AI Tools Don't Scale, Systems Do: Bonterra’s Approach to Content Orchestration

Welcome and Introduction

Steve Kearns: Hey there, everyone. We are so excited today to be joined by Neil Grosso, who's customer advocacy marketing manager at Bonterra. And we're here to talk to you more about this concept of AI systems and content scale, specifically how we worked with the brilliant Neil here at Bonterra to build a content system, not just sort of a one off, kinda content brief with AI. So with that, we'll go ahead and jump into our housekeeping. So feel free to drop questions into the Q and A. We will go ahead and make sure we answer them after the presentation wraps. We'll probably have put about ten minutes of Q and A toward the back half of this webinar. But we're very much looking forward to sharing Neil's story with you today. So with that, maybe we'll start with a quick round of intros of who's all on the call. So hey there. My name is Steve Kearns. I lead Jasper's customer advocacy function. I've been with Jasper for about seven months now. Actually, I used to be a Jasper customer, and very excited to be here with you all today. Maybe I'll pass to Sara Mo to introduce herself as well.

Sara Mo: Hey, y'all. I am Sara Mo. I am a product marketer here at Jasper. I am going to be talking through what's been happening in the market and what led us to build this tool that Neil has been finding success with.

Steve Kearns: Maybe, Neil, I'll have you do a quick intro right now too.

Neil Grasso: Sure. Hey, everybody. As Steve mentioned, my name is Neil Grasso. I'm the customer marketing manager at Bonterra. Right now, I oversee two main sections of our customer marketing function. The first is what I call marketing to our customers where we are owning the campaign management side of the customer marketing function. The other is our customer advocacy side of the business where we do all things customer advocacy program, customer reference program, and then specifically, as we'll talk today, case study development and review generation response.

Market Context and Content Systems

Steve Kearns: Awesome. Well, thanks so much folks for the introductions. We're super excited to be here with you today. And just so you all know, you will get a recording of this in your inbox afterward if you have to hop off early. So do not worry. You will get all of these resources in addition to the customer story that we published with Neil in your inbox following the wrap of this webinar. So we're looking forward to diving in. So with that, let me start by sharing sort of what we're seeing in the market right now. We have this incredible opportunity here at Jasper to work with customers across the B2B and B2C ecosystem, different sizes, different verticals. And generally speaking, what we're seeing in the market is that there's a lot of folks that have adopted AI, scaled AI tools even. But folks are producing a lot of one off content with the LLM. So think your Clogs, your ChatGPTs. Even a lot of folks are using Jasper sometimes to only produce your ads, your emails, your blog posts, social posts, web pages, which, fast backwards, I should say, about five or so years, was a pretty revolutionary concept that you could take something from your briefing stage to sort of that final output stage on a one to one basis in a matter of seconds or minutes. Pretty revolutionary idea. But what we're here to talk about today is customers and individuals, companies that have moved beyond this one off content production to actually building what's called a content system or a content pipeline. So this is what we do here at Jasper and customers like Neil do such a fantastic job at, which is the concept of taking something from sort of a shared context layer — your concept or your brief — running it through a system of execution like Jasper, and then really being able to scale your content in an on brand governable way, anything from ten assets, a hundred assets, a thousand assets. What that's gonna do is take you from this kind of one off mindset that so many of us have fallen into the trap with AI tools, to building a repeatable content system that can help you free up time to do some of the things that you may not necessarily have had time to do before. A lot of that is why we built the system that we built, and folks like Neil have been taking brilliant advantage of it. So we're excited to share his story with you. But with that, I'm gonna pass to Sara Mo to talk a little bit more about the shift between the one off content creation that we're seeing from different parts of the market and what our vision is for where we think marketers can be successful with AI tech.

Sara Mo: K. Thanks, Steve. Alright, y'all. Just as Steve talked about, we're seeing a shift in how marketing operates across this broader landscape. We're moving from manual execution to intelligent workflow, from siloed SaaS stacks to agentic systems of execution. We're moving from marketers as doers to marketers as orchestrators. The shift in question goes from how do I create this asset to how do I build the machine that creates it? And that's the shift that we're looking at in the market today. So last year, this was the reality for marketers. And when we were looking at how can we help marketers solve problems, how can we support this shift, we asked ourselves: how are we gonna solve this problem of the reality where content demand had already doubled, seventy percent expected it to grow another five times, and more than half of marketers are spending over forty percent of their time managing reviews. So demand was multiplying, and this is still the reality. Demand is multiplying, and headcount is not increasing. The operating model just hadn't yet changed. So that's the real problem. And options were limited. On one hand, you had copilots that can give you quality with the right inputs, but there was no system and no governance. And on the other hand, you had automation that gave you scale, but there was no marketing intelligence attached to it. So you're forced to choose quality without scale or scale without quality. So that's the reality that we understood when we were trying to solve this problem for marketers. And I'll just talk a little bit more about when we're talking about scale, what do we actually mean? This is a scale that all you marketers know exactly what I'm talking about. You have three assets, but you need three variants and five languages across five segments for five different channels. Just that right there is over a thousand pieces of content, and that's the modern math of modern marketing that no team can keep up with manually. So I'm gonna toss it back to Steve and Neil to show how they built this system in practice over at Bonterra and what the wins are that they're getting.

