Turning Customer Conversations into Action: How CivicPlus Uses Staircase AI for Proactive Customer Success

How's everybody feeling? Nothing like being the closer after a ripper of a Pulse party. Really appreciate the timing of all that. But so happy to be back here at the Pulse stage.

As Elliot mentioned, I am the VP of account management and customer success at Civic+. Formerly a gangster, I spent almost eight years here. I was present for pretty much all of the acquisitions, including Staircase, which was one that I was extremely excited about and have brought that into Civic+, as one of our first products dedicated to customer success. I'm really excited to show you what we're up to and what we're doing under the hood with it in a more guerrilla warfare style way, I would say.

Fun fact about me as well, too. Gosh, I've been in software for an extremely long time at this point. But prior to that, I was in the music industry, and I played in a hardcore punk band called Rad. And missed those days, but it kind of reminds me a lot of what software is like right now.

I don't really think there's ever been more of a punk rock era in software than that we're facing in AI at the moment and all of its disruption and all of its opportunity. So it's an extremely exciting time to be doing what we're doing and sharing with the community. So a little bit about Civic+. We are a GovTech organization with about 13,000 parent accounts that are really sort of core government.

Core, and what I mean by core government, is local, state. We do very little federal. But we do everything from hosting your town or city's website and securing it. So wherever you live, if you go on your city or your town's website, there's actually a pretty good chance that that's us.

And we're really sort of like the front door to your city and all of its connected services. So beyond websites, we do things like alert entire communities in a crisis if there is a tornado or any sort of natural disaster. We also social media archive to make things available for FOIA requests. We have asset management products for cities to be able to manage their lights and their fire hydrants and all of their city's assets.

So it's extremely rewarding and exciting work to be doing to be able to show up in people's everyday lives and feel like we're making a difference. What that sort of looks like for us internally is our CS and our account management organization is combined. Combined, we're about 95, 100. About 40 CSMs make up that load.

So if you think about what coverage looks like, pretty intense off the bat. 13,000 customers, 40 CSMs start to do the math and go a little crazy. And we have 15 different product lines, as I mentioned. So I just mentioned a few, but a pretty sprawling product line.

And if you think about what our accounts can kind of look like from a data perspective, it can get really hairy really quick. So if you think about the city of Las Vegas being a customer, of which they are-- thank you, Vegas-- the city of Las Vegas might use us for a website, but the sheriff's department might use us for their social media archiving, or they might use us for mass notifications or something else. So if you start to kind of think about all the permutations of what these customer relationships can look like, they can be really, really thick, like fast. Like a single customer might work with a support person, three CSMs, an account manager.

So when I came in, a huge goal of mine was to be able to sort of cohesify and simplify our model by bringing the two teams closer together, but also leveraging technology to help understand where we needed a segment, how we sort of group these products together, how we think about them moving more as a unit, and really sort of getting outside of what we call product instances. You could use that language for basically relationships, if you think about that in Gainsight Speak, and trying to pull all that together into a more cohesive motion. So this is just an example of a big problem statement that we had coming in. So Civic Plus is a tried and true sales force shop.

Basically do like everything in sales force. And that has been the motion for-- and sort of the go-to answer for Civic Plus pre-staircase. So within that as well, too, it makes getting a single view of customer health really difficult. But also, the really only sort of window that we had into customer health at all, pre-staircase, were telemetry signals for Pendo.

And if you can think about what that looks like across 15 different products without a whole lot of standardization, those telemetry signals were pretty thin. And they lacked a lot of context. And if you think about sort of like the primary go-to-market message for staircase, it's really this. The true early warning signals of churn don't typically lie in usage.

By the time you see a 20% usage drop, likely all your save plays are worthless. So a true early warning sign of churn lies in that buried customer email that you were never CC'd on. It's in that ticket that was unanswered. It's in that frustrated message to a below the line or above the line person.

