The Future of Content Pipelines: Scaling with Quality, Resonance, and Performance

Introduction and Market Overview

Loreal Lynch: Great. All right, we're going to jump into our first session where we're going to be talking about the future of content pipelines. We'll talk about what we mean by content pipelines and really want to focus on how to scale with quality, resonance and performance.

But before we jump into content and what we're building at Jasper, I just want to take a really high level view and look at the market and I'll start with showing this slide and I'm going to start with a line that you may have heard before, which is, in these unprecedented times, I don't know if anybody's ever seen this chart. It's kind of wild. You can see there's a direct Correlation between when ChatGPT hit the market and what's happened since then with Jobs and then what's happened since then with the S&P 500. It's wild the impact that AI is having on the overall economy. It's kind of crazy, this world that we're living in.

And so what does that mean for all of us as marketers and as business people? Here's something else that's interesting. Models can increasingly handle high knowledge work. AI models are exponential. Every new model expands what's possible. It's estimated that by the year 2030, 50% of marketing workloads are going to be able to be done by agents. Not only that, but recently there was this analysis done where they looked at industry expertise, actual industry human experts, and compared that with model performance and they found that models are nearly at par with industry experts.

So it's crazy. What does this all mean? What does this mean for marketing?

The Great Marketing Shift

Loreal Lynch: We're experiencing what we're calling the great shift. The great shift in marketing. And not even just in marketing, it's across all business. Some of it is. It's actually not bad. It's welcome change. I think for a lot of us, the first one is our processes are moving from being manual to being intelligent and automated. That's a very welcome shift. There was a survey recently done by Asana that said that 63% of a marketer's time is spent on routine, low leverage work. If you think about going and pulling a report every week and then writing an executive summary to send that out to your team. If you think about going and resizing 15 banner ads for the different formats that need to be done, this is rote work. This is things that humans don't need to be doing. AI agents should be able to automate that away.

So the move to automation from manual is Definitely a welcome change. The second thing that I think is also a welcome change in my opinion is the move, the collapse of the Martech stack, the collapse of the enterprise stack. There's all these SaaS based tools that are kind of having a little bit of a reckoning right now. If you think about what's happening in the market as more people adopt agents and they build their own agentix systems through APIs and MCP servers, there's all this automation happening. It's kind of like I was at this investor event recently and One of the VCs said all these big SaaS platforms, they have to think about if what they've built, is it a strategic benefit or is it a liability?

Because you really have to think about how these companies are going to evolve in the agentic era as people are building their own systems. And so there's a lot of composability happening. We're doing it ourselves, within Jasper, within our team, about what we can build, what we can automate.

And everybody wants custom, everybody's workflows are custom, so it needs to be composable. So the whole stack is in the middle of a massive shift. And then the last thing that's changing is the very nature of our roles and our identities as marketers. We've always been doers, we've always been the ones rolling up our sleeves, doing the work, planning the thing, writing the thing.

And now agents can do a lot of that for us. And so our roles change, we actually get elevated into more strategic roles. We're no longer the doers, now we're the orchestrators. We have team members that have been ICs individual contributors that now are managing several agents.

And so what does the skill set need to look like as we bring on these new types of employees? They need to be more strategic, need to be thinking at a higher level, need to be systems thinkers, thinking about orchestrating across systems. And along those lines, our prediction is not that marketing jobs are going to go away, but that the nature of the roles are going to shift into more technical roles.

And I don't mean technical in terms of writing code, I mean technical again in terms of being systems thinkers, being able to think not just about writing a piece of content, but how do you create a system that allows multiple people on the team to write multiple pieces of content? This is something that I encountered recently. I actually had a creative person on my team, a creative director who said, hey, we need to hire a copywriter.