Bonterra Overview and Review Challenge

Steve Kearns: Amazing. So then, Neil, we can go ahead and dive right in. What you'll see folks here on the left side of your screen are just some of the things that Neil has been able to achieve by creating this content system, specifically through the lens of automating review workflows. So with that, Neil, I'll pass it off to you to share a little bit more about yourself, what you look after at Bonterra. I know you gave us a quick intro at the top, but feel free to share a little bit more about some of the different things that you look after and then what Bonterra does as a company.

Neil Grasso: Yeah. Hundred percent. And I meant to kinda mention that before. Apologies. One of the things, the way that we describe Bonterra internally is we are in the social good space. So we have a variety of platforms that serve as solutions for nonprofits as well as for profit institutions, like financial institutions and banks for the CSR side of things. We also offer a variety of case management solutions for both the public sector as well as nonprofit service organizations. We are split — I work on a team with my supervisor, Susan Bell, who is excellent, as well as my colleague, Kathleen Connolly. We are split by product category for a lot of our different workflows. But for review generation and response, this is something that Kathleen and I work all encompassing, and we're kind of working together to divide and conquer here because we have so many products that get so many reviews on a monthly basis. So we wanted to work with Jasper to help solve a problem we had with the review response protocol. Basically, it's always best to respond to reviews as quickly as you can. Ideally, you would do it the moment the review comes in, but when you have so many different products across so many different networks — G2, TrustRadius, we also have some unfeatured customers as well as Capterra's networks — it just can be very overwhelming. So we chose Jasper for this because we had tried to previously do something similar where we were going to automate a good chunk of our responses through ChatGPT. We built some pretty complex prompts, but it just really, really wasn't working all that well. This was a project that we viewed as a high volume, repetitive, but brand critical project. So it was a perfect candidate to try to work with a team like Jasper and a system like Jasper to improve the outputs after we had failed to really get the inputs we were looking for on ChatGPT. So, yeah, that's what we partnered with the team on. We worked in Grid to do so. I wanna just also call out that I became a power user of Jasper both from the actual content creation and document format as well as on Grid, but I'm not an AI expert. I'm someone who just really bought into using Jasper and tried to work with the team to get the best possible outputs for this project, and we wanted to deliver in terms of speed and quality. And we were able to do so. I'm really happy about that success. So, yeah, just really excited to talk about this project and all the ways that we use Jasper to achieve that efficiency.

Steve Kearns: Yeah. Neil, I appreciate you sharing that. It's interesting. I actually started my career in social marketing. And one of the expectations that was set very early on as I was learning the ropes was that, especially in the world of brand comms, when someone takes the time to communicate with a brand or communicate about a brand, it's best practice that they get a response. Now, of course, with the proliferation of messaging into different digital platforms and what have you, you've got things you gotta monitor. You've got outbound. You know, it can become pretty overwhelming pretty quickly to actually keep that charter of someone communicates about your brand or takes the time to leave a review, and then you respond. So it's something that is, to your point, a high value task in that it's important for the user to get a response. But it's not necessarily something that, I would imagine, a lot of those responses kind of look very similar. Talk to me a little bit more about how you built the system.

Building the Review Response System

Neil Grasso: Yeah. Yeah. For sure. And you hit the nail on the head there. It's one of those things where I describe it as it's high value in terms of the actual content being there when it needs to be there. But from a strategic lens, I would say that the majority of review responses are not where a marketer should be spending most of their time, especially as we try to build out programmatic systems and programmatic content systems. So when you think of review response, the majority of your reviews, thankfully for us, are happy customers that don't have any complaints. But we saw it as kind of like a tiered system where that might be eighty to eighty five percent of the reviews we see across our products. But there were also three other categories that we wanted to flesh out. So it's something that we originally tried in ChatGPT, but it really didn't work out that well when we tried to put this into play and tried to get review responses that were what we expected from a quality perspective. And we wanna make sure that our linking strategy and everything else that would be associated with a review if a human were to be doing it needed to be there. Otherwise, the value of that review response just plummets. So to give you an idea, the way we saw it was we thought that there were four major buckets — we labeled them A, B, C, and D — of the kinds of reviews we saw across our networks. So bucket A was a great review, no specific feedback. Just what you mentioned, Steve. It's like, hey, love Bonterra Network for good. No complaints. Super thrilled with the experience. And that would only necessitate a very standard review response of, thank you so much for your positive feedback, we really appreciate it, and then basically a sign off. Bucket B was a great review but with minimal feedback. So they overall had a great experience, but they have a product enhancement request, or they're a little confused about how to do something. So that needs to have some sort of escalation route to get them the answer they need. Then we had bucket C, which is a critical review with some vague feedback. These are pretty common if someone doesn't have a great experience or just a mediocre experience. And then of course bucket D where somebody has a very critical review with specific feedback. That was a kind of review where regardless of what systems you build, I'm a personal believer that the human has to get back involved for that kind of a thing and then escalate whether that be to understand the client situation, see who else has worked with them, etcetera. But like I said, the majority of reviews are super, super positive. So I wanted to build something that could think about this problem systematically or programmatically, and Grid was the solution for us. There was no code — it was a no code interface, so I didn't have to bring in engineering support to build this. And I was able to build in Jasper IQ so that our brand voice and all the information across our knowledge base was baked into the workflow so that all of the responses remained on brand. And Steve, I'm gonna kick it back to you, but if you have any questions about how I did that specifically, I'd love to answer.