It's lack of exposure and visibility. And folks not understanding or being able to leverage those signals quick enough to be able to make a difference. So when we sort of turn the lights on with staircase, which is an analogy that I've used a lot, when I stepped in, and my teams here, they'll tell you this, one of the first things that I started to talk about a lot was they kind of felt like we were playing tackle football in the dark. It was really difficult for each one of our internal people to know exactly what's going on in each individual account who's doing what.

So the concept behind staircase was we're going to illuminate the field. That's what we're going to do first. We're going to expose everything. We're going to show everybody where everybody is.

And we're going to be able to start getting signals instantly. And one of the best things about staircase, if you look at it in its comparative landscape to the rest of the ecosystem inside of Gainsight, I don't know if anybody are staircase customers here. If you are, raise your hand if you're not, if you're thinking about it, just so I can know. But one thing that you should know, and you probably do if you're a customer, is that implementation is really, really easy.

You start getting signals in a matter of hours. There is no such thing as like a six week, eight week implementation project, none of that. You're plugging in and you're going. And one of the first things that we did when we saw that was just a signal from a mid-sized account where we were able to start seeing an instant timeline.

So you're probably intimately familiar with the concept of a timeline with Gainsight. In Staircase, it's all AI driven. And you start to see bucketized events that come out of the box with Staircase. And one of the things that always terrified me when I was at Gainsight, and it terrifies me still today at Civic Plus, is you see negative conversation in red on the timeline.

And then you see nothing, nothing, nothing, nothing, turn. So if you illuminate that and you see that account by account, one of the best things for us was being able to really have a light bulb moment without going through implementation, without going through training, just leaning into new tools and leaning into AI, being able to visually see where those critical events are, and giving us the ability to get ahead of something that had nothing to do with any sort of thin signal that we might have gotten from telemetry, or any of the other things that we were plugged into before. This is a big light bulb moment for everybody on the team and created a lot of trust out of the gate. So that created a real catalyst for change within Civic Plus.

So this was kind of the mandate and what we were looking at. And if you think about GovTech and where its place in technology is, some of these problems are older school. Like you've seen this slide probably a million times in a million different permutations. We don't really suffer from any significantly special problems that anybody else might do.

But we are in a really fortunate position in government technology because of a couple of things. One, it's really, really tough to become a government vendor in general. And when you do, they're exceptionally sticky because it's even harder to leave. And if you talk to government agencies, their list of people that they are approved to buy from is pretty short.

So it's a great position to be in because we get RFPs for things that have nothing to do with us by the sheer fact that we are a trusted partner of local government for a myriad of different services and software solutions. So they come to us for virtually anything that they have-- any need that they have to discuss. So what does that do in terms of our numbers? Well, I mean, it creates a really sticky base.

So we have great GRR out of the gate. That's awesome. But underneath that, there's really sort of a whole lot of areas-- hides areas of opportunity. So one of the most exciting things about the prospective joining Civic Plus when I took the role was I learned that out of our 13,000 parent accounts, the average product that each parent customer owned was 2.3.

So if you talk to an account management leader, that is an absolute dream because that tells you that the revenue opportunity in the base is super significant in that inversion that you want to see in SAS that everybody goes for, which is the holy grail, which is where you punch the upper echelons of your TAM. And the growth opportunity you use in your customer base, that inversion happened at Civic Plus. So the opportunity is here. And because of the GRR being so good, it hid really, really areas of ripe opportunity for us to be able to go after the big, big price.

And because GRR was so good, being candid, it enabled some bad habits and some things that we know that we have to go change. Because GRR is so good, we have a bunch of volume-- we have a bunch of overwhelm. We've got a bunch of fragmented systems. We're reactive.

And things are fairly inconsistent in terms of playbooks. So our sort of roadmap for the rest of the conversation, want to talk a little bit about the blind spot more specifically and how we started tapping into contextual conversations versus telemetry as a road forward, what our big bet was, as well as how we started to operationalize this in the business, what to look at when you're trying to operationalize, and where we're headed with this thing. And of course, if everybody has any questions, I'd love to hear. So the blind spot, really, as I mentioned, totally had to do with the fact that we really had only a single view into customer health, and that view being Pendo.

