YouTube to Airtable: captions and images ready to post
Posting a YouTube video is the easy part. The annoying part is turning it into three different social posts, hunting for a decent visual, and then trying to remember where you saved everything.
Social media managers feel it first, because the requests never stop. But a content creator juggling filming days and a busy virtual assistant managing drafts will run into the same wall. This YouTube Airtable automation turns every new upload into captions and images you can actually use.
You will see how the workflow watches your channel, generates platform-ready copy plus AI visuals, and stores everything neatly in Airtable so you can review, tweak, then post.
How This Automation Works
The full n8n workflow, from trigger to final output:
n8n Workflow Template: YouTube to Airtable: captions and images ready to post
flowchart LR
subgraph sg0["Get YouTube Uploads Flow"]
direction LR
n0["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Get YouTube Transcript"]
n1@{ icon: "mdi:robot", form: "rounded", label: "Generate Social Posts + Imag..", pos: "b", h: 48 }
n2@{ icon: "mdi:swap-vertical", form: "rounded", label: "Structure Posts & Prompts by..", pos: "b", h: 48 }
n3["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Generate Instagram Image"]
n4["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Generate Facebook Image"]
n5["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Generate LinkedIn Image (Dum.."]
n6["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Upload Instagram Image to Ai.."]
n7["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Upload Facebook Image to Air.."]
n8["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/httprequest.dark.svg' width='40' height='40' /></div><br/>Upload LinkedIn Image to Air.."]
n9["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/airtable.svg' width='40' height='40' /></div><br/>Save Instagram Caption to Ai.."]
n10["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/airtable.svg' width='40' height='40' /></div><br/>Save Facebook Caption to Air.."]
n11["<div style='background:#f5f5f5;padding:10px;border-radius:8px;display:inline-block;border:1px solid #e0e0e0'><img src='https://flowpast.com/wp-content/uploads/n8n-workflow-icons/airtable.svg' width='40' height='40' /></div><br/>Save LinkedIn Caption to Air.."]
n12@{ icon: "mdi:play-circle", form: "rounded", label: "Get YouTube Uploads", pos: "b", h: 48 }
n12 --> n0
n0 --> n1
n4 --> n10
n3 --> n9
n10 --> n7
n11 --> n8
n9 --> n6
n5 --> n11
n1 --> n2
n2 --> n4
n2 --> n3
n2 --> n5
end
%% Styling
classDef trigger fill:#e8f5e9,stroke:#388e3c,stroke-width:2px
classDef ai fill:#e3f2fd,stroke:#1976d2,stroke-width:2px
classDef aiModel fill:#e8eaf6,stroke:#3f51b5,stroke-width:2px
classDef decision fill:#fff8e1,stroke:#f9a825,stroke-width:2px
classDef database fill:#fce4ec,stroke:#c2185b,stroke-width:2px
classDef api fill:#fff3e0,stroke:#e65100,stroke-width:2px
classDef code fill:#f3e5f5,stroke:#7b1fa2,stroke-width:2px
classDef disabled stroke-dasharray: 5 5,opacity: 0.5
class n12 trigger
class n1 ai
class n9,n10,n11 database
class n0,n3,n4,n5,n6,n7,n8 api
classDef customIcon fill:none,stroke:none
class n0,n3,n4,n5,n6,n7,n8,n9,n10,n11 customIcon
The Problem: Repurposing YouTube content takes forever
Repurposing is supposed to be “quick.” In reality, it’s a messy chain of small tasks that steal a whole afternoon. First you grab the transcript (or you don’t, and you try to skim the video at 1.5x speed). Then you write captions that sound different on Instagram, Facebook, and LinkedIn. Then you need images, which turns into rummaging through old folders, screenshots, Canva drafts, and half-finished ideas. By the time you’re done, the momentum from publishing the video is gone.
It adds up fast. Here’s where it breaks down once you start doing this every week.
- You rewrite the same idea three times because each platform wants a different style.
- Transcripts and notes live in random places, so future posts are hard to find and reuse.
- Image creation becomes a bottleneck, especially when you want visuals that feel consistent.
- Manual copy-paste into your content tracker invites mistakes, and those mistakes are weirdly hard to catch.
The Solution: New uploads become organized post drafts in Airtable
This workflow watches your YouTube channel’s RSS feed for new uploads. When a new video appears, it sends the video URL to Dumpling AI to extract the transcript automatically. Next, OpenAI takes that transcript and turns it into three platform-specific captions (Instagram, Facebook, and LinkedIn), plus three distinct prompts for images that match the video’s topic. Then the workflow generates realistic images via Dumpling AI using those prompts. Finally, it creates one Airtable record per platform with the caption, and patches each record to attach the matching AI image. You end up with a clean library of ready-to-post drafts, all in one base.
The workflow starts with a new YouTube upload. From there, the transcript becomes the raw material for social captions and image prompts. Airtable becomes your “single source of truth” for what’s ready, what needs edits, and what you’ve already posted.