Steve Kearns: Yeah. No. Absolutely. So I've got a couple of questions. First, a thought in response to that, and then I've got a question on IQ specifically because maybe there are folks who are tuning in who don't know a ton about IQ. But it's kind of the foundation through which you would build a lot of this content and be able to scale it. So I'd love to hear how you would describe IQ to another marketer who might be kinda struggling with that brand governance or consistency or scale with their AI tool. But what I'm hearing from you — and you let me know if this is a correct interpretation — is look, as you seek to build a content system, you have to find the patterns through which you would do your work, right? Which is, hey, reviews are showing up in one of four different buckets or five different buckets, and they're going to generally look like that. If you build something that can then help you systematize that, it starts with a fair bit of work up front, which is, hey, here's what these reviews are gonna look like, here's how I would recommend responding to them. And then only once you define that system can you actually get to the outputs you want.

Neil Grasso: Yeah. Yeah. A hundred percent. I would say to kinda answer your questions one by one there, for the IQ layer, that is something where — think a lot of teams, and we have thirty plus team members on Jasper across our customer marketing programs team and our corporate marketing team — it can be difficult to adhere to a brand guideline just in general, even if you have some expert level writers. So on the first piece, what was great about Jasper IQ is it was able to take our brand guide into account for all of our products and our brand at the Bonterra level so that we weren't — and I think most importantly, you kinda wanna make sure you're not saying anything you shouldn't. So that's really, really crucial. And then on the second piece to the Jasper IQ, the knowledge base is great. But to your point, Steve, what was really important was taking examples of review responses that we have delivered by hand manually in the past and saying, hey, this is what good looks like, and we're going to build a system to help you detect what you should be looking for, what's important, then what should be included in your response. By anchoring it with that example of, hey, this is what a response in each of these four buckets would look like if it were considered a great response — we would totally fit the bill and make the customer happy and make me happy as the marketer. Like, that was critical, and it was really great that Jasper has that built into its ecosystem. It's all about how you think about solving this systematically.

Live Demo: Bonterra Review Workflow in Jasper Grid

Steve Kearns: Yeah. Could you actually take us for a tour and show what it is you built? I think it might help show folks what the system of execution looks like and what it looks like in practice.

Neil Grasso: Yeah. Absolutely. I just think I have to — you might have to stop sharing so I can start sharing. There we go. Okay. Great. So can everybody see my screen here? Let's see.

Steve Kearns: I can see it.

Neil Grasso: Cool. Awesome. So yeah. So what you'll see on the left columns here before Bonterra Review Analyzer — this is the grid, and this is kind of how we wanted to approach these problems. We got a Looker report from G2. It's one of our review networks, and we actually took all of the reviews across multiple products. I think there's like four or five products mixed in here. Along the left side, I have cut a little bit here because I don't wanna put our customers' information on display, but all of the important pieces are still visible. So what you're seeing is all of the information from those Looker reports that came from those reviews — product, the star rating, and then some of the key questions they asked during that customer survey — all of that is taken into account, and then there are two agents running here on the grid. So what you're seeing in the column titled Bonterra Review Analyzer is what we wanted to do is almost have a trust or verification step bubble. Is the bot — is Jasper in this case — detecting what we would be detecting if we were sitting in the role and trying to do this manually? So does this fall into bucket A, B, C, or D? What was great about this was when we first started out, we wanted to make sure that that step was right. And when it wasn't, we were able to make those corrections so that in the future, the logic would be applied in a more efficient way for the second agent, which is the actual review response generation where we have entitled here as Bonterra Feedback. So this was great because it was able to take in and ingest all the information from the review itself. It was then able to tell us, hey, this is what we think is the analysis of what bucket it falls in. So in this case let me see if I can full screen this. Oh, hold on. There we go. So it's able to give us a breakdown of all the things it took into consideration. It was able to properly categorize this as positive but constructive. And then when it actually generated the response, it had a review that followed our guideline of addressing the individual's name. In this case, they were anonymous reviews, so it just says hi there. It breaks down all of the positive parts of the review as well as points to the right location in terms of the training link for the inquiry that they had. And that is like — I said it's like ninety percent of the review response game. Like, eighty to eighty five percent of your reviews fall in that first bucket. And just being able to react and respond properly helps us so much so that we can focus on some of these that are — we have a lot of four and four point five and five stars here on screen. But for those that are two, three, maybe even one stars, that allows me to focus all of my manual time and review response on the most critical cases. And that is just such an efficiency upgrade compared to where we were just a couple of months ago.

