The exact AI playbook (MCPs, GPTs, Granola) that saved ElevenLabs $100k+ & helps them ship daily

The exact AI playbook (MCPs, GPTs, Granola) that saved ElevenLabs $100k+ & helps them ship daily

How I AIJun 2, 202544m

Luke Harries (guest), Claire Vo (host)

“Everything is a launch” growth philosophyGranola → custom GPT case-study pipelinePrompt engineering: roles, must-do rules, examplesRepurposing: blog → X thread → LinkedIn → founder voiceBuild vs buy for SaaS tooling; marketers using CursorPrompt-driven localization via GitHub Actions + CMS buttonMCPs: tool exposure, local context, agentic workflows

In this episode of How I AI, featuring Luke Harries and Claire Vo, The exact AI playbook (MCPs, GPTs, Granola) that saved ElevenLabs $100k+ & helps them ship daily explores elevenLabs’ AI marketing playbook: automated launches, localization savings, WhatsApp MCPs Luke Harries (Head of Growth at ElevenLabs) shares three practical AI workflows that make marketing execution faster, cheaper, and more systematic: automated case studies, in-house localization, and an MCP-based WhatsApp assistant.

ElevenLabs’ AI marketing playbook: automated launches, localization savings, WhatsApp MCPs

Luke Harries (Head of Growth at ElevenLabs) shares three practical AI workflows that make marketing execution faster, cheaper, and more systematic: automated case studies, in-house localization, and an MCP-based WhatsApp assistant.

He demonstrates a “case study in minutes” pipeline using Granola transcripts plus a custom ChatGPT “Copy Editor” GPT trained on ElevenLabs’ tone, formatting rules, and examples—then repurposes outputs into tweet threads and other launch assets.

He describes ripping out a $40k/year localization SaaS plus $100k+ in agency spend by building a lightweight translation server + GitHub Action with per-language prompts, keeping human review only for sensitive pages.

Finally, he explains MCPs (Model Context Protocol) and shows an unofficial WhatsApp MCP that downloads chats to a local SQLite DB so Claude can summarize, search, and message—hinting at a future personal assistant built on WhatsApp, email, and calendar.

Key Takeaways

Automate the launch checklist, not just the copy.

ElevenLabs treats each feature as a full launch with value props, messaging, audience, and distribution assets. ...

Get the full analysis with uListen AI

Use transcripts + summaries together to avoid “lossy” content.

Luke pastes both Granola’s structured summary and the raw transcript into the GPT. ...

Get the full analysis with uListen AI

The best editing happens in the prompt, not the output.

When outputs miss the mark (weak headings, not enough stats, wrong format), they update the underlying instructions so every future run improves. ...

Get the full analysis with uListen AI

A custom GPT is a scalable brand voice enforcement layer.

Their “ElevenLabs Copy Editor” encodes tone (serious, research-led), style rules (American English), and formatting requirements, backed by strong example posts/tweets. ...

Get the full analysis with uListen AI

Distribution is part of the asset—generate it as a system.

Luke pairs creation with a repeatable workflow (e. ...

Get the full analysis with uListen AI

Build vs buy can win when SaaS blocks prompt control.

They replaced a localization vendor largely because it didn’t allow editing the AI prompt, making quality impossible to tune. ...

Get the full analysis with uListen AI

MCPs hint at flexible, chat-native workflows beyond rigid automations.

Luke contrasts static Zapier-style flows with chat-driven tool use: when the task changes (e. ...

Get the full analysis with uListen AI

Notable Quotes

When you're editing, as much as possible, try and edit the underlying prompt rather than the actual output.

Luke Harries

This saved us $40,000 a year for the tool, so immediately canceled it. Over $100,000 in agency costs.

Luke Harries

If we're just using ChatGPT for the reference of what's better, why don't we just use ChatGPT for the whole thing?

Luke Harries

What an MCP is, it's a Model Context Protocol... which enables anyone to expose tools to AI agents.

Luke Harries

I think a personal AI assistant really only needs your WhatsApp, your calendar, and your email, and then it knows everything about you.

Luke Harries

Questions Answered in This Episode

In your “Copy Editor” GPT, what are the highest-leverage instruction rules (e.g., headings, stats, tone) that most improved first-pass quality?

Luke Harries (Head of Growth at ElevenLabs) shares three practical AI workflows that make marketing execution faster, cheaper, and more systematic: automated case studies, in-house localization, and an MCP-based WhatsApp assistant.

Get the full analysis with uListen AI

How do you validate that the GPT is staying faithful to customer claims (ROI/time-saved) and not exaggerating when turning transcripts into polished case studies?