Each one of our products is really unique in its architecture. It's unique in how it's built from a code-based perspective. It's unique in the sense that logins and typical things that come out of a Pendo readout are more difficult and easier to get. We don't really have a consistent view of being able to do this across each one of our products at a standard that is relatively the same.

So Staircase really sort of plugs that gap for us in terms of being able to pull in all that contextual data and make that conversational. So we're able to overlap that and enrich the signals from Pendo that we get. And what this has really enabled us to do is move at a level of speed that we've never been able to move to before. And how we started thinking about operationalizing this is really in sort of five steps.

So the first step is really just getting everything plugged in. So making sure that when you're looking with Staircase, any sort of signal that you can get it given-- so Slack, email, Gong, literally anything where you see customer signals, that's where you want to be able to play Staircase, because it's exceptionally good at bringing them all together and even better when you layer on an MCP like Claude or Chat GPT. So that's step number one. Step number two is starting to get into a rhythm around how you're detecting the signals and looking at that like a macro and a micro view.

So what I mean by that is within the second step, there's actually a little bit of a split there. Because what AI really allows you and enables you to do is go from 30,000 feet to 2 feet in an instant. So you can get these ultra crisp, clear signals at an account specific level. But you can also start to identify trends.

And where these trends really start to help is being able to level up what you're trying to do in a digital one-to-many motion. So that split is like, how do you apply this, and how do you operationalize this at an account level? And then how do you start to identify cohorts that are going through similar things in your company and leverage not just the MCP, but the data to be able to guide those cohorts through an orchestrated journey that you control? The third is then meet your people where you're at.

I don't know if everybody had a chance to listen to the glean CEO. But that's a concept that I genuinely believe in, is like meet your people where they do work. For us, some of the things that I've seen some early wins in in terms of how we operationalize this is the ability to pump in automatically all of our findings and the play directly into a Slack, because we have so many people that are running these plays and that are responsible for different things across our entire portfolio, being able to meet them all in Slack, tag them in a single thread, drop in the signal, all of its context in the play from our staircase MCP directly into Slack has been a game changer and something that I want to double down on. The fourth is getting your reps in and being able to really start to scale it.

And then the fifth, super, super important, particularly with staircase, is make sure that you're training it, right? Give it feedback and make sure that you're killing false positives and keeping the trust and the integrity in the data. So it's like anything else, right? If you're trying to drive adoption of a solution, all it takes is a few negative experiences where a rep takes a look at something and they don't necessarily trust it for them to start to disengage.

So I would say that this is true with relatively any AI native offering or whether it's staircase or anything else, but engaging, giving that feedback, communicating with the company, communicating with your CSM if you're working with staircase specifically, flagging those false positives, documenting them, sending them, so that the model gets better and better and better. That's super important. What does it actually look like to operationalize? So the hard part, the easy part is like buying the tool, getting it turned on.

The hard stuff comes in operational, when you're trying to operationalize anything, is really has to do with the RACI. So you gotta get the basics down. The foundation has to stick. You have to build that trust amongst your teams.

And they really have to be the ones driving it on the ground floor. And that's the super hard part. I think everybody can get behind the fact that with staircase and with all these tools that are available to us on a daily basis, the amount of output that you can get and the amount of things that you can proliferate with it are amazing. But how you actually drive value is that repetition and folks like really getting into it and understanding the foundation and the basics of what they're supposed to do when, right place, right time.

There's sort of never any substitute for that in my opinion. And no matter what or how sophisticated AI is, if there's a human in the loop, they have to know what they're responsible for and they have to be able to execute. So for us, a couple of the things that made it stick as I mentioned was the signal really landing in Slack being more actionable outside of a dashboard. Every layer it was tied to a play and not just a flag.

So we're able to follow up that context with really rich suggestions about what to do. That really puts like the end user more or less in like that editor seat. They're able to assess the play, try to understand if it really applies, how they would augment it. That's all great and part of the feedback process as well too.