Steve Kearns: Yeah. This is great. Thank you for sharing that, Neil. I mean, one of the things that I wanna drive home for folks who are listening is that you — and again, it's in the spirit of building a content system — content systems are all based on the inputs and the context that you're gonna give the system of execution like Jasper, which will then give you the scaled outputs you need. So at the front end of this, you set up the system, which is actually rooted in several agents, which are shown across this top row here — the Bonterra Review Analyzer, the Bonterra Feedback. And it sounds like you went ahead and you actually created those agents to be able to classify the reviews that are coming in into one of those several buckets you mentioned. And then the next agent is actually producing the feedback that you would then review and then publish.

Neil Grasso: Yeah. I mean, you hit the nail on the head again. I would say where I give my team and myself credit was that because we had tried this before in ChatGPT, we did have lengthy instructions of how we wanted things to be formatted and what success looked like. It just simply was not working with those vertical LLMs as you described. So I also wanna shout out your team on the solutions architecture side of Jasper. They were so willing to meet with us multiple times, tinker with our existing prompts, and then build something for us where we were able to get these two agents that now are extremely reliable for our purposes.

Authenticity, Brand Voice, and Time Savings

Steve Kearns: It's, yeah. We've got a great team over here that are helping a lot of our customers kinda build some of these things that might be a little more custom. One of the things that I always advise our teams to go in with as we're talking about this concept of content scale and inputs that result in scaled outputs is also just asking our customers what problems might they have, or what are the things that they have not yet solved for that they would love to alleviate off their plate. And I think the review response one is huge. So we talked about managing content across reviews, we talked about the concept of scale and building a content system. Talk to us a little more about the authenticity of the responses. How did you maintain that quality and that brand voice while you were automating at scale? I know you touched upon IQ briefly, but tell us a little bit more about what are some of the kinds of things that you put into the IQ layer to make sure that your responses are feeling very human.

Neil Grasso: Yeah. I mean, I would even say, beyond feeling. And look, the brand portion of it and the guidelines are important to adhere to. I would also say in the prompting itself, one of the most important things about review response is not glazing over somebody's comment about one of the things that they did not have a great experience with or vice versa. I think I would almost say it's similar to if someone was a really good listener in a conversation. It would be like, hey, you brought up four points, I'm going to address all four. I might not have the solution to all of your problems in my response, but I'm going to address them individually and try to point you in the right direction and make a good faith effort in that regard. You know, I've said this, and we'll probably get a little bit into this later, but when it comes to remaining authentic in the era of AI content, the analogy I always give is that we have historically been mixed between seeking human content and seeking automated content. A lot of us on this call today will go and read the news. And there are a lot of people in those spaces that you want to hear their perspective, their work, their written materials, and you wanna read them. The reason for that is because people are people driven for a lot of cases. On the other hand, there are times when automated content has always been a solution. You think about when you go to get a restaurant reservation and you book something online — you'd rather get an automated message letting you know that your table is ready or it's not than wait for someone to manually do it just to tell you, hey, you missed your time or something like that. So I think that now it's just about building marketing content that meets folks where they are. When it comes to a case study build or something more creative and really insightful, they're going to want something where the human has had more intervention. But for something like this, especially with eighty five percent of the reviews being like, hey, we're just super happy with the product, they want that notification that someone has seen it and someone has responded to it, but they don't need a super in-depth response. So yeah, that's kind of the philosophy that we approach this with.

Steve Kearns: Right. And I mean, it's such a good point you make in terms of the — there's a lot of different things in marketing where time to market or speed to market is equally as important as sort of what a message says, if not sometimes even more important. Because I've been in a world where I've managed community management style replies, and you can sit on those for a couple of weeks if you have other fire drills that are coming up or other things that are deemed higher value for you to be spending your time in. But the opportunity cost of that is that your customer doesn't hear from you for a couple of weeks, and it might actually negatively impact their experience. So I think it's a really astute point on where actually AI generated content or automated workflows might be a better customer experience. And that doesn't blanket across everything, to your point, but it's a great analogy that quite honestly I hadn't really considered before we had this conversation. So talk to us a little bit about some of the results that you saw in terms of time savings. And then also, has that enabled you to free up your time to do other things? And if so, what are those other things looking like?

