Make.com Content Automation (2026): Build a Full Workflow on the Free Plan
📑 Table of Contents
🎯 Quick Summary
Make.com’s free plan is tight — 1,000 operations per month, 2 active scenarios — but it’s enough to run a real content automation workflow. The key is designing lean: one trigger, one research call, one AI call, one publish step, one update step. Five modules. Around 8–10 ops per article run.
Make.com is where the workflow actually lives. It’s the thing that watches your spreadsheet, fires the API calls, and delivers the finished article to WordPress. But it’s also where most people get stuck — not because it’s hard, but because there’s no single guide that covers this specific workflow from start to finish.
This is that guide. Step by step, module by module, with the exact settings that keep you inside the free plan’s operation budget.
What the Make.com Free Plan Actually Gives You
The free plan has three constraints worth understanding before you build:
- 1,000 operations/month — every module execution in a scenario run costs 1 operation
- 2 active scenarios — you can have more scenarios saved, but only 2 can be active (scheduled to run) at a time
- 15-minute minimum interval — the fastest your scenario can check for new work is every 15 minutes
The 15-minute interval is fine for a content workflow. You’re not building a real-time system. The 2-scenario limit means you should build one well-designed scenario rather than splitting logic across multiple. And the 1,000 op limit, at ~10 ops per article, gives you roughly 100 runs per month — plenty for a growing content site.
The Scenario Structure
The workflow is a linear chain of five modules. No branching, no loops (for the basic version), no complex logic. Build it in this order and test each connection before adding the next.
Google Sheets (Filter Rows)
↓
Tavily (Search)
↓
Gemini AI (Generate Content)
↓
HTTP (POST to WordPress REST API)
↓
Google Sheets (Update Row)
Each module passes data to the next. The Google Sheets row provides the article topic. Tavily uses that topic to fetch research. Gemini receives the research plus your prompt and writes the article. The HTTP module posts the result to WordPress. The final Sheets module marks the row as done so it doesn’t get picked up again.
Module 1: Google Sheets — Filter Rows
This is your trigger. Instead of using the “Watch Rows” module (which uses a webhook or polling cursor), use Filter Rows — it scans your sheet on every run and returns rows matching your filter condition. Simpler, more predictable, and easier to debug.
Key settings:
- Spreadsheet ID: your Google Sheet’s ID from the URL
- Sheet Name: whatever you’ve named your content tab
- Filter: Column B (Status) equals “Write Approved”
- Limit:
1— critical. This processes one row per run, keeping ops usage predictable - Table contains headers: Yes
Module 2: Tavily — Search
Add the native Tavily module (search for “Tavily” in the module browser). Connect your Tavily API key when prompted.
Key settings:
- Query: map in your article title from the Sheets module. Add context:
{{Sheets_module_ID.`3`}} 2026 review pricing plans benchmark features vs alternatives - Search Depth: Advanced
- Max Results: 8
- Include Raw Content: Yes
- Chunks Per Source: 3
- Start Date:
2020-01-01
Module 3: Gemini AI — Generate Content
Make.com has a native Gemini AI module. Add it and connect with your Gemini API key from Google AI Studio.
Key settings:
- Model: gemini-2.0-flash (fast, free tier, good for long-form)
- Role: User
- Content: Your full prompt text, with
{{tavily_module.results[]}}injected at the research data section and{{sheets_module.`3`}}for the article topic
The prompt is the most important piece of this entire workflow. A weak prompt produces generic output regardless of how good your research data is. Prompt engineering is covered in depth in Guide 4 of this series — read that before finalising this module.
Module 4: HTTP — POST to WordPress REST API
This is the module that does the actual publishing. Use Make.com’s built-in HTTP → Make a Request module — not a WordPress-specific module, which adds unnecessary complexity.
Before you build this module, set up your WordPress connection:
- In WordPress admin, go to Users → Your Profile
- Scroll to Application Passwords
- Enter a name (e.g. “Make.com”) and click Add New Application Password
- Copy the generated password immediately — it won’t be shown again
- Your credentials for Basic Auth are:
your_wp_username:the_application_password
HTTP module settings:
- URL:
https://yoursite.com/wp-json/wp/v2/posts - Method: POST
- Headers: Authorization: Basic [base64 of username:app_password]
- Content Type: JSON (application/json)
- Body: JSON object with title, content, excerpt, status, slug, meta fields
{
"title": "{{sheets.article_title}}",
"content": "{{gemini_cleaned_output}}",
"excerpt": "{{parsed_excerpt}}",
"status": "draft",
"slug": "{{parsed_slug}}",
"meta": {
"_yoast_wpseo_title": "{{parsed_seo_title}}",
"_yoast_wpseo_metadesc": "{{parsed_meta_description}}"
}
}
"status": "draft" until you’ve verified the output quality is consistent. Switch to "publish" only once you trust the full pipeline. One bad auto-published article is harder to clean up than a draft sitting quietly in your WordPress backend.
Module 5: Google Sheets — Update Row
The final module marks the processed row so it won’t be picked up again on the next run. Use Google Sheets → Update a Row.
Map the row number from your Filter Rows module and update column B (Status) to “Published” or “Drafted” — whatever value your filter condition doesn’t match. This is what prevents the same article from being processed twice.