So those are a couple of the things that started to help the operationalization of this whole motion start to stick. But being candid as well too, we're still on our journey and we're still figuring it out. So for us, where this is really heading. So outside of really leveraging Staircase honestly as like a core CS platform within Civic Plus itself, we have deployed not just GPT to our entire, all of our AMs and all of our CSMs leverage it on a daily basis, but we're also leveraging Clod.

And specifically with the Clod MCP, there's been some really great things that we have created that we're starting to standardize. And the use cases just keep coming. That's one of the best things about this. Our folks are super excited about it.

My leaders come to me like every week being like, look at this awesome thing that I built. And all the things are awesome. And they're really, really helping us. I think for us, it's about trying to make the right bets and choose the right things that we want to level up, show the rest of our teams that are helping us systematize what this looks like across the rest of the business.

So there's more to that. How you really do that has to do with how you get things behind a firewall, how you really, you get the rest of your systems interconnected with it. So we're seeing great output, not just at a leader level, but at a rep level as well too, of things being created all the time. But these are a few things that we have decided are sort of the next three plays for us really.

So CHS is our reference to what we call a customer health score. So pre-staircase, the way that customer health score is worked essentially, were these telemetry signals that I've talked a lot about. They get fired off once a month and they get exported into a Salesforce case. And it's essentially like a zero to six level.

And if something is like gets to a five, that's when it's like, okay, like house on fire, we have to bring everybody in and there's an alert that goes out. But by the time that happened, obviously it was too late. So we had to start figuring out contextual ways to put more meat on the bone behind these signals to be able to give people real context behind the signals that they were seeing so that they weren't necessarily reaching out, like they weren't even necessarily knowing what they were reaching out for. With staircase, we were able to fill all those gaps and really like create a very nice experience for people to go to and understand what the risk is automatically.

Better than that, we're able to automate like a lot of the outreach, the initial outreach to at least get the signal off the ground and the workflow started for the rep. So that's use case number one. I have some visuals that we'll share in a minute. The second is the dark account playbook.

So this is a funny concept that we started talking about a while ago when we were trying to build this, which was pretty much everybody, no matter what business you're in, has a digital or one-to-many tier and you just might not know it. Your digital one-to-many tier for a lot of people are just your customers that don't talk to you. So like, how do you fix that? And in a one-to-many fashion, that's one way that AI can be super, super, super powerful.

Not just with staircase, but if you think about being able to leverage an MCP like cloud, being able to take these signals, hook in an odd, like create from nothing to like a very articulate five-step sequence, rig that into outreach, like through the same MCP connection, now all of a sudden, like you're able to take these cohorts, and again, I ask folks to raise their hands if you're thinking about staircase or if you're staircase customers, but there's this concept of out-of-the-box reporting there, which is actually really, really good. And if you go into the topical stuff that is there, that's where you can really find rich context around cohorts of your customer base that are experiencing nuanced things that you might not have known about before. That stuff comes in staircase right out of the box. You plug that into MCP and all of a sudden, like you have a very sophisticated digital motion, and you need like literally two tools to do it.

And what used to take like weeks of saying, okay, I wanna reach out, I wanna do a dark account campaign, you gotta go talk to customer marketing, you gotta go get all the right materials, right? You have to figure out who owns outreach, you have to get them to help you stitch together like all these different messages. Like before you know it, you're looking at like a week of burn time and if it gets off the ground at all. So what this has really done for us has made the cost of piloting things like literally nothing, right?

And particularly in a digital tier where we're trying to activate a long tail that we don't communicate with a lot, that's super, super valuable. So that's a big one for us and one that we're really, really focused on because we just do have so many customers, especially when you look inside of our architecture, as I mentioned, we've got 13,000 parent accounts, but we have this sprawling array of relationships and customers underneath that within those parent accounts. And that for us is typically where we see a lot of the dark motion. So we wanna activate that part of the base and we're leveraging Staircase and Cloud MCP to do it.