Neil Grasso: Yeah. I'm actually gonna build on this from a little bit of our last QA there, and then I'll build it into this one. So when it comes to this content, Grid is not replacing my judgment. It's executing on it, which is great. And to bring that into the context of what does that look like in terms of time saving — well, previously, I'd have to do all of these reviews by hand, respond to all of them manually. And like you said, if I've got other bigger fish to fry, maybe somebody doesn't hear back from me for a week or a month or so, which is not ideal. But out of a recent round of thirty reviews, I was able to basically push, I think it was twenty five of them, without really having to do any editing at all because everything just worked. I think maybe I had to tweak like one word, but nothing crazy. And then four needed light editing where I actually had to get in. Something might have been a little bit not clear, but so I made a little bit of a light edit. And only one really needed extensive editing, but it wasn't because of anything Jasper did. It was because of a low rating with very specific complaints with context that Jasper wasn't going to be able to track. So in terms of those results, what we were able to previously get done in a week I was able to get review response for an entire month done in one business day. Like, that would have previously taken me a week plus, and now we've shrunk that timeline considerably. To speak to what that has allowed me to do now — I think that for a lot of marketers, AI poses a concern and a threat. And I understand why that might be, but for the first time in my marketing career, for a lot of marketers, we've always heard the phrase do more with less as budgets get cut or other economic factors play in. AI and tools like Jasper are the first thing where it's now like you can do more with more, because the more — I'll never say that it's a one to one replacement for headcount. I wouldn't say that it's there yet. But what I would say is that it gives the single marketer the opportunity to have outputs at scale with like a team of ten. So that team of ten should be thinking, well, can we be at the scale of a team of a hundred? Like, that's how I like to think about it because you have so much more of an opportunity to do that. And so now what we're able to do is, like, once we're able to automate a chunk of this workflow and we're still focusing manually on some of the more specific fringe cases, I start to think to myself, well, now can I spend more time within the whole review management bucket? Can I generate more reviews? So we have maybe there's going to be some more manual work there, but there's also going to be a lot more automated, and we're going to have a much bigger footprint on these sites. Or do I focus more of my attention towards generating far more case studies at a faster clip, and then spend my time within that process getting really strategic and emphasizing certain things in the sequence so that sales can be enabled to sell those products and focus on those use cases. So yeah, it allows you to really expand your depth and your strategic workflows that are across all of your programs instead of these one offs. It's like, no, now we built a case study engine that allows us to be so much more beneficial and efficient internally. So that's how I'd like to think about it.

Steve Kearns: Yeah. I think it's, you know, illuminating to hear you talk about it through that framing because that is — we find that our most successful customers are starting to think about their ability to scale their own impact in that kind of way. And there's always been, since the first day I started working in marketing — and this is ten, fifteen years prior to AI really ever coming on the scene from a consumer tech basis — there was never enough time in the day to do the things I wanted to do, and there were always things that were kind of an opportunity cost of, hey, I'm gonna invest in this, which means I'm not gonna get a chance to do that. So the more ways that we can find to unlock doing what we do best, which is getting in touch with our customers — if you're in a customer marketing capacity like the two of us are — I think the more impact we could have, which is great.

Neil Grasso: Hundred percent.

Next Use Cases and Transition to Jasper Grid

Steve Kearns: So with that, maybe the final question I'll leave you with before I pass back is, are you starting to pilot your next Grid use case? Or what do you think about using Jasper for in the coming weeks and months, now that you've kinda tackled this reviews workflow?

Neil Grasso: Yeah. I'd say two things that come to mind. One is we're looking to expand which networks we have this set up for. So right now, it's just set up for G2, but TrustRadius and Capterra are where we're looking to grow next so that, hopefully, where I'd like to be in a couple of months is all of our reviews that fall into bucket A — that take up again eighty to eighty five percent of them — those are all automated for the most part, and I can focus on the ones that are kind of those more critical complaints across all of those different networks. I think I'd love to be there in a few months. And then to kind of practice what I preach, talking about improving the systems themselves, I then like to take a look at what we're doing within each of these review flows, trying to see if we can make any improvements. And one example would be right now, we take all the reviews that Jasper provides, and we are copying and pasting them because there are some limitations on the review vendor end, like on those sites, there's not a lot of ways to automatically upload them. But I've been working with the solutions architecture team, and they've suggested there might be some workarounds for this so that once you feel good about the edits that you've made, you can push them live right from Jasper. It's a future state thing. It probably won't happen as quickly as we can expand to the other networks, but I think that it's a great way to then show my internal leadership — hey, this is a problem that we were not solving very well at all, and now we're solving at a hundred percent rate of response. And look how quickly we're able to do all of it thanks to a tool and a team like Jasper.

