Every week, this blog writes itself, posts to LinkedIn, and lands in your inbox. We thought you might want to know how.
There's a certain irony in an AI consultancy manually writing blog posts, scheduling social media, and sending newsletters. We noticed it too. So we built a system to handle most of it automatically — and then we thought, what better way to show what's possible than to explain exactly how we did it.
This isn't a technical manual. It's a plain-English look at what we built, what tools we used, and why any business with a marketing function should at least understand what's now possible.
The problem we were solving
Consistent content marketing is one of those things that everyone agrees is important and almost nobody does well. Not because people lack ideas, but because they lack time. Writing a blog post, turning it into social media copy, designing a header image, sending a newsletter — done properly, that's a full day's work every week.
We wanted to produce genuinely useful content for Australian business leaders on a daily and weekly basis, without that content consuming our team. The result is what we call BRAiKING NEWS — a system that researches AI news, writes social posts, drafts blog posts, generates images, and distributes everything with almost no human intervention.
Almost. The human still reads it before it goes out. That matters, and we'll come back to it.
What the system does
Every morning at 6am, the system wakes up and scans the internet for AI news that's relevant to Australian businesses. It selects three stories, writes a two or three sentence summary of each, explains why it matters for business leaders, and queues them to post on LinkedIn and BlueSky throughout the day.
Once a week, it takes the most interesting of those stories and expands it into a full blog post — complete with a header image generated specifically for that topic. It then writes a newsletter combining the week's highlights and the blog link, creates a draft campaign ready to send, and posts to the AiGILE company LinkedIn page with the blog image attached.
Everything gets reviewed before it goes live. A human reads it, decides it's good enough, and presses send. That's the only step we haven't automated — deliberately.
The tools involved
You don't need to understand the technical details to appreciate the picture. Here's what's doing the work:
Make.com is the engine room. It's an automation platform that connects different software tools and tells them when to talk to each other. Think of it as the conductor — it doesn't make music itself, but nothing happens without it.
Claude by Anthropic does the writing. Every piece of copy — the social posts, the blog drafts, the newsletter, the LinkedIn content — is written by Claude based on instructions we've carefully crafted. Getting those instructions right is genuinely skilled work. The difference between a prompt that produces generic AI-sounding waffle and one that produces something you'd actually want to read is significant.
Google Gemini generates the blog header images. Each week it creates an original editorial-style photograph based on the topic of that week's blog. We gave it guidance on our brand colours, the tone we're after, and the kind of imagery that works for an Australian business audience. It takes about 30 seconds.
Google Sheets is the database. Every story researched, every post written, every blog drafted — it all lives in a spreadsheet that serves as the system's memory and control panel. It's where we review content before approving it.
MailerLite handles the newsletter. The system creates a fully formatted draft campaign, ready for us to review and send with one click.
LinkedIn and BlueSky receive the finished posts directly from the system, on schedule, without anyone manually copying and pasting.
The bit that took the most work
Writing the instructions for Claude was by far the most time-consuming part of the build. You can point Claude at a news story and ask it to write a LinkedIn post, and it will. But left to its own devices, it will use phrases like "game-changer" and "leverage" and "revolutionise" and produce something that sounds exactly like every other AI-generated content you've ever ignored.
Getting it to write in a specific voice — direct, commercially minded, occasionally dry, never using buzzwords — required a lot of iteration. We ended up with detailed guidance that covers not just what to write, but what never to say, how to open a post, and what the audience actually cares about.
That investment pays off every time the system runs.
Why this matters for your business
The specific system we built is designed for content marketing, but the underlying approach applies almost anywhere you have a repetitive process that involves finding information, making decisions about it, and producing an output.
Customer communications. Monthly reporting. Meeting summaries. Tender responses. Compliance checklists. The tools exist to automate significant portions of all of these — not to remove the human judgement, but to remove the manual labour that surrounds it.
The human review step we kept in our system isn't a failure of automation. It's a deliberate design decision. AI writes a first draft in seconds. A human decides if it's good enough in minutes. The total time invested is a fraction of what it was before — but the quality control stays in place.
What it cost to build
The software tools involved are not expensive. Make.com, Claude, Gemini, MailerLite — the combined monthly cost for a system like this is well under a few hundred dollars. The investment is in the setup: understanding what you want to automate, configuring the tools, and writing the instructions that produce the right output.
That's exactly the kind of work we do with clients. If something in this post sparked an idea about a process in your business that could run more efficiently, we'd enjoy the conversation.
This blog post was drafted by Claude, reviewed by a human, and published to Squarespace as part of the system described above. The header image was generated by Google Gemini. Total hands-on time to produce this post: about four minutes.
Not sure where your business stands with AI?
Find out your AiDOPTION Score — a free 10-minute diagnostic that measures your AI readiness across Strategy, Technology, and People. You'll get a personalised score and practical recommendations.