decision based on data, and you go look at the data, and the data wasn't accurate. And so, yes, the simplification was extremely important.

And then thinking about the fact that we all have limited resources, whether it's time, whether it's our talented teams. So putting those resources in the right places to solve the right problems, it's got to be driven by data that ultimately the executives and the board believe in. So we had to have a trusted story based on data, in fact, that we could stand behind and put in front of the team to say, hey, we need to invest in this particular product area because we see that there's actually data behind. We've got a renewal problem potentially in a particular area.

Or we need to go think about this contractual change in our commercial negotiations, or, or, or. So you've got all these conversations that you can have and drive actual playbooks, a strategic set of prioritized playbooks that everybody can get around. So then the whole company is putting their team members behind a set of playbooks that will actually retain your customers, and you see a turnaround, and you're not just relying on anecdotal information, because that's typically where we were right before this. So Joel, nothing's ever perfect the first time, right?

So what were some of the tricky parts of this? What did you have to adjust as you went? Anybody have any data quality issues in your systems? Anytime you rolled out-- like I said, this was an incredibly technical challenge, but it was a new process, where we're bringing multiple elements together.

Well, kind of interesting things happen, where, like, you see that someone's got a transition CTA or a churn CTA, and it had a medium severity. And so we were picking up medium severity on all CTAs. Like, that's just a medium-- obviously, it's not, right? A churn CTA needs to always be high.

So there was one thing that we had to pick it up, and that was something that made us look kind of foolish, right, that we weren't picking up all churn CTAs, but we were using the severity. So that was one thing. The closed CTA that's unsuccessful is still a problem, right? And so we were picking up only open CTAs and the severity level.

So we needed to actually look at closed CTAs that were unsuccessful to actually influence this metric as well. So that was another thing. We talked about renewals. They were doing at risk or on track.

That was kind of the only two things. And then, turns out, as we talked more to the renewals team, there was even more nuance. Oh, if it's at risk in its best case, that's really risky. But if it's at risk and it's commit, well, there's risky.

But we still think, like, that was another thing we actually put into the logic. And then there was just some system limitations. We were going through a bunch of system changes, and there was a lock on what we could actually change in our system. So we actually ended up having to do it in Snowflake.

We actually did this logic in Snowflake and Tableau and actually pushed it to Gainsight. So not typically what we would have done. But lots of little things that we had to kind of change as we went to make sure that we were picking up things. And we weren't looking silly.

And people like Sarah, they're going to beat us up if we don't come in with something that makes sense, if they see an attorney CTA and it's not even showing up as a very high risk, then that's going to be a problem. The last thing-- this was kind of a funny one. Our senior leadership might often have a very educated opinion about their recent experience with a customer and what they think the risk should be. And so we had to create a separate field called adjusted consensus risk that reflected the will of the leader to make sure that it was actually in there and that we actually, at account level, we assessed a customer as per the executive's opinion.

And so we didn't override all of the judgment that had been put in. We needed to have something separate so we could bump those two up against each other. And now I think we created something called overall risk to basically take the worst of those two. So a few changes that we had to be flexible on, for sure.

And I can tell you from the individual contributor side, and just as one instance, I had this customer with some architectural challenges. There were some things that we had identified where we were like, we might have to get everybody on the same page here. So in utilizing this model, we have sales, we have CS, we have product, all on the same page. Are we going to have to make some adjustments?

Are we going to have to change the customer's SaaS strategy? So just wait to have something like this in place so that everybody is on the same page. They understand the severity of the risk. And then that way we can start being proactive and reactive to the issue at hand and just get in front of it and do what's right for our customers and protect the revenue too.

So Joel, in a model, it's only as good as what you do with it. So what does that operating rhythm actually look like? Could you walk us through what you built? Yeah.