Steve Kearns: Yeah. No. That's great. And I think that's where a lot of our customers are headed now — how do you automate the full system once you're happy with what those outputs are gonna look like? So how are you then pushing to the system that you need to actually go ahead and publish in? It's one I've heard before in chatting with other customers, so I'm glad that that's something you're thinking about as well. So with that, I'm sure folks are gonna have questions for you later. Feel free to drop them in the chat. I am going to steal back the screen share here, if you don't mind. And if folks can see this, I'll go back into our slideshow mode. And I'm gonna welcome back Sara Mo to share a little bit more about Grid and maybe even zoom out a little bit. And the big headline here is that Grid is not just for the review use case we just talked about. There's a couple of different ways that you can use a tool like this. So with that, Sara Mo, I'll pass to you.

What Is Jasper Grid

Sara Mo: K. Thanks, Steve. And thanks for chatting with us, Neil. Y'all just saw a really amazing use case for Neil and Bonterra through their review responses. But I just wanna anchor us back in what actually is Jasper Grid, and what is the system that Neil is using to power these wins for his company? So this system is called Jasper Grid. That's what we built to solve the problems that we were talking about earlier in the session. And it's the system inside of Jasper that operationalizes marketing work. It's a no code spreadsheet like interface where you configure your workflows once, and then you run them at scale. You're not prompting over and over. You're building a repeatable system across campaigns, AEO, GEO discoverability, refresh workflows, personalized content, product description pipelines. It's all configured with your IQ, your agents, and your data. And Grid turns AI from a tool that you use into an infrastructure that your team runs on. So we're gonna go to the next slide. And as I just said, you can use Grid, Jasper Grid, for essentially anything. Honestly, Grid is incredibly flexible, so you can do much more than what's on the screen and what Neil just did. Across all of these use cases, the pattern's the same. You import your structured data. You connect and configure your IQ and your apps. You connect columns sequentially, and then you run the workflow. So it's not prompt engineering. It's workflow engineering. And I wanted to make sure that y'all saw yet another use case today. Neil's is great, but I wanna show just another one in case that one doesn't resonate with exactly your job. And so we're gonna do the hot topic of everything in AI and marketing right now, and we are gonna show you how AEO and GEO works in Jasper Grid.

Jasper Grid AI Visibility Demo

Narrator: Search has changed. AI now decides what gets surfaced, decided, recommended. Here's how Jasper uses agents to scale AI visibility in minutes. We'll start in the Jasper Grid, already loaded with our content. Each row represents an existing blog post. Before we optimize for AI visibility, humans define the guardrails. Then Jasper's agents get to work. First, we add brand intelligence — brand voice, audience, and product information. That becomes the guardrails for Jasper's agents, shaping how the content sounds and what products to reference. It's not buried in a prompt. It's explicit, reusable, and governed across every piece of content. Next, begin orchestrating agents, starting with the rewriter. The best part — every agent is automatically grounded in your brand voice, so it already sounds like your team wrote it. And because it's guided by your guardrails, audience aware and product informed, the content is both strategic and accurate. With Jasper's agents, you don't need to be a prompt engineer or set up complex rules. Just add your brand intelligence and go. Behind the scenes, the agent restructures the content for modern discovery, all guided by the brand intelligence we attach. Let's validate quality on a single row before scaling. You can immediately see clearer structure, more answerable sections, and content that's still unmistakably on brand. Because the agent is grounded in your strategy and content rules, optimization is consistent, not dependent on who wrote the prompt. This is how SEO teams scale updates across hundreds of pages without manual rewrites. Once quality checks out, agents take over execution, refreshing the rest of your content in minutes, not months. Every post is restructured for modern discovery — clear answers, stronger semantic signals, AEO ready FAQs, GEO relevance, and updated metadata. The result: more answerable content, more citations, and greater share of voice across search and LLMs. Jasper isn't just helping you rewrite content. It's turning your existing library into an AI optimized growth engine.

Sara Mo: Okay, y'all. That was Jasper Grid for AI search visibility. Normally, we would do a live demo, but Steve and I are traveling for one of our academies, and we're both tuning in from a hotel room. And so we were not trusting the hotel Wi-Fi to get you that live demo. So hopefully you enjoyed the prerecorded one. But truthfully, any use case for marketing can be built in Grid. We have the foundations. We have courses that you can take if you wanna build it yourself, and we also have partners here at Jasper to help you build, if you want someone else to build it for you. So I'm gonna hand it back over to Steve, and I think we're gonna transition to Q and A.

Q&A: Getting Started with AI Content Systems

Steve Kearns: Awesome. So I got a few good questions that came through in the chat, so we'll be able to answer those one by one. So Neil, I will direct this first question to you. Where should marketers first look if they want to identify how to use AI to create a scaled content system? Like, what kind of steps should they take? Should they be thinking under a specific category of work or looking under some kind of rock to get started?

