Why I'm Building My Automation Stack on Make.com
Blog post description.
AIAUTOMATIONAI AGENTS
Geoff Bunce
3/22/20264 min read
I'll be honest with you. I didn't spend weeks evaluating automation platforms, writing comparison matrices, and arriving at a carefully considered decision. I chose Make.com because it solved a specific problem I had, at a cost I could justify, and every time I've pushed it, it's pushed back with enough capability to keep me moving. That's the whole story but it's worth unpacking, because the reasoning behind it says something about how I think about building. The problem I was actually trying to solve My situation is probably familiar to anyone trying to build something on the side of a demanding day job. I'm Head of Solutions at Paragon ID, which means full SDLC responsibility: requirements, architecture, deployment, maintenance, managing a development team, alongside product strategy and tender responses. I coach football on weekends. I have three Labradors who require more management than the development team. Alongside all of that, I'm building Bunce Group, multiple independent revenue engines designed to reduce my reliance on a single employer and eventually give me genuine location independence. That means running Fidelis News, Dog and Outdoors, and Rook6 Lab simultaneously, in whatever time exists around everything else. The maths here is simple. I cannot manually do everything that needs doing. Any task I'm doing repeatedly by hand is a task that's stealing time from something that actually requires my thinking. Automation isn't a nice-to-have in this setup, it's a survival mechanism. Why Make.com specifically The first thing that sold me was the visual interface. I've built things in code my whole career, I'm not intimidated by logic but there's something genuinely useful about being able to see a workflow as a diagram rather than a script. When a scenario breaks at 2am, I want to see exactly where it broke and why, without hunting through logs. Make.com gives you that. The second thing is the HTTP module. This sounds mundane but it matters enormously. Any platform with a clean HTTP module means I'm not limited to its native integrations, I can connect to anything with an API. WordPress REST API, Claude API, NewsAPI, whatever comes next. The tool becomes a layer of logic that sits between services, and the services themselves are interchangeable. That flexibility is non-negotiable for how I build. The third thing is cost at the scale I'm operating. I'm not running enterprise workflows processing millions of operations. I'm running lean pipelines for a set of early-stage businesses. Make.com's pricing at that scale is sensible. Zapier would have cost me three times as much for the same functionality, and it would have been less flexible. The concrete example: Fidelis News which is the most developed pipeline I've built so far is for Fidelis News, my independent news publication covering UK and global affairs, with a focus on civil liberties, sovereignty, and the stories that get quietly ignored elsewhere. The target is two to three articles per day, each 1,200 to 1,800 words, SEO-optimised, and ready for editorial review. Doing that manually would take most of a working day. The pipeline I've built routes NewsAPI into Make.com, passes relevant stories to the Claude API with a structured prompt, and pushes the output as a draft post to WordPress via the REST API complete with SEO title, meta description, and focus keyword pre-populated in the Yoast fields. The whole thing runs automatically. I review drafts and publish; I don't write from scratch. Make.com is the connective tissue in that pipeline. It handles the routing, the conditional logic, the error handling, and the data transformation between services. It's not glamorous work, but without it, none of the rest of it functions. The honest bit, it's not perfect and I've hit walls. The most recent one was a JSON escaping issue when embedding dynamic variables into the Claude API payload, the kind of thing that works fine in testing and then falls over in production because a news headline contained a character that broke the JSON structure. The fix wasn't complicated once I understood it (build the prompt as plain text in a Set Variable module before it gets anywhere near the HTTP call), but finding it cost me time I didn't have. The documentation is inconsistent. Some integrations are better supported than others. And there are moments where something that should be simple requires three modules and a workaround. But here's the thing: the ceiling is high enough that it's worth the friction. Every wall I've hit has had a solution within the platform. I haven't yet reached a point where Make.com fundamentally couldn't do what I needed, I've just had to figure out how to ask it correctly. That's a tool worth keeping. The bigger point I've been influenced by The Sovereign Individual for a long time, the thesis that technology creates the conditions for individuals to operate independently of large institutions, if they choose to build that way. Automation tools like Make.com are a practical expression of that. They allow a single person to build and operate systems that would have required a small team a decade ago. That's not a small thing. The gap between someone who understands how to connect services and automate logic, and someone who doesn't, is enormous and it compounds over time. Every pipeline I build is leverage. Every hour I get back from automation is an hour I can put into something that genuinely requires my judgement. Make.com isn't the point. The point is building systems that work while I'm not watching them. Make.com just happens to be the best tool I've found for that so far. What's next The Fidelis pipeline is nearly complete. Once it's running cleanly, the next phase is building similar automation layers into Dog and Outdoors' content workflow and into Rook6 Lab's internal tooling. I'll document all of it here as it develops. If you're building something similar, or thinking about whether automation infrastructure is worth the investment at an early stage, feel free to reach out. It's a question worth thinking through carefully, and the answer isn't always yes.
Geoff Bunce
Geoff Bunce is an operator and builder with 10+ years in smart ticketing technology. Founder of Bunce Group: the parent company behind Fidelis News, Dog and Outdoors, and Rook6 Lab. Head of Solutions at Paragon ID. Writing, building, and documenting the journey at geoff.buncegroup.com.
DOg and outdoors ltd
© 2026. All rights reserved.