So we kind of had to rally around. It's one thing just to have a number and nobody really cares about whether something's very high, high if you're not actually doing something with it. So we actually start prioritizing our discussions in our meetings. And so we had a weekly ops and CS meeting where it's basically Sarah and I and some of my team making sure that we're aligned on which customers we really need to actually down select so that we can actually have a good conversation about those customers in more detail.

And we're using Consensus Risk to help define that. Then we had a biweekly command center, which is the meeting where people like Trent and his management will come and actually have to talk about the situation with those accounts to Sarah and some of her leaders to make sure that we're actually inspecting the story behind what's being told. And then also, we actually have mitigation plans that are there to actually resolve the issues and not just talk. We don't want to be just weathermen.

We want to actually have some ability to change the outcome. And then kind of the really big one is once a month, we have a meeting with our CEO. He cares a lot about this, which is one of the reasons we were able to rally people around this. And he wanted to actually go, in many respects, count by account.

Give me your red and orange accounts, right? And I want to go through every single one of them and see if I agree with your mitigation strategy. So the taxonomy that we had allows to kind of unify around what are we going to actually prioritize with. And then once we have that priority in alignment, it basically allows us to get quickly to actually solve some of the problems that we're having and we're having problems for our customers.

And Sarah, let's talk about the risk reviews for a second. So what actually happens when you go into a room and sales says, on track, and CS says, I'm worried. Who ends up winning that conversation? So it was actually an interesting journey, as you can imagine.

Everybody's got their opinions, and especially about the top most important renewals and accounts that are coming in. We have a concept of up for renewals, or a UFR spreadsheet basically shows us, hey, these are the top ARR customers that you have, regardless of renewal expiration. And we go through those, as you mentioned, Joel, based on severity of risk and based on what mitigation activities we've been taking. And Trent, to your point, oftentimes, especially at the beginning when we were still setting up the operational cadence, there's a perception of-- I think maybe Kim, on stage, was talking about, well, Gainsight's got all these insights that CS is withholding.

And Salesforce has all these insights that's really the source of truth. And so what we were able to do is say, hey, we've all got these different insights, and we probably are telling a more holistic story. So ultimately, everybody wins. You have some friction-filled conversations.

You get through it as you get the operating rhythm moving. But then you start to build trust, because you start to see, oh, wait, this person has an insight because they talked to this particular person in the account. This person is looking at our usage data and can talk about, hey, they're not adopting exactly the way that we would advocate for in a best practices sort of way. So we need to guide them differently.

And the account rep might be talking about, hey, I just had an event. Everything's great. We went out for drinks. It was fantastic.

So you bring all of these stories together, and you actually have a more robust discussion around, is the account plan that you set out at the outset of the year actually still accurate? Are you still going after the right stakeholders in the customer, in the account? And are we really driving the right sorts of actions? And so ultimately, we're trying to drive value for our customers.

This is all great. It's internal conversations. We're pointing out where there's risk. But ultimately, we want to make sure our customers are successful, because if they're not, we're not.

So ultimately, everybody wins. The entire internal team, it drives insights into our product strategy. Our engineering teams are now joining these calls to say, hey, we're starting to see trends. In addition to NPS data, we see all these trends of insights and conversations that you're bringing to this forum that are driving our top risks.

Let's go do something about that, and we can avoid systemic churn reasons in the future. So the customer ultimately wins, because we guide them in a better way, but we win as well. And we needed an arbiter, and so that arbiter ended up being this concept, this model that we had created. And I think there was a little bit of an aha moment for some of the sales folks as something that they thought was on track, and we had rated, or at least the model had rated, as more negative.

And it ended up churning. So I think you ended up getting believers like, hey, maybe I should actually buy into this a little bit. And honestly, the level of intensity from our leadership to basically care a lot about this topic together, allowed us to basically kind of bond a little bit on, hey, we needed to actually have a really good story when we have this conversation about these accounts. And I think probably more than ever, that this allowed, this wasn't just consensus risk, but that pressure allowed us to kind of create some a common effort to be able to have good discussions about those questions.