He demonstrates a “case study in minutes” pipeline using Granola transcripts plus a custom ChatGPT “Copy Editor” GPT trained on ElevenLabs’ tone, formatting rules, and examples—then repurposes outputs into tweet threads and other launch assets.

Get the full analysis with uListen AI

What did the per-language localization prompts include (glossary terms, forbidden translations, formality rules), and how did you test translation quality across languages?

He describes ripping out a $40k/year localization SaaS plus $100k+ in agency spend by building a lightweight translation server + GitHub Action with per-language prompts, keeping human review only for sensitive pages.

Get the full analysis with uListen AI

You said the localization SaaS couldn’t edit the AI prompt—what minimum level of prompt/control/observability should buyers demand from AI-enabled SaaS vendors now?

Finally, he explains MCPs (Model Context Protocol) and shows an unofficial WhatsApp MCP that downloads chats to a local SQLite DB so Claude can summarize, search, and message—hinting at a future personal assistant built on WhatsApp, email, and calendar.

Get the full analysis with uListen AI

For the GitHub Action translation workflow, how do you handle versioning, rollbacks, and detecting “bad” translations before they ship to production?

Get the full analysis with uListen AI

Transcript Preview

Luke Harries

When you're editing, as much as possible, try and edit the underlying prompt rather than the actual output.

Claire Vo

I like that you have here the, "You are a," and then gives a very specific identity and job to be done at the top of this. And then you have very specific instructions where you say, "You must do A, B, C, D," and it's quite particular.

Luke Harries

This saved us $40,000 a year for the tool, so immediately canceled it. Over $100,000 in agency costs. I think the highlight, though, is just not having to deal with more SaaS vendors, more agencies, constantly trying to get upsold. [upbeat music]

Claire Vo

Welcome back to How I AI. I'm Claire Vo, product leader and AI obsessive, here on a mission to help you build better with these new tools. Today, we have a great conversation with Luke Harries, Head of Growth at ElevenLabs. Luke shows us how to make everything a launch by making everything automated with AI. He shows us his secret flows for generating case studies and tweets on the fly, how he saved his company tens of thousands of dollars by, yes, being a marketer that coded in Cursor, and explains what an MCP is and how he hooked it up to WhatsApp. Let's get to it. This episode is brought to you by Orkes, the company behind open source Conductor, the platform powering complex workflows and process orchestration for modern enterprise apps and agentic workflows. Legacy business process automation tools are breaking down. Siloed low-code platforms, outdated process management systems, and disconnected API management tools weren't built for today's event-driven, AI-powered, cloud-native world. Orkes changes that. With Orkes Conductor, you get a modern orchestration layer that scales with high reliability, supports both visual and code-first development, and brings human, AI, and systems together in real time. It's not just about tasks. It's about orchestrating everything: APIs, microservices, data pipelines, human-in-the-loop actions, and even autonomous agents. So build, test, and debug complex workflows with ease. Add human approvals, automate back-end processes, and orchestrate agentic workflows at enterprise scale, all while maintaining enterprise-grade security, compliance, and observability. Whether you're modernizing legacy systems or scaling next-gen AI-driven apps, Orkes helps you go from idea to production fast. Orkes, orchestrate the future of work. Learn more and start building at orkes.io. That's O-R-K-E-S dot I-O. Hey, Luke, thanks for joining.

Luke Harries

Thanks for having me.

Claire Vo

In 2025, we've talked a lot about vibe coding, Cursor this and v0 that, but we have not talked enough, I think, about vibe marketing. So what do you think the future of an AI CMO is in the next couple years?

Luke Harries

There's all these tools like Lovable and Cursor, and the rate of software production's going to go exponential, but it's not gonna matter if no one's actually using your tool. And so what's important is actually getting the product into market, getting people to know about your new features. At ElevenLabs, we have this launch process. So basically, every new feature we do, every new model, we run it through this massive checklist, which is like, okay, we need to first work out what are the value props? Then we need to work out what's the core messaging. Then we need to work out who's it for. Then we need to turn that into the blog post, the X post, and it, it takes a lot of time, these massive launch processes. And so the thing I'm really excited for the AI CMO is being able to go from every single new feature or new product and translating that into your entire launch process, making the assets, making the videos, making the images, but then also going beyond the launch. So what are those then evergreen channels that you'll be testing? And so, uh, let's say ElevenLabs, we launched the best speech-to-text model. Okay, we need to be running Google Ads for that. So then it will spin up, understand all the various keywords, spin up the Google Ads. It will optimize the landing pages. So I think this entire thing is gonna change massively, um, and we're already using a few of these different workflows, and I'm excited to talk you through them.

Install uListen to search the full transcript and get AI-powered insights

Get Full Transcript

Get more from every podcast

AI summaries, searchable transcripts, and fact-checking. Free forever.

Add to Chrome