Neil Grasso: Yeah. To use your analogy, I'm going to start by actually saying the rocks that they should look under first, and that would be — I would probably stray away from doing a lot of comparison on LinkedIn or other thought leadership platforms. Not not this one. This one you should listen to. But the reason I say that is because I think for a lot of people, they look to either individual contributors or companies on LinkedIn that walk through a step by step of a situation that they don't have within their own workflows. So I'd say look internally to your teams. And I think the question you wanna ask yourself — at least this is how I did it — was, you know, what's an area that is still a high value output but is pretty cyclical and is one that we can have as a starter project that can establish a layer of trust so that once we see those results, then we can start experimenting with this in some of our more critical and highly sensitive workloads. I wouldn't recommend, for example, starting off by automating some of your most important brand or thought leadership content because if you push that live without reviewing it, there probably will be some consequences. But at the same time, I think that if you wanted to take a look at whether it be review response or maybe something that's a long form asset that you want to use Jasper to get some efficiency gains, I would say, for one, just look to see how much time is it taking. Two, what are areas that you think that by feeding Jasper with all of the tools that we've talked about today — whether it be using the IQ layer or using Grid — how can we speed up and get reliable outputs, and then focus a lot of our time on fringe cases and review response or sequencing in a long form asset. And then once you really optimize that process, you can start to basically do that for the next process you see. Take an ad hoc example of requests you get, stretch it out, take a look at all of the most painful parts of it now, see where Jasper or any other AI tool can help, and just improve your processes, and then increase the bandwidth of the program so that you can do more and deliver more impact.

Steve Kearns: Yeah. I think that's a — thank you for breaking it down kind of into those steps. I think it's a great call out to look for a workflow that has a high value output, but maybe the human marketer's input or ability to execute that at scale maybe feels diminished or not a great place to spend time. So I'm sure everyone can kinda think of some kind of workflow or a task that kinda eats away at them that is important, butmaybe not the most important, super time consuming, and maybe not the most creative. So I love that. Question from Michelle, and I think I'll actually pass this one to you, Sara Mo. What's the best way to get started with Grid and identifying use cases if your team is already using Jasper?

Sara Mo: Great question, Michelle. So I think this goes back to something that Neil was saying essentially about understanding what is a high value output that tends to be repetitive. So Jasper Grid is great at scaling. It's great at understanding logic and building a system that essentially is sequential, between the IQ, the information, all of the inputs to get you outputs that are kind of normalized to your expectations. So I would say if you're already using Jasper Grid, what are things that you're doing over and over again? What is something that you wish that you could automate? A few use cases that we see are personalized emails — personalization is great to do in Jasper. Product descriptions, another one that we see a lot of success within Jasper. And then any sort of web content that's driving SEO, AEO, GEO, whether you're doing a web refresh to drive you to change your current content to be optimized for this new world of AI visibility or if you're trying to spot areas where you could build new content. If you're a product marketer and you're doing a campaign launch, that's something that we use Jasper for all the time here at Jasper and we see our customers do a lot. If you're a PMM who's doing competitive work, building a competitive grid that allows you to create battle cards and sound bites for your sales team. So you can kind of do anything in Grid. It just has to be something that is repeatable and that you have an expectation for output.

Q&A: Subject Matter Expertise and Proprietary Content

Steve Kearns: Yeah. Thanks, Sara Mo. So Neil, this question is from Brian. How do you address the need for content to feel not just proprietary, but that requires a lot of subject matter expertise? So such as with enterprise B2B.

Neil Grasso: Yeah. I mean, it's an excellent question. It's one that I try to answer every day. So Brian, it's a great ask. What I would say is this is another area where I highlight this is the role of the modern marketer in the AI era — your high value subject expertise and your ability to ask questions to your customers or your prospects or internally to find answers to questions where the data doesn't live somewhere that AI is currently scraping. That's a huge opportunity, not just in marketing, but across the board. And I think that when you get those answers and you do that research — I was an ex journalist. When I put on my investigative journalist hat and I go and do that work, I feel really fulfilled in doing it, to be honest. And also, it's what will improve the outputs that you're getting out of whether it be a vertical LLM like ChatGPT or Claude or, in this case, Jasper.

Q&A: Integrations and AI Visibility

Steve Kearns: Amazing. I think that's a great segue into our next question to kinda open up the hood a little bit of how the tech works. So we got a question from Jill, and Jill says — and I think, Sara Mo, this one will be for you. We use SEMrush and Taku, and I actually haven't heard of Taku, so maybe I need to look that up afterward. But we hop between the two for AEO, GEO, and SEO as well as meta upkeep, and we hop back to implement, but going between the two is exhausting. So is Jasper a combination of tools like this? Sara Mo, your thoughts there.