And then consensus risk was something that we could align on and not have a bunch of different assessments of what risks would be. So I think it really had an impact of actually making sales and CS come together as a result. Yeah, and I could say like from an IC perspective, like I think that I was a little bit worried that something like this might override judgment, but what it actually ended up doing is giving my instincts a seat at the table. Like we could confidently say, I think that there's a huge risk here and here's why.

So, and Sarah, while Joel and his team were building cadences and fighting over spreadsheets, what did you need to see from a leadership perspective before you were willing to put this model like in front of a board? Yeah, as a customer success leader, we've got to be able to go talk credibly about the health of our base. And so I think previous years, there was a lot of anecdotal conversation around, oh, well I've talked to this customer and this is one of our top customers and they're saying this. And that was driving the strategic discussion around, oh, we think we're gonna land here or here on on-time renewals or on GRR.

And so to be able to say, hey, let's go pivot, right? We all know we need to do better. We've got a ton of rich data insights from support, from customer success managers, from sales, right? Let's bring all that together and have data-driven conversations and actually project where we think we're gonna land, right?

In terms of GRR on-time renewals, we needed to trust the data. We certainly had a lot of kind of back and forth, as you'd mentioned, Joel, right around, do we need to tweak this or that, right? And how the algorithm is looking at these different things by vertical, by product, right? By on-prem versus SaaS, by how many times have they engaged with our community or gone to an event?

Are they an advocate for us, right? So no surprise, all of these kind of normal customer success health score metrics. And so having trust in that data was extremely important because I'm gonna go put myself out there, right, as a CS leader and say, hey, these are our top reasons for risk and this is why. Here's the data behind it.

And so if we didn't trust the data and we didn't actually go through the friction-filled conversations, right, to hash it out and make sure that everything was accurately reflecting the true situation, I had a lost credibility right away. We're still on the journey. We're still learning and we're still tweaking and working together as an organization. But to your point, Joel, it's given us a thing to go and rally around, right, in terms of, hey, this is a model that we know our CEO cares about, we know our executives care about, and we need to make sure that it's robust, so let's keep learning from each other.

We're not asking anybody to change the way that they're working in the tools that they're leveraging every day, but we're getting smarter about how do we go work together, right, and get to a shared outcome. And Joel, you know, I said at the beginning of this, having that empirical data and having a reliable way to identify that risk is huge. So for the skeptics in the room, what does that actually look like? What's the math behind it?

And what does the model produce? Yeah, so having been in finance in my prior life, I do understand kind of what the story that they're trying to build and they're building a financial plan, and what this model ended up doing is it really helped bring up a bottom-up perspective, our assessment on individual renewals that were coming in the year, and give them a different viewpoint of what they were doing as more of a top-down model for renewal rates in particular. And so what this model did is we had kind of a two-vector, so we had products. You guys probably have products in your organization that are more or less risky inherently, right, because you're investing more or less in them, and so some of them are just gonna have more risks because we're not spending as much there and we kind of expect in some of these cases for those customers to leave.

And then we're investing a lot in some of our modern stuff, and so we expect higher renewal rates, and so we have history of that, and that was one of the vectors, right? And the other vector was when you're looking at our assessments, so consensus risk was the other vector. So you had this two combination. You got an inherent risk of the product stack, and then you got individual intel on each of these customers, and so we were able to take a probability matrix for each one of those upper renewal amounts for each quarter and be able to come up with a kind of a math-driven, bottoms-up renewal forecast, which was really kind of helpful for our finance team who was super optimistic relative to kind of our initial view, and this allowed us to kind of work together to say, "Hey, are you being super optimistic?

"Am I being pessimistic?" And we just did our second round of that this year, and it was even more helpful to that entire process. So that, I think, is where the math comes in. I think the other thing, how many of you have some sort of machine learning model or kind of health score that you guys are using that's all based on old-school AI, right? The biggest problem, you guys probably know this, is explaining those things, right?