Because even though we use Jasper, we still need kind of that point person that's in charge of managing the Jasper inputs and helping us be able to scale. And I was a little mixed on hiring a copywriter because obviously the nature of what we do it would be what would that say to the market if an AI content company is hiring a copywriter? But I agreed because we needed to. We have so many launches, product velocity, we have multiple campaigns. We're trying to move towards more outbound and account based personalization.

So as I interviewed these candidates, I would ask them, tell me about your process for using AI. How do you use AI in your own workflow? And a lot of them said, well, I use it to ideate or I use it to overcome blank page syndrome.

And I thought that was good. But what I was really looking for, the candidates that really impressed me were the ones again that were those systems thinkers. I've created a prompt library so that I could give this to my team so that when they write, they sound like when I write.

And how do you scale yourself as a copywriter because the craft is still so important so that you don't end up with this AI slop. But again, how do you just think a little bit differently Instead of creating a piece, create systems that create pieces. And that's really, I think, the shift that we're seeing and what I predict is going to happen within marketing organizations again by the year 2030, is that we're going to see a lot of agents and we're just going to see this shift happening across marketing roles and skills.

Succeeding in the New Content Landscape

Loreal Lynch: And so in this new world, how do you really succeed? And not just succeed, but stay ahead? How do you build a moat? How do you remain differentiated? There's kind of three things that we've seen successful customers doing. The first is actually being able to activate that data and that AI intelligence that is really the critical component of personalization optimization. Being able to reach the right, the right audience at the right time. You've got to be able to harness that data and that intelligence. The second piece is getting your message everywhere. It's very competitive right now. There are so many channels, so many places where your customers are looking, searching, discovering.

And so how do you distribute your message everywhere across channels, formats, regions? And the third thing is as AI can level the playing field when it comes to campaign creation, anybody can create unlimited campaigns at scale, scale now in theory, with AI. And so I think that makes brand increasingly important. How are you really going to differentiate? It's through brand, it's through trust, it's through creating meaningful connections with your audience building loyalty.

And our point of view is that what makes all three of these possible, the foundation for that is content. Content is relevant for all of these. And that's why we're all here, and that's why we're passionate about what we're building at Jasper. We really see content as not just part of the marketing transformation, it is the marketing transformation that needs to happen.

And this slide illustrates why. All of you are well aware that you have one campaign concept, but that does not mean one piece of content. One concept can easily, and we see this with customers, can easily proliferate into over a thousand pieces of content. Once you take multiple variants, multiple channels, multiple audiences, multiple brand voices for if you have sub brands within your portfolio.

And this is very hairy and very messy. And it's interesting because when people first get started with AI, they think and you see all these studies and it says, what are you using AI for in your organization? And people say, I'm using it for content, but what they mean is, I'm using it to write one piece of content. An individual might be using it to write an email or to write a blog post, but nobody can do this. This is still a big challenge that people haven't figured out.

And this is what we're laser focused on solving this at Jasper.

The Challenge of Scaling Content

Loreal Lynch: Because when you scale content, and if there's one thing I've learned about kind of seeing this change that's happened over the last year or2 as AI tools become more accessible, it's that creating content is easy. Creating really shitty content is easy, to be honest, but creating good content at scale is damn near impossible. It's really, really hard.

And so we joke. We have this triangle we put together. Most marketers are lucky if you can do one of these. Quality, speed or scale, maybe two, it's really hard to do all three. That's really what we're building at Jasper is we're focusing on being able to create a system that allows you to generate AI powered content pipelines so you can create content that delivers it's high quality, your team can work faster to deliver it, and ultimately can scale across audiences, regions, brands.

And so that's what we're here to talk about today and to talk about how we're doing that and what we're delivering to you. We have a couple of exciting launches today and I want to bring Brian Sao, our Chief Product Officer, onto the stage to walk us through the platform a little bit.

Jasper Platform Overview

Bryan Tsao: All right, thanks, Loreal. So, hey, everyone, my name is Brian. I'M our Chief Product Officer here at Jasper and I'm really excited to share with you a little bit of what we're building to help you all make high quality marketing content very fast and at a high scale.