What You Get: Automation vs. Results
| What This Workflow Automates | Results You’ll Get |
|---|---|
|
|
Example: What This Looks Like
Let’s say you publish 2 YouTube videos a week and repurpose each into 3 posts. Manually, a realistic run looks like 20 minutes to pull the transcript, about 30 minutes to write three captions, and another 30 minutes to find or create three images, so roughly 1.5 to 2 hours per video. With this workflow, your “work” is basically just letting it run, then spending about 10 minutes reviewing the three Airtable drafts and making them sound like you. That’s a few hours back every week, without lowering quality.
What You’ll Need
- n8n instance (try n8n Cloud free)
- Self-hosting option if you prefer (Hostinger works well)
- Airtable to store captions and image attachments.
- Dumpling AI for transcript extraction and image generation.
- OpenAI API key (get it from your OpenAI account dashboard).
Skill level: Intermediate. You’ll connect a few accounts, paste API keys, and map fields into Airtable.
Don’t want to set this up yourself? Talk to an automation expert (free 15-minute consultation).
How It Works
A new video hits your channel. The workflow is triggered by your YouTube RSS feed, so it reacts as soon as an upload appears.
The transcript is pulled automatically. n8n sends the YouTube URL to Dumpling AI’s transcript extraction endpoint, which returns the text you’d normally have to hunt down or copy manually.
Captions and image prompts are generated in one pass. OpenAI turns the transcript into three captions tailored to Instagram, Facebook, and LinkedIn, then creates three prompts designed to produce visuals that match the content.
Airtable becomes your draft hub. The workflow creates one record per platform (caption first), generates the platform-specific image, then updates each record to attach the correct image file.
You can easily modify the caption prompt to match your brand voice based on your needs. See the full implementation guide below for customization options.
Step-by-Step Implementation Guide
Step 1: Configure the RSS Feed Trigger
This workflow starts by polling a video RSS feed for new items.
- Add the Retrieve Video Feed node as the trigger.
- Set Feed URL to
https://rss.app/feeds/FmN1GIARUeCC6ELF.xml. - Keep the default polling schedule unless you need a custom interval.
Step 2: Connect the Transcript Retrieval
Next, the workflow fetches the video transcript for each RSS item.
- Add the Fetch Video Transcript node and connect it from Retrieve Video Feed.
- Set URL to
https://app.dumplingai.com/api/v1/get-youtube-transcriptand Method toPOST. - Set JSON Body to
{ "videoUrl": "{{ $json.link }}", "includeTimestamps": "true", "preferredLanguage": "en" }. - Credential Required: Connect your httpHeaderAuth credentials.
- Ensure Send Headers is enabled and includes Content-Type with value
application/json.
Step 3: Set Up AI Content Generation
The AI step creates platform-specific captions and image prompts from the transcript.
- Add Create Social Copy & Prompts and connect it from Fetch Video Transcript.
- Confirm Model is set to
gpt-4o-miniand JSON Output is enabled. - Keep the system message content as provided to ensure the JSON schema output is consistent.
- Credential Required: Connect your openAiApi credentials.
Step 4: Map Platform Fields and Generate Visuals (Parallel)
The workflow maps AI output into fields and then generates images for each platform in parallel.
- Add Map Platform Posts and connect it from Create Social Copy & Prompts.
- Set field assignments to the following expressions:
- Instagram →
{{ $json.message.content.instagram.post }}, Facebook →{{ $json.message.content.facebook.post }}, Linkedin →{{ $json.message.content.linkedin.post }}. - IG Image Prompt →
{{ $json.message.content.instagram.image_prompt }}, Facebook Image prompt →{{ $json.message.content.facebook.image_prompt }}, LinkedIn Image prompt →{{ $json.message.content.linkedin.image_prompt }}. - Map Platform Posts outputs to both Produce Facebook Visual and Produce Instagram Visual and Produce LinkedIn Visual in parallel.
- For each “Produce” node, set URL to
https://app.dumplingai.com/api/v1/generate-ai-image, Method toPOST, and the JSON Body to the provided prompt structure. - Credential Required: Connect your httpHeaderAuth credentials to all AI image request nodes (7+ HTTP nodes are used in this workflow).
Step 5: Store Captions in Airtable and Attach Images
Each platform caption is saved to Airtable, then the image URL is attached to the created record.
- Add Store Instagram Caption, Store Facebook Caption, and Store LinkedIn Caption nodes, each connected from their respective “Produce” node.
- For each Airtable node, set Operation to
create, map Content to the corresponding expression (e.g.,{{ $('Map Platform Posts').item.json.Instagram }}), and set Platform to the literal value. - Credential Required: Connect your airtableTokenApi credentials to all Airtable nodes.
- Add Attach Instagram Image, Attach Facebook Image, and Attach LinkedIn Image nodes connected from their respective “Store” nodes.
- Set each URL to
=https://api.airtable.com/v0/[YOUR_ID]/[YOUR_ID]/{{ $json.id }}and use the provided JSON Body to attach the image URL from the related “Produce” node. - Credential Required: Connect your httpHeaderAuth credentials to the Airtable attachment nodes.
[YOUR_ID] in the Airtable API URL with your Base ID and Table ID in all “Attach” nodes.Step 6: Test and Activate Your Workflow
Verify the workflow end-to-end before enabling it for production use.
- Click Execute Workflow and ensure Retrieve Video