You can get something that the science guy said, "Yep, the R-square is really good on that, "and these metrics matter, "and they actually are predictive." But actually explaining what you do as a result is really hard, and so by bringing this as one of the three ways of explaining consensus risk, it allowed us to bring it in and basically use it without having to put so much spotlight on it, because as soon as you put that spotlight on that machine learning model, it's hard, and people like Sarah was like, "What do you want me to do about that?" It's telling me it's negative, but I can't really tell what to do next. And so that's one of the things I liked about this consensus risk, it has allowed us to use a metric that we hadn't really been used much to actually give really good insights and allow it to be an extra buffer to what was already some good assessments being done for our renewal reps and our CSMs. So Sarah, last question before we get to lessons learned. So now that this is live and running, what is this actually unlocked for the rest of the business that wasn't possible before?

Having a consistent way of talking about the risk in our base, I talked about this before, right? We've got R&D, we've got product management, sales, customer success, all rallying around the data and looking at, okay, we do see these trends, and we believe that the data's telling us where we need to go focus. So driving that kind of top down alignment on where do we actually spend our dollars and spend our resources in terms of product roadmap investment, right? Or offerings that might be different in professional services and customer success, right?

Those are conversations that we're able to have because we actually have a robust set of data to go talk about it. It doesn't have to be, hey, we're doing this because these are the sales plays that we've launched for the fiscal year, right? Or these are the things that our top customers at our customer advisory board are telling us, so we just need to go do this. We're able to go really have an informed conversation.

And this certainly puts us in a position to get, we're not completely out of it, but to get more so away from the completely reactive response to risk and your renewal to a more proactive conversation, right? So now we have these early signs and triggers, right, that are saying, hey, red flag over here, right? You've got a CTA and Gainsight, but you also have this data that's telling you to go run a specific play that allows us to intervene in the renewal risk much earlier. So we're getting ourselves out of this kind of, not completely, firefighting mode.

There will always be those surprise situations that pop up, right? And so in quarter, you know, in quarter we have a surprise situation pop up, you have a stakeholder change, right? You've got something that's really out, macroeconomic factors, right, that impacts a customer. We can have those discussions around, okay, what are the commercial negotiations we need to go drive with that particular customer?

But the systemic, are they actually adopting our products? Are they actually getting value out of our products, right? We can go drive those types of conversations in a much better way. All right, so we've told you about what we built and why it worked.

But now let's talk about what we might've gotten wrong, because I think that that's also very important, right? So Joel, if you had to do this all over again, what would you have done differently in the first 90 days? Yeah, Sarah mentioned that we were not born in the cloud, so we were not a SaaS company to begin with. And so some of these concepts of talking about renewals and talking about bad news, you know, sales guys, typically and gals, don't like talking about bad news as much.

And so that was a bit of a change. And so I think I probably underestimated how much, if they felt like they were under attack a little bit in some of these meetings where they weren't giving a lot of, we weren't connecting a lot on, hey, this is going to be a considered commercial contractual for this big renewal. And so they weren't ready for the conversation at that level of intensity. So the CEO asking, hey, what are you guys gonna do about these commercial contractual problems?

It seems to be like 30% of our problem, right? So what we ended up doing, and we probably should have, it would have been great if we'd have done this to start, is basically have those conversations and kind of argue it out, you know, those controllable factors I discussed, product challenges, contractual commercial, customer experience, those sorts of things are going to have a finger pointing sort of potential. And so having those conversations earlier, more often, before we got to kind of a high stakes scenario is definitely something that I would have done differently. I think one of the things I think is really gonna help us is we're actually gonna pilot staircase here pretty soon.

We think that that's just gonna be, I think it's gonna be a huge game changer because that's gonna be something that will tie in some of that unstructured data sets that we don't currently take advantage of. And I think it's kind of table stakes now. If you're not using some sort of tool like Staircase to actually look at those unstructured data sets, whether it's help tickets or transcripts from calls or the email traffic, we think that's going to be another instances where it's gonna help us get more insight to sales and sales is going to be able to use those insights and we can collectively align on what the real risk is and or unfortunately a churn reason might be. And Sarah, what surprised you the most when this was live?

