Inside Rival: How We Use What We Build
This blog is part of Inside Rival, a series where our team shares how we use Rival in our everyday work. From favorite tools to AI agents and automations, we're showing you what we're building, how we're using it and what we've learned along the way.
I'm a creature of habit, especially when I've spent a considerable amount of time figuring out a process that works. So when it came to building HTML emails, I had my system down. Was it the most efficient? Probably not. Did it involve bouncing between ChatGPT, Claude, and more browser tabs than I'd like to admit? Absolutely. But it worked, and I'd gotten pretty good at it.
I'd use ChatGPT for research, copywriting, and figuring out how to structure an email. Then I'd move over to Claude to work through the design and HTML. If something wasn't quite right, I'd go back to ChatGPT, make adjustments, and return to Claude to try again. A little copy here, some design there, a few rounds of troubleshooting, and eventually I'd have a finished email.
My own little Frankenstein operation.
And honestly, I was fine with it. The last thing I wanted was another platform to learn, another process to figure out, or another AI solution promising to make my life easier. I already had a system that got the job done, and with a lean marketing team and a never-ending list of priorities, switching things up wasn't exactly at the top of my list.
But then our team built an HTML email agent in Rival, and my carefully constructed process was about to get a serious reality check.
And the best part? We're working toward making it available to everyone. Stay tuned.
If It Isn't Broken, Why Am I Fixing It?
Here's the thing about working in marketing with a lean team. You get pretty resourceful. I'm not an HTML designer, and I certainly don't know how to code, but I know how I want an email to look, what I want it to say, and how it needs to represent the brand. So I figured out how to make it happen with the tools I had.
And to be fair, ChatGPT and Claude were incredibly helpful. Without them, I probably wouldn't have been building HTML emails myself in the first place. But looking back, I was doing a lot more work than I realized.
Every email involved multiple steps, multiple platforms, and a fair amount of copying and pasting. I'd get the messaging right in one place, move to another for the design, and then spend time making sure everything actually worked together. Sometimes the copy looked great, but the layout needed work. Other times the design was exactly what I wanted, but one small adjustment turned into another 20 minutes of prompting.
Still, I knew the process, and I knew how to get the result I wanted. The idea of starting over with something new sounded more exhausting than sticking with what I already knew. Which, looking back, is probably how a lot of inefficient processes survive. They work just well enough that we stop questioning them.
Yes, I Work at Rival. No, I Didn't Try It First.
When our team built an agent in Rival specifically for creating HTML emails, you'd think I would have been first in line to try it. But honestly, I was hesitant. Not because I doubted the technology, but because I'd already gotten comfortable with my laborious process. I knew how to get the results I wanted, and learning something new wasn't exactly a priority.
Which is especially funny considering I was one of the people pushing our team to build this agent in the first place. Apparently, complaining about my duct-taped workflow was easier than actually letting it go.
But that's when things changed.
The agent handled everything from start to finish. Research, copywriting, design, and HTML, all in one place. No switching between platforms or piecing together outputs. What used to take me a considerable amount of time was done in a fraction of it, and the quality of the finished email genuinely surprised me.
One of my favorite things about working at Rival is getting to experience what we're building firsthand. And this was one of those moments where I finally understood the difference between using AI to help with a process and using AI to actually execute it.
My Frankenstein approach relied on me to connect all the pieces. ChatGPT could help with the research and copy, Claude could help with the design and HTML, but I was still the one managing every step, moving information between platforms, and troubleshooting when something didn't work.
Rival takes a different approach. By combining AI with deterministic functions, the agent can handle specific tasks within a more structured workflow. Instead of relying on me to repeatedly prompt, copy, paste, and piece everything together, the process happens in one place. I still have creative control, but I'm no longer responsible for holding the entire workflow together.
I spend a lot of time thinking about how to position Rival and communicate its value. But experiences like this make that part of my job easier. I'm not just talking about what our technology could do. I'm seeing firsthand how a different approach to AI can change the way we work.
Maybe the Best AI Workflow Is the One You Don't Have to Manage.
Looking back, I'd gotten really good at using AI tools, but I was still doing a lot of the heavy lifting to bring everything together. I hadn't realized how much time I was spending managing the process until I didn't have to anymore.
I'm still making the creative decisions, reviewing the copy, and making sure everything represents our brand. I'm just spending a lot less time getting there. And with everything else on my plate, that's a pretty big win.
So I think it's officially time to retire my Frankenstein approach to HTML emails. It served me well, and we had some good times.
About the Author
Lauren is the Manager of Brand & Strategic Marketing at Rival, where she focuses on brand positioning, go-to-market strategy, and connecting emerging AI technology to real business challenges. Working alongside Rival's developers gives her a unique perspective on both the possibilities of AI and the challenges of making it accessible. She brings a strategic marketer's perspective to an industry often dominated by technical conversations, with a focus on making AI practical, approachable, and genuinely useful.