The last piece is definitely more of like a frontline coaching, management, performance, sort of always on motion, which is just like how easy it is to be able to really, really help a rep understand what the key things are in their account base and how to get after it and work in like two weeks sprints. So one of the things that we did was literally just pull customer account lists, the book of business lists, four reps, insert it into our MCP, ask it to retrieve the top 20 risks and the top 20 opportunities within that book of business specifically and create a two week sprint plan for each and every one of those accounts in that book of business. And then all of a sudden you've got like a totally ready to go like roadmap for a rep to focus on for however long you want your sprint to be. And that's also been a game changer for us in terms of being able to really highlight for reps at that 30,000 foot view going down into two feet about what is actually happening in their accounts on the ground.

And I will say as well too, we're working on adoption because we went from like Salesforce only to this, like pretty much overnight. So some of the reps are still getting used to it, but it's working out really well. So those are the three things that are top of mind for us and sort of what's on our roadmap in terms of use cases. But here is a handy visual of what our new case workflow looks like versus the old.

So we're able to go get a dash around all of our customer health score cases and be able to create messaging sort of out of the box that is nuanced to each one of those situations. And when you think about like how big of a difference and what that looks like, like when somebody was literally, all they had before was a zero to six and now they have this and they're able to go in and see rich context behind each one of those signals, there's pre-baked ready to go messaging for them to kick off. And we are sending that out on that behalf. If the customer replies, reps on it, they're off to the races.

If they don't, on goes the next message. If they don't reply after that, maybe we go back to Pendo and we send them in an app. And if they don't reply after that, they run through the sequence. But this has been a big game changer for how we sort of put a lot of context again behind very, very thin signals and not just get an sort of exponential value out of like an archaic zero through six system, but really sort of take it to the next level.

So lessons for any CS leaders that are trying to do things like this now. So start with something that you really can't afford to lose, like a big renewal and work backwards from there. Nothing really puts the feet to the fire, like leveraging a big event, right? So if you're able to hit your wagon and try to apply some of these workflows or these concepts to a time bound event, like a big renewal, that's a great test case.

And one of the things that I've told customers over the years at Gainsight is, gosh, what better way to justify the value of CS than tying it to revenue? So if you're going to try a new motion, tie it $2. That's my suggestion, always. The second again is the tool is the easy part.

The work comes in really getting the human buy-in, right? So there's been so much, and there always is at Pulse, talk about what the humanity element of what we all do, how that intertwines with the tools that we're using, and that truly is the hard stuff, right? How you get your teams working together, how they understand their roles and responsibilities. The way I've sort of been putting it with my team in how I sort of think about leveraging all the AI that is available to us is I would love to see you try to like leverage Claude to automate 50% of your job that you know is bullshit, and then go 110% on the 50% that matters, right?

And that's sort of our biggest concept and our motto in terms of how we think about trying to adopt this. The third is that AI again can be your sharpest analyst, but not the decider. There are false positives. There's, I'm sure everybody is probably a little bit fatigued from AI slop, right?

There's like a lot of AI slop out there. That's where the human element really comes in. You don't wanna sound formulaic, you don't wanna sound canned, you wanna sound like yourself. You have a personality, you have a voice.

Your voice matters. Don't let the robot take your voice, right? So for, I think one of my biggest takeaways from just the era that we're living in and the things that I've seen, especially over the last 24 months or so, the last couple years, is really the companies that are gonna win in the next decade at NCS aren't gonna be the ones with the most data, but the ones that are able to leverage it to their advantage the fastest, right? So that's what this whole era has been able to do for all of us.

It's something that I'm really excited about, particularly in the GovTech context, because it's no secret the government tends to be the least behind, or the most behind, rather, in terms of adoption. So being able to bring these concepts into our space and hear how elected officials, city clerks, all the folks that we talk to on a daily basis, how they're thinking about leveraging AI, it's a hell of a lot different than a B2B context, but to be able to bring that speed to government is something that is super exciting, because as we all know, government can be a bit slow, and GovTech shouldn't be. So that's also the motto. All right, that's what I had.

(audience applauding) I'm ending a bit early, but totally happy to take any questions. I don't have a Slido, but if anybody has any questions, you're more than welcome to just ask or find me after. Can I answer anything? Does anybody have any questions?