Did the data reveal anything that was really surprising to you? Yeah, so I know Joel talked earlier about the different levels of risk, right? So we go from monitor to low to medium to high to very high and they're all color coded. And so when you start to look at the heat map, right, of our customer base, I honestly was surprised by the amount of red that we had in our customer base the first time that we were looking at the data.

And when it comes down to maybe the strategic accounts where you have a lot of people that are handholding those customers, right? And you see all of this red and that customer is you're kind of like, wait a second, we've been talking about this. We have all these account calls about this particular customer. Why is there so much risk here, right?

And so it took us some vetting back and forth to say, okay, well this risk is because of a particular project that's going on versus maybe consensus risk at an overall level. So we had to educate the different account team members, right, to say, okay, if you're saying consensus risk is very high, it's because we expect that customer to churn or significantly downsell. The other piece that was a little bit surprising to me quite honestly was in our onboarding phase. So we have an onboarding team that lives in our customer support organization.

And that team was put in place because we really needed to ensure that we were handholding our SaaS customers as they were going through environment activation and professional services implementation, whether it's partner driven, internal, self-led. And that team has been kind of shepherding customers through that process for the last couple of years, right? We've had that in place. And what we started to find was that there was actually risk in the onboarding phase that we weren't raising up as a trigger to talk about why do we have a customer that's paused for six months?

They purchased, they activated, the environment's sitting there and they're not doing anything. And we haven't really had a conversation with the account team to understand the why behind it. So that deep dive into our onboarding team actually helped us parse out, oh, well, there's a good reason. They're going live with one thing and then this is next, or this is a bad reason.

They were sold something and that was kind of an afterthought. They didn't really have an ROI or a business case that they were really trying to pursue. So now we need to go resell it in the account. So those are very different actions that you would go drive.

But if we hadn't dug into the data, we wouldn't have, we would have just had churn, right? Happened later. Yeah, that makes sense. All right, so it's your turn now.

We have a couple minutes left and just want to make sure that we get to discussing the questions that are the most important to you. Any questions? There's some right here. Go live as well.

Real quickly, guys, if you go on Slido, you should be able to pull this room up and drop your questions in there. They'll pop up on the screen. But while you're doing that, if anyone does have a question that you want to just blurt out, I can walk over to the mic. Yeah, we can take one of these.

Did you run into any issues with adoption of these CTAs? If so, how did you ensure that users were taking the appropriate action when working and closing the CTAs? I think there was some education that we ended up doing quite a bit with, I think you probably were part of some of those education. This is what risk categories in particular.

I think we did, right? So that six, we wanted to basically not just be a churn set of reasons, we wanted to be a pipeline of the risk to be the same as the churn reason so that we had mitigation plays and that if it ever got, unfortunately, to a churn situation, there'd be no doubt that obviously this was the churn reason. So there was a lot of discussion about that. And then I think risk severity, there was a little bit of nuance you were just discussing.

Like, is this really a churn risk or is this really just kind of like a relational risk and what severity level should that be? Because if you put it at high, it's going to immediately go into this bucket of, and it's going to get a lot of focus and do you want that level of focus? So that was one. Yeah.

And do one in the room? Room, yep. So, did you have the opportunity to influence compensation across sales and CS as a lever to help align there? Is that part of your journey?

Yes. We have the opportunity. I think it's really on sales, it's really whether or not they're going to get a, how much of the renewal base will they actually get in their quota, right? And that's going to determine almost everything about how interested they are in this topic.

We're going all ACV from a sales perspective, but it does include the protect number. So I think there's a sufficient amount of interest in that. Yeah, and as Joel mentioned, having top-down interest in driving a change here, right? And saying, hey, we are a SaaS company, we are embracing AI, right?