Staying Inside Your Operations Budget
With five modules and Limit: 1 set on your trigger, each scenario run costs exactly 5 operations (one per module). If you add RegExp parsers to clean Gemini’s output, that adds 1 op per parser. A scenario with 10 modules = 10 ops per run = 100 article runs per month on the free plan.
The scheduling sweet spot for free plan users is running the scenario every hour. At 10 ops per run, 24 runs per day = 240 ops. Over 30 days that’s 7,200 ops — but you only have 1,000. So in practice: run on-demand by activating the scenario manually when you have approved rows, or schedule it to run once or twice per day rather than continuously.
🚀 Start Building Your Scenario
Make.com’s free plan requires no credit card. Create your account, start a new scenario, and follow the module order above. The whole thing can be running in an afternoon.
Sign Up for Make.com Free →No credit card required
❓ Frequently Asked Questions
What is the difference between Filter Rows and Watch Rows in Make.com?
Watch Rows uses a cursor to track which rows it has already processed — useful for high-frequency data. Filter Rows scans the sheet on every run and returns rows matching your conditions. For a content workflow where you manually mark rows as approved, Filter Rows is simpler and more predictable. Watch Rows can miss rows if your cursor gets out of sync.
How do I set up WordPress Application Passwords?
Application Passwords are built into WordPress core since version 5.6. Go to Users → Your Profile → Application Passwords section. Enter a name for the connection, click Add New, and copy the generated password. Use your WordPress username and this password for Basic Auth in the Make.com HTTP module. Note: Application Passwords require HTTPS on your site.
Why does my Gemini output get cut off in Make.com?
The Gemini AI module in Make.com sometimes defaults to a low max tokens limit. Check your module settings and either remove the max tokens limit or set it to 8,000 or higher. Long-form articles (2,000+ words) need significant token budget. If output is still being cut, check that Gemini Flash’s context window supports your full prompt plus output length.
Can I add error handling to stop a row getting stuck?
Yes. Make.com supports error handlers on any module — right-click a module and select “Add error handler.” For production use, add a handler on the HTTP module that updates your Google Sheet row status to “Error” if the WordPress post fails. This prevents the same row being retried indefinitely and gives you a clear list of failed runs to investigate.
When should I upgrade from the Make.com free plan?
Upgrade when you consistently hit 800+ operations per month, when you need more than 2 active scenarios (e.g. a separate keyword research workflow), or when the 15-minute minimum interval becomes a constraint. Make.com’s Core plan at $9/month gives you 10,000 ops and unlimited scenarios — the right move once the free workflow is validated and producing results.
Latest Articles
Browse our comprehensive AI tool reviews and productivity guides
Claude for Small Business Review (2026)
Anthropic's Claude for Small Business ships with 15 ready-to-run AI workflows inside tools like QuickBooks, PayPal, HubSpot, and Canva. We break down what it does, who it's for, and whether it's worth your time.
Generative Engine Optimization (GEO) 2026: How to Get Your Content Cited by ChatGPT, Perplexity & Google AI
Traditional SEO gets you ranked. GEO gets you cited. With 60% of searches now ending without a click and AI Overviews slashing organic CTR by 58%, getting your content into AI answers is the new growth channel. Here's the complete playbook for 2026.
Perplexity Projects Explained: New Workflow System
Perplexity Projects are changing AI research with a new workflow system that enhances productivity and streamlines complex tasks.
Bika.ai Review: No-Code Agentic Database for AI
Is Bika.ai the no-code agentic database solution you've been searching for? This review breaks down its features, pricing, and potential.
Gumloop Review 2026: Drag-and-Drop AI for Founders
A comprehensive Gumloop review for non-technical founders, evaluating its drag-and-drop AI capabilities, pricing, and suitability for business automation.
LangGraph vs AutoGen: Advanced State Management 2026
Compare LangGraph and AutoGen for advanced AI agent state management in 2026, detailing benchmarks, pricing, and real-world application differences.
Commonstack AI: Intelligent Model Routing Guide
Discover how Commonstack AI optimizes LLM usage with intelligent model routing for cost savings.
Clawbot AI Review 2026: Multi-Agent Orchestration Compared
An in-depth look at Clawbot AI versus CrewAI for multi-agent orchestration, examining their capabilities, pricing, and ideal use cases.
Claude Code vs n8n: Connecting AI for Auto-Healing Pipelines
Explore Claude Code vs n8n for agentic workflows, detailing their strengths in code automation and business process integration.
DeepSeek V4 Review 2026: The Largest Open-Weight Model Ever — Pro, Flash, Benchmarks & Pricing
DeepSeek V4 Review 2026: The Largest Open-Weight Model Ever — and the Biggest Disruption to AI Pricing
Gemini 3.5 Ultra Review: Google’s 10-Million Token Sovereign — The End of the Context Wars? (May 2026)
Gemini 3.5 Ultra completed global rollout across all Google One AI Premium accounts and Enterprise API tiers. Benchmark data sourced from Artificial Analysis v4.2, Google DeepMind Technical Reports, and independent stress testing from NivaaLabs.
Grok 4.3 Review 2026: xAI’s Cheapest Frontier Model — Benchmarks & Verdict
Grok 4.3 launched May 6, 2026 with a 40% price cut, 1M token context, native video, and a 321-point Elo jump on agentic benchmarks — but still no persistent memory at any price.