So here's how we do it. I want to start by talking about our LLM optimized foundation.

LLM-Optimized Foundation

Bryan Tsao: This is the foundation of our platform, how Jasper sits on top of ChatGPT or Gemini or Claude to generate the marketing content that you all get when you use Jasper. Over the last year, we've invested a lot into the tooling that we use to test the quality of the different models. Most people don't know this, but at this point our team is constantly running split tests against cloud, pitting it against ChatGPT, pitting it against Gemini to find out for every type of marketing content, whether that's long form content or short form content, like an Instagram caption, or the suggestions that are AI powered inside of our platform, trying to at all times figure out what is the best LLM for each specific marketing task. So, for example, oh, and we're constantly testing our user responses to those pieces of content so we know if a user copies a piece of content or uses it as context in a future generation or gives it a thumbs up.

So we're really always measuring what is the best model for every given marketing task. For example, right now we're actually using in production Cloudsonnet 4 to generate a lot of our long form content. We found that it's the best at writing something that has complex ideas, maybe a few paragraphs and it's been working really well for us. Right now we're testing a lot of the next generation or the latest updates of models to see what might be better. We're testing Gemini 2.5 Pro. We are testing clots on it 4.5 and we're testing those two against our current generation best model.

And what's really interesting is we're also finding there's not just one winner right now. Gemini 2.5 Pro has actually pulled ahead over the last month in generating content quality or high quality content. Except for in one case where we know that Claude 4.5 sonnet is better at generating content when you attach knowledge inside of the Jasper platform.

So we're constantly testing all of these different little detailed aspects of which model provides the best content for any given use case. And the thing is, what we're finding is that month over month changes are happening much more rapidly. A couple months ago, in August, for example, at that point we were probably doing about 85% of our content generations. With ChatGPT, we were using 4.0. That was the best of the time. Today we're only sending 17% of our content generations to ChatGPT because we've seen Claude and Gemini pull ahead. Our current One is Claude 4.0 sonnet, but if the results of our current split test hold up, we're going to be changing a lot of Those to Gemini 2.5 Pro. What we found is that the models are getting better and changing so fast so frequently that the best model for any given job is going to be different from week to week.

And one of the things that we really want to provide as the core foundation of our platform is making sure that you're using the best without having to change vendors every time there's a shift in the market.

Jasper IQ and Integration

Bryan Tsao: On top of the LLMs, I'm sure you've heard us talk about the IQ layer where you can configure your brand voice, configure your audience segments, upload your company knowledge. We think that what distinguishes high quality marketing content from, from just any content that you could get from any LLM is that marketing content is about your company, it is targeted towards an audience and it is in your brand voice. We've invested a lot into making our IQ layer the best place to configure and manage those and to optimize those things for LLM for AI generations of marketing content. We also recognize that with the explosion of AI tools, Jasper is not going to be the only place where marketing content is created. Yesterday morning we had a bunch of our CMOs SVPs of marketing for our customer advisory board right here at this hotel.

And the biggest theme that they talked to us about was Jasper's really awesome. But hey, our company's rolling out Copilot or chatgpt and suddenly everybody's a marketing content creator. Everybody is writing an email in chatgpt and all these companies are running, really celebrating, really giving a lot of pressure to people like, hey, use AI more, use AI more, use AI more.

So now everybody is just using these tools wherever they can to do whatever they want, right? So with the recent launch of our MCP server, a big part of our vision is we want to bring Jasper IQ tools to everywhere that content generation happens for marketing. Because l' Oreal talked about speed, scale and quality. To me, quality means consistency and output, no matter who generates it, no matter where it is generated.

Because yeah, somebody goes into ChatGPT, generates an email, sends that email out to a customer or a thousand customers or 10,000 customers. That is your marketing output, right? So we're really excited that with our Jasper MCP server, you can now integrate Jasper's IQ and any custom apps you have into any MCP compatible service.