We are doing the right things to drive value and ROI for our customers. And having an executive team that actually looks at the compensation plans for each group. We've had some really refreshing conversations this year, I would say, on what are we motivating our teams to go do? And what are we going to go measure, right?

In terms of success. One area would be our professional services team, right? And so it's been very refreshing to have that kind of rallying around, hey, we're going to go measure what we care about, and we're going to go and award people for driving the right types of behavior. Super important though.

How's your definition of what counts as high-risk change since you first rolled this out and what caused you to revisit it? I think there was some nuance that you mentioned, Sarah, but I think one of those things was, hey, we need to make sure that we're, we have lots of products. And so not every, just because one product, you have a customer that has three different types of products, but we need to make sure that we're calling that risk out at the right product level. And then we kind of had to actually change the way we were doing the weighting.

Initially, we had no weighting, right? It was like, if you had $100,000 of, that was high-risk, and then you had $900,000, that was medium, different product, we basically called the entire account high-risk, and that was creating a lot of consternation. So I think it wasn't so much the definition of risk, but at the count level, whether or not we were going to roll it up as a high-risk. Yeah, great.

The fine-tuning was important. For sure it was, because that's where there was a lot of emotion. Maybe you went out here, like there? Yeah, who's creating the playbooks and the CTAs that you guys are now coming up with, and what's the process for that?

Yeah, it's actually, our customer success team leads, the creation of the playbooks, and we've got leaders across our customer success management organization that are driving particular product focus areas. But the good thing about this year and the rallying around consensus risk is that we're not doing it in isolation. I think in previous years, we might have tried to do some things and not necessarily had all of the key stakeholders at the table to say, hey, is this really the right approach? And oh, by the way, for us to be successful driving this play, we actually need product management to help us with some roadmap items, or R&D to help us with some tooling to help with data loading or migration or whatever it might be.

And so this past year, we didn't start off that way. We really evolved into a place where we have product management, R&D, sales, right, customer success, professional services, all sitting around the table talking about what are the things that are actually gonna move the needle, whether it's reducing the time to go live, or reducing the time to value, right? So it's a shared kind of conversation, but very prioritized in terms of these are the product areas that we need to go focus on, because they're gonna impact our retention risk most. Just to piggyback on that, what is that direct addition of like, do you tie it to total revenue or risk, or logos, or what's the word about the right time?

Yeah, and so I guess it would probably be two tracks, because I would say when you have your top customers, you're always gonna have some separate track around your customer advisory board members, or the customers that are speaking at every event, right? You're gonna have a special track for those. But for the bulk of the rest of the customers, right, we are looking at this data, our consensus risk data, and we are prioritizing it based on ARR at risk, right? And looking at it by vertical, by product area.

Timing of renewal. Timing of renewal, right? So we start to say, okay, there's a sweet spot here, right? If you're looking at a turn risk within your same quarter, you're probably in a commercial negotiation.

At that point, you're not gonna be able to go drive a product conversation of, oh, let's go adopt this, right, over the next four weeks. That's not gonna move the needle. They've probably made a decision already, and now you're trying to combat that decision, right, and in terms of the risk. But if you're looking two, three quarters out, hey, we can actually go influence one the roadmap, and we can influence the way that we're implementing the product, and so we can influence the outcome, right?

So there's definitely an art to the conversation. I think sales has a better appreciation of the different plays that you can run earlier that risk is identified, compared to what maybe they, I mean, they like to be the heroes. Everybody likes to be a hero and come in and save the day, but the answer is to actually get it due to your mitigation well in advance, right, so that you never get to a place where you have to be a hero. We'll get a quick couple of these.

