- The Content Engineer
- Posts
- A Simple LinkedIn Content System | (Guide + Gumloop Automation)
A Simple LinkedIn Content System | (Guide + Gumloop Automation)
Turn raw conversations into differentiated LinkedIn content

Welcome back, content engineers.
Today’s newsletter:
Deep Dive: Turn interviews into weeks of unique LinkedIn content
News: HubSpot plugs into ChatGPT, Anthropic CEO predicts 20% job losses, Sweden launches content engineering diploma
~ 5 min read ~

DEEP DIVE
How to Turn Raw Conversations into Differentiated LinkedIn Content
Nobody likes LinkedIn. At least from what I can tell. And yet, millions of people spend countless hours scrolling through it every day.
So, if you’re a business owner or creator, it’s worth engaging with, even if posting feels deeply, almost existentially embarrassing (just me?).
Anyways, to take the pain out of the process, Mike and I have built a simple Gumloop automation that turns interview/meeting transcripts into weeks of unique LinkedIn content.
Note* This one’s purpose-built for marketers and agencies looking to scale founder-led content programs (but it works for solo creators too).
Here's the flow:
→ Interview your founder/client (or yourself)
→ Structure questions around the funnel
→ Upload transcript to Gumloop automation
→ Get 3–8 voice-matched posts in a Google Doc
→ Doc auto-delivers to a Slack channel for editing
→ Slack channel doubles as a content library/archive
You can grab the automation here: https://www.gumloop.com/pipeline?workbook_id=dZLACddEyCgVw88ZuipV9w
Let’s dive into the details 👇
The System
The full system has three components:
Scheduled interviews – Book 30-minute conversations weekly or bi-weekly. Consistency matters more than frequency.
The automation – Analyzes your past LinkedIn posts to learn your voice, then transforms new transcripts into 3-8 formatted posts. Delivers everything to Slack in 90 seconds.
Human review – Edit posts on your schedule. I recommend batch editing right after the interview.
Part 1: The Interview
The quality of your posts depends largely on the questions you ask. Structure them around the funnel:
Top of Funnel Questions (TOFU): Cast a wider net with broader, more relatable content. If you're interviewing a founder or client, these are the kinds of questions you can ask:
Looking ahead, what excites you most about where your company is going?
What's the craziest thing that happened while fundraising?
What hiring mistake do you see every founder make?
These posts drive reach and get the algorithm on your side. Examples 👇


Mid-Funnel Questions (MOFU): Build authority with industry-specific thought leadership
Why is [common industry practice] completely broken?
What's your contrarian take on [industry trend]?
How do you think about [specific problem your ICP faces]?
Walk me through your framework for [X]?
This is where you become the go-to resource for your ideal customer. Examples 👇


Bottom of Funnel Questions (BOFU): Convert attention into pipeline:
Walk me through [client name]'s results after working with us
Why did we approach [product feature] differently than competitors?
What specific problem does your latest feature solve?
Less engagement, but necessary to actually drive revenue. Example 👇
Another example, focused on showcasing a product 👇
Part 2: The Automation
Here's what actually happens when you upload your transcript:
Step 1: You open the Gumloop interface and upload two things:
The meeting date (for tracking)
Your transcript file
Step 2: The automation reads your transcript AND pulls in examples of previous LinkedIn posts from a live Google Doc (this is how it learns the person's voice).
Step 3: AI analyzes both inputs – the new ideas from your transcript plus the voice/style from past examples
Step 4: It generates 3-8 LinkedIn posts, formats them in a new Google Doc, and drops the link to the doc in your Slack channel
The whole sequence takes 90 seconds.
Part 3: The Human in the Loop
The final step is pretty straightforward. Review the Google Doc that lands in your Slack channel. Edit it on your own schedule.
If you're working with a busy founder, they now have every conversation converted into posts, waiting in a dedicated channel. They can pop in whenever they have five minutes, choose a post, edit it, and hit publish.
Nailing the Samples
The results from this Gumloop automation can be exceptionally good if you feed it enough LinkedIn post samples, so make sure to set up that node properly.
Example output 👇

Start Building
Most AI tools rewrite your ideas. This one doesn’t. It pulls your actual phrasing from real conversations and turns it into content that sounds like you, just tighter and more shareable. No generic voice. No lost nuance. Just your thinking, cleaned up and ready to post.
Good luck!

BLUE LINKS
CEOs keep writing the same AI memo – here's what they're really saying:
Behind the ultimatums, execs are basically asking for help
HubSpot becomes first CRM to plug into ChatGPT
Plus, OpenAI rolls out several other MCP connectors like Google Drive and Calendar (all of which continue to work better in Claude imo)
Anthropic's CEO warns we're sleepwalking into a white-collar bloodbath.
Dario Amodei predicts 20% of entry-level jobs vanish in 1-5 years… this one reads like a PR push (Anthropic is currently looking to raise money, with AI safety initiatives as a centrepiece)
Sweden’s Berghs School launches AI content engineering program.
Taught in English, their 'one-year' AI diploma costs 16K USD
Looks like Google's new AI Mode will get trained on PageRank (i.e. the same criteria used to prioritize content in Google Search)
SEO's death greatly exaggerated? Yes and no… will be diving into this in an upcoming issue

THAT’S A WRAP
How did you like today's newsletter? |
Your vote helps me create better content 🙏
Till next time,
- Niko & Mike