So that could be ChatGPT, Copilot is MCP compatible. And as more and more marketing tools come onto the standard, that really brings the reach of Jasper out to where all the people on all of your teams are working.

Studio and Customization

Bryan Tsao: Next, I want to talk about some updates we're making to Studio. So Jasper's apps, which I'm sure many of you use already do, I think a pretty good job out of the box at helping you create marketing content. But I also know that I'm sure every single one of you, every one of your marketing teams has your own way of how you want to write a blog post, how you want to write an email, how you do a press release, or any given piece of marketing content. I'm sure the whole point of it is not that you are all doing the same kind of blog post. Studio solves that problem by making it really easy for any of you to customize any of our existing apps with your own instructions, with your own output examples so that the blog posts or the marketing content that you get out of Jasper sounds like you. Over the next few months, we're going to be making this system even better. We're going to be adding tools like web scrape and web search.

So when you're using these apps, you have the power of being able to go out into the Internet and do research and add information that you didn't enter in right out of the box. We're also going to be releasing the option for you to use Jasper Tune system prompts with any of the apps that you create to generate much higher quality output. So all the things that we've learned about how to prompt an LLM to do high quality content, we're going to release to you all as something you can add to your own apps to make sure that the custom apps that you build on your own with Studio are of the highest quality. Lastly, we're going to be releasing access to Jasper Studio created apps within our API and MCP Server so that as with iq, no matter where people on your marketing team or people within your company are generating marketing content, they can use these highly customized apps that you've invested in. Right. Again, that's the quality. Quality is consistency, no matter where something is made.

And MCP in our API is how we're bringing that to consistency across the entire marketing stack.

Jasper Agents

Bryan Tsao: We're also going to be gaing three of our agents, starting with the optimization agent. So the optimization agent is about SEO, AEO and GEO right now. It's going to come with a great integration that we have into Semrush.

So any of you who are Semrush customers, you can oauth in directly to your Semrush account and be able to use the optimization agent to do keyword research. Apply that keyword research to creating SEO optimized website content. You can do competitor keyword analysis. We have SEO scoring all these things coming straight out of the box with the optimization agent. We've also built our own in house AI search visibility capability.

So you're going to be able to see for any given term or any given piece of marketing content how visible that would be in a chatgpt or in a cloud, or in a perplexity. So really excited to launch this agent that can really help you all write content that boost visibility into your company, into your brand, into your products with SEO and the LLMs. The next agent we're going to launch is the personalization agent.

And this is really about connecting into Salesforce and also in the future, other marketing automation platforms like your HubSpot or your MarketOS and using bringing in customer data from those platforms into Jasper so that you can make personalized emails or landing pages targeted towards individuals right inside of Jasper. This is actually something that l' Oreal's team and my team work together on inside of Jasper to do our own personalized content. So I'm sure many of you have received emails from us over the past few months, especially maybe before you converted to being a customer. We're now at a point where we're leveraging Jasper to personalize almost all emails to inbound leads, with custom email copy for every single person based on research that the agent does on their company and also a custom landing page to try to drive higher signups, higher demos. I'm sure that's something that many of you are on the hook for in your marketing teams.

So really excited to be bringing the personalization agent to the market. And then lastly, we have the research agent. This is actually our most commonly used agent, our most popular one, and we find that a lot of great campaigns start with research and we want a way to bring that research directly into Jasper, into the canvas.

So when you're coming up with that campaign brief, you don't have to go back and forth between ChatGPT and any other tools. It's bringing it right into the Jasper workspace. So those are our three agents which sit on top of the canvas which we released over the summer.

Introducing Jasper Grid

Bryan Tsao: I'm now really excited to announce the next big thing we're going to be launching. So that's the Jasper Grid. Over the summer, we launched Jasper Canvas to help with speed. We think that that's the best tool for those of you who are launching campaigns. I see a lot of service titan folks in the crowd. I know, for example, you all are doing a big launch in Australia. You have a bunch of campaigns that you're doing international expansion.