What ground rules did you set to make sure that there was a collaborative approach and outcome? I think primarily it was just basically, the ground rules were, hey, the data needs to speak, and then we need to actually have a civil conversation about what the reason is for this, and then what the right mitigation efforts go on to be. I don't think it, once we actually got to that point where we're gonna do this, guys, it's like we're not running away from it, I don't think there was a lot of pushback. Right, and I will say so last year, our CRO and I got together and said, hey, we're gonna go drive this as a collaborative conversation.

And so we said, hey, what are our top 10 accounts that are at very high risk? We're expecting them to turn. There's likely little that we can do to save the account. And we went through account by account reviews with everybody that was involved, whether it was historical relationships, current relationships, right?

And we started off every call saying, this is not a finger-pointing exercise, we're in this together, let's go bring up everything. Nothing is off the table, don't be afraid of backlash. We'll just have an open discussion around what's not working, what we feel like fell down, and then let's quickly pivot into, okay, so what? So what do we do about it, right?

So from a cultural perspective, we just open up that door to say, we're not here to make each other look bad, right? We're here to actually make sure our customers are successful and ultimately we're successful. So it set a tone, I think, for a very collaborative year. Yeah, credit to the CRO that he was leaning into that, for sure, that made all the difference.

What gainside automations and CTAs are coming out of this and what, you know, I think we're not doing a lot of new stuff, we just basically modify our thoughts about what the risks of variety or what the CTA should be. I mean, I don't know how, when you come out of one of those conversations with Command Center, how much do we actually end up changing the system or do something different in gainside? A whole lot, I don't think so. I think it's really just refining what we believe to be true in the system, right?

Where did AI play a role in building this model or overall process? What AI tools were used? I think, I mentioned the machine learning model was already a part of the, and as a long running model, we just got better advertising for it, basically. We actually used it more effectively because of consensus risk.

Assessing what the right reason code should be, but we used machine, not machine learning, but just traditional LLM type logic there to kind of look at what was popping up and then how did that compare against industry best practice. And then, from a reporting perspective, I've been doing quite a bit of cool stuff just using Co-Pilot and Excel, which is extra powerful now if you haven't used it in a while. So I think we ended up doing some really good presentation layer stuff that was, made all that work that's getting done actually pop and actually help that prioritization process. Yeah, if you can imagine being in a meeting with the CEO and being asked to go through the top 50 up for renewals and talk about very specific details of what's going on in the account, what do they own, where do we have risks, what do we do about it, that's a huge set of data to bring together.

And so, certainly, you helped arm us with a dashboard that allowed us to kind of click through to say, okay, we know exactly what the install base looks like, we know exactly where product risk exists, and then we've got a narrative that's fed from Gainsight from all the CSMs putting in their weekly timeline entries, right? It feeds into this dashboard that helped us have a really robust conversation. So we were never really caught flat-footed. You saw that tab, one of those views was a Tableau dashboard that we'd spent a lot of time working on, because we thought maybe we can get self-service as a, we still have hopes of making this data self-serve for potentially even the CEO.

But we had a lot, at least in this first year, a lot of interest in making sure that we were wordsmithing and saying things just the right way. So that ended up being something we used internally a lot, but not necessarily for a large audience. I would use Excel to kind of take our manipulated kind of story line and actually turn it into like, man, I wasn't using any of my own formulas, I was using prompts only, so very powerful. What about you, anything else on this topic?

Don't think so from my side, and it looks like we're at time, so. How long, we actually just started literally a year and a couple months ago, right? Because we had just finished our planning process and we had rolled it up one more time, and so we started about 13, 14 months ago. And it's one thing to actually show it's another thing to actually operate based on it, and that's what we've done this year, is like our CEO is a, what's the medium, and high and very high, and I want those on the list, and I want them basically sorted by the renewals that are coming up the next two fiscal years, and I want to sort it by the different products, and then, so I mean, it's becoming kind of a rallying cry and a common language.

It's fantastic top-down support to drive the right type of focus for our organization. Absolutely. Yeah. Awesome.

Thanks everybody. Thank you guys so much. Thank you. Thank you guys, appreciate it.

(audience applauding)