Sara Mo: Yeah. Great question. So SEMrush is accessible in Jasper. You can actually just log in to your SEMrush account through Jasper, and so you can bring all of that insights and data to your grid already. Highly recommend doing that. It's also accessible in our optimization agent in Canvas and in Grid. So highly recommend using the same SEMrush integration that's already in Jasper. Now, I'm not sure exactly what Taku does, but I do wanna talk about something that is on the road map for Jasper. This whole Grid, AEO, GEO world — we are not just stopping here. What we have launched in the last call it month now is we have three different templates that are designed and customized with fourteen of our agents to get you the optimization, originating new content, and outranking your competitors' content. So those systems are already existing in Jasper. That is how you actually action to get the results that you're looking for. Now what's coming, and it's already in beta so some of your friends that are customers might already have access, and it should be live within the next month for everyone, is our AI Visibility Hub. And what that is going to do is surface insights across what your visibility is actually in the LLMs themselves. So across ChatGPT, Claude, Perplexity, Google Gemini — they're all going to be shown in Jasper, and that data is then going to be able to be immediately executed on in Jasper. We're gonna have this new GEO agent that's essentially a role based agent. So it's super powerful to be able to make strategic recommendations and then execute without you having to do a lot of work or any of the setup in greater than any of the other systems. So Jasper's really becoming the key place to start your AEO and GEO efforts. Like I said earlier, it is the hottest topic in marketing right now, and Jasper's making that easy for you. And the thing that is making Jasper different in our approach to that is we are making sure that we are influencing what the LLMs' perception of your brand is. Our job is to not just create a bunch of content and keep racking up that AI slop like we've seen a lot of folks doing lately. Our job is to make your brand consistent across your reviews, across your website, across your emails, across any piece of content that is out there, because that's truly what's going to drive the LLMs' perception of your brand into the future and essentially retain that position over time. So Steve, I know that was a long answer, but hopefully that was some good information for the folks here.

Q&A: Translation and AI Data Sourcing

Steve Kearns: No. No. No. That's great. And I've got one more question for both of you, and then we can go ahead and wrap up. So Sara Mo, this one is also from Julia. I need to produce content in English and French with Jasper Grid — does Jasper allow you to have a French translation column? And how accurate is translation? I know in my experience using Copilot it doesn't always produce accurate translation. And if you could also talk a little bit about how and where we source our AI visibility data from. And then Neil, I'll close it out with you one final question, and then we'll thank folks for their time.

Sara Mo: Okay. I'll try to be quick here. Yes. You can absolutely translate to French in Jasper. Our translation agent is competitive with the best translation agents on the market through DeepL and Google, etcetera. So yes, you can translate to French in Jasper, and you can trust it. Next question was our AI search visibility data. So this is something that is actually pretty commoditized across the market, and anyone can access it via APIs. So we essentially have done that and brought the answers to you. There are some companies out there who get their data in a little bit sketchier ways by having Chrome plugins that scrape your entire search data on your screen, and we are not doing that version. We are doing the safe, publicly available APIs version. So that will be available in Jasper here soon.

Q&A: Final Tip and Closing

Steve Kearns: Awesome. Well, thank you, Sara Mo. I think folks are walking, hopefully, with sharp knowledge on how they can consider using Jasper. And Neil, I'll close out with one final question for you for those who are tuning in. What is one tip you have for marketers here as they start to think about scaled content systems? Maybe within the context of Jasper, what was a really big unlock for you that you'd recommend folks take a look at?

Neil Grasso: It's a good question. I'd say it kind of returns to some of the things we talked about before where it's like being able to identify where you think a system like this will produce the most value. And I think being able to tell the story behind it as well. If you are going to do any sort of automation in marketing and you pay very little mind to the output and immediately show someone in leadership at your company that you're able to automate this, but the output is really poor or the response is poor or the performance is poor, I don't think you've done your job as a marketer. I think that you need to take a look at — okay, let's evaluate the process, not just try to push for the outputs. And when you do that, I think it just changes the game entirely. And a tool like Jasper is built far better for that than a ChatGPT or a Claude where a lot of those are simple input, simple response, simple output, and that's great. That's fine for one off, ad hoc requests. But I think that when you're trying to drive real internal efficiency and still be able to achieve great output, you have to look at your process, and you have to be able to say, hey, where are we really going to make improvements? And then that's what's going to lead you to be able to go to your leadership team and say, hey, look, we've improved from A to Z. It's like a completely different situation. That's what we did with review response. I was able to show my leadership team, hey, we went from some responses here and there to one hundred percent. That's a big number that jumps off the page. And when they saw the content, they were happy with it as well.

Steve Kearns: Yeah. Well Neil, I think that is a perfect way to wrap our conversation. We really appreciate your time. Hopefully folks on this webinar found your use case inspirational and valuable. Even just beyond the work that you're doing with Jasper and with Grid, I think these are all great tips that other marketers should be hearing as we think about how we navigate kind of this new frontier with scaled AI tech. So we appreciate your time. For those of you who are joining us, we appreciate your time. And with that, we'll follow up with a couple of different resources, namely a link to our Grid course. If you all wanna get certified, you can become certified content engineers like the three folks on this call — myself and Sara Mo are instructors. So you'll hear more and more from us. We'll also send you a link to the customer story that we produced with Neil, so you can read a little bit more about his story and best practices. So with that, thanks again for joining us, and we'll see you soon. Take care.