And we think that the canvas is the best way to launch campaigns at high velocity for us. The grid is about scale, right? It's going to be able to take the content that you need to execute those campaigns, right?

So you have a concept, you need different types of content. You need emails, press releases, website content, social media, whatever it is. And now maybe you need that to target different audiences or different languages. Right? I've talked to a number of folks in the crowd about this. Definitely Vinny, definitely Nick from Dell.

And the grid is the tool that we've built to scale these campaigns out across all of the different audiences and variations. Doing assets by not 10 at a time, not 50 at a time, but 100 at a time or 1,000 at a time, because that's the amount of content that you actually need to launch these campaigns. So later, I'm actually really excited. Daniel, the lead PM on this project is going to give you an in depth demo of the feature.

But maybe just as a little tip, content operations could move as fast as your ideas. What if scale wasn't the problem but the Unlock Jasper Grid transforms how marketing teams produce content. It connects your data, strategy and creative into an intelligent automated system built to scale. Generate optimized blog posts for every engine, automatically written, localized and ready to publish. Deliver endless content variations for every audience segment without endless manual work. Plan. Once launch everywhere Grid keeps messaging consistent across ads, social and email. At the click of a button, instantly generate product descriptions and visuals that are perfectly aligned with your brand. Jasper Grid solves for scale, transforming how marketing teams work, create and grow. It's not more content, it's a faster, smarter way to make it. This is how AI content pipelines come to life. All right, so Jasper Grid. I don't know about you all, but one thing you will soon learn about me is that I'm a huge spreadsheet nerd, big Excel. If any of you actually see my laptop's right There. It's a Dell PC, so shout out, Dell. I'm a customer of yours as well. I've always been. I'm one of the two PC users in our engineering, product and design team at Jasper because I want to use Microsoft Excel.

So just even on a personal note, I got to say, I'm so excited to be building and launching this for all of you and I can't wait for you to see the longer demo and start using it. So if you have any feedback, anything, definitely find me. I could talk about it for days, but I can see there's only 12 minutes left in this, so I'm not going to keep going on that.

Putting It All Together and Closing

Bryan Tsao: We're going to wrap this up in the next couple slides, so let's put it all together. Right through Jasper, you are getting the best of ChatGPT, Claude and Gemini right out of the box. And as that change it going to change it under the hood for you. It's not something you need to worry about. If your company only uses Gemini or only uses Copilot, or only uses ChatGPT, and you're worried that you're not going to get the best marketing content, with Jasper, you don't have to worry. You're getting the best of everything straight out of the box. On top of that, we've invested in the IQ layer as the best place for you to create, manage, configure your brand voice, your audiences and your company information. Through Studio, Agents and Grid, we're building out the systems that you need to go beyond chat and truly scale your campaigns out. Right. Canvas, again, you can create those, like take a campaign brief, right? Create the initial versions of all of your different content types. That's about speed of launching campaigns.

And with Grid, you're going to be able to scale those campaigns out for every variation of content that you need for different audiences or different languages. And then lastly, we really believe with our MCP server, we're extending Jasper out to wherever the work happens. And that's quality. Talked a little bit again about how all the proliferation of different chatbots, different AI tools, we're taking Jasper to all of those tools wherever the work happens, through a combination of native integrations, custom integrations and our MCP server. We think the core again of Jasper is that iq, right? That's the thing that really defines your marketing content, right? What is the brand voice? What are those audience segments and how do you target them? What is the information about your company or your product or your service they are trying to bring to market and we really want to extend high quality generations again across your entire marketing staff.

So with that, it's a little bit of speed scale quality, you know, that we're really excited to work on and share with you all and. Yeah. All right. Thanks, everyone.