Who's leading our change management? (audience member speaking off mic) Yeah, hey, that's actually a really good question. So we've talked a lot about forward deployed engineers, and that being our concept. So we have an AI center of excellence that we've had, and we have made a decision very, very recently, literally within the last month, weeks, where we're taking that concept and we are extending it to every division of the company.

So every team is going to have its own dedicated AI forward deployed engineer that is taking all of the stuff that we're building, assessing it, helping us assess its actual value, and then if it's something that we wanna make a greater bet on, they are pushing that up the chain to help us systematize it. Oh, I do have slide questions. Okay, what all data sources did you connect to Staircase 2 in order to mature the confidence level of your signals? Okay, so obviously the first one and the biggest one is always gonna be email, right?

That's where you're gonna find the most hidden signals and probably the most surprising things because you're not CC'd on company emails all the time, right? So there's a lot of stuff that you just don't know because of that sheer fact. So the first one and the biggest one is always gonna be email. The second one that was huge for us is obviously Salesforce.

We have gone integrated into our instance of Staircase. We have SharePoint, Teams, basically everything that we were able to turn on, I said yes to if that answers the question. How do you avoid signal fatigue with so many notifications, especially with false notifications? Great question.

This has been something that Claude has helped a lot with. We still have a lot of work to do with this, but I hear the signal fatigue thing. In fact, I used a moniker yesterday about Staircase, which is like, man, if you've got a little bit of ADHD, Staircase can be like, man, it's a gateway drug. You can go so deep, so fast on virtually anything and it's easy to go down a rabbit hole because you're like, oh my God, what happened then?

And then, oh wow. It's like you're on this weird wild world of discovery, basically across everything that you find. And it can be super easy to just get lost in the sauce with that, right? So Claude MCP has been exponentially helpful with that because you can just literally say, show me the top 20, create the top 20, tell me why they're the top 20.

You can really cut out noise with Claude and with chat UBT in general. How did you get your CSMs to trust the signals coming from Staircase, especially early on when they were used to relying on their own gut instincts? Man, that is also a really great question. I think, and I'd be really curious to hear any one of my leaders back this up or not, but I think that they were so floored by the fact that this stuff actually existed that they were like, oh my God, I went from being able to see nothing to being able to see everything, that's great.

I don't necessarily know if we have that big of a trust problem with that. But yeah, I think that just the sheer fact of it showing up and going from not having any tools to having all of them is something that they're still getting used to, something that we're still getting used to. So how did you handle change management for the launch of Staircase and workflow changes? We are implementing now that we have very hands-off CS leaders, so I'm worried that my work is going to ultimately fall flat.

Here's my take on this. I think that tools like Staircase aren't gonna push anybody into a space where you feel like the work is flat. If anything, I think it highlights what work is worth doing. That's also one of the things that is great about all the tools that we all have available to us now as well too is that being able to identify that 50% of the stuff that I said, you should try to kill off in your role with AI, that's where I think the magic is.

So I wouldn't think about it too much as like, is my work going to fall flat? I would think more about that mantra of what is the 50% of my job that I want to go 110% in on. That's how I would think about it. When you started, did you roll Staircase out across your whole CS team at once?

Or did you start with a specific segment and how did you decide where to begin? Big bang, just wait for it. Yep, it's got everybody in there. With government, data regulations become increasingly complex and varying just by state by state agency.

How is Staircase ensuring compliance with differing requirements around governments, retention and PII? How have you operationalized this within Staircase data ingestion architecture? Well, okay, that's one that could be above my pay grade. But I will say, so getting Staircase through security at a government tech organization is a process that I don't care to relive.

It was hard, it was really hard. Getting anything through a government tech organization is really, really hard. And the truth of the matter is, is that people that are above my pay grade helped me do that and I didn't have a whole lot to do with it. So I can't give a whole lot of depth around some of the back end parts of that question, and I don't know if that answer's useful to you at all, but much like any other GovTech company, we went through a really rigorous process and I also have apologized to my alma mater more than once for how bad we abused them during it.

So given your public sector focus, where's the, sorry, I lost that one. What if there are strong term signals from an account that is not CS-owned, part of your long tail, or digital program, who do the CTA's get assigned to? Great question, so my digital leader, who's sitting right here, her name's Marissa Shepherd, she dishes them out. So she's like our overseer for the whole digital sector.

And when we have names on those accounts, in names only for stuff just like that. So if, for lack of a better word, something hits the fan in the digital sector, Marissa's there to tap that named resource, despite the fact that there is no expectation that they're going to work that account like any other that is in their book that is not digital, they are there for human intervention, and we have that plan, and that is a part of their overall book of business, although not related to expansion quota, runinal quota, anything like that. Doesn't really go into the numbers in terms of how we think about getting to plan, but wanna make sure that there is a human in the loop no matter what, government kind of expects that. So if something crops up in that segment, we dish it out to a name.

Okay. What part of your dark account play seem to have the biggest impact? We all know it's really hard to digitally re-engage dark accounts, so curious if you've had any specific motion that seemed to really resonate with customers. So, candidly, we're early days, but one of the things that I was the most, the most kind of shocked by was how awesome Claude is at taking fairly little context from a whole variety of disparate places, piecing it together and putting together an unbelievably personalized email.

So in the dark account play that I showed you, like pre-staircase, I mentioned too many times now, zero through six signal, right? But it's also able to say, "Hey, you still have this ticket over here." And also, we know we talked to you about this other product a few months ago, and we know that very civic plus-specific lingua franca here, but we know you're coming up for a redesign with your website. It can take all that context and key into the most important parts about it and make it sound super, super personalized. The last campaign that we ran, Marissa sent me the early stats, and within an hour, at least we knew that the majority of them had opened.

We had a 75% open rate and a high click-through rate as well. We weren't measuring how this was done in a human way before because we weren't doing anything from a one to many. We weren't leveraging outreach. It was always a hand-to-hand thing, but I can absolutely tell you that it wasn't working well before.

So the fact that we're able to get that really, really account-level context and all the things that wrap around it without any effort, and then being able to drop that into a message that's really custom-tailored to the person, awesome. You mentioned your instance uses relationships to represent an account on active products. How do you go around the staircase limitation? Okay, great, great question.

Answer, that limitation is like no more very soon. Know that to be factual because I asked while I was here. I think that if you are using both Gainsight and CS, that the relationship issue is solved. If you're using just staircase like I am, it's coming.

But also, one of the things that's been really good about this and how I've done it kind of in a punk rock gorilla-wear-first-star way is we just get the product instance owners, accounts from Salesforce, just like the raw data file, like a CSV, you throw it into cloud, you use the MCP and you say, I want you to take all the context you can find around these sub-accounts. They are connected to this master list. And the results that we've been able to get out of that, I have genuinely questioned if we actually need to go the relationships route because it's been so good. It's harder, you gotta go pull reports manually, you gotta feed those reports manually, you gotta be really smart about your prompting, you gotta iterate, you gotta edit for slop, but we've done it pretty effectively as well too.

Okay, given your public sector focus, how did you get internal approval to use staircase without GainSight having FedRAMP certification? That's also one that's probably above my head. So I wish I could give you an answer to that question, but I can't. Do you see any reason to pilot staircase with a few CSMs for several weeks once purchased or go live with CSMs all the way?

I'm a Band-Aid rip it off guy, if you can't tell. Yeah, so just do it. What's your biggest pushback on using staircase? I'd say I hear Signal Overwhelm, I hear I'm not quite sure what to do or whether this play is actionable or not.

My sort of answer to that is those are all kind of excuses to be honest. Like it's like you can literally ask your MCP and Claude like what should I focus on within this data set? And the answers are fantastic. I know now we're over and I burned through a lot of questions.

I don't know if I got them, it's all of them. I didn't, but I'll certainly answer. Yeah, anyway, thank you guys. Hope you enjoy the rest of the conference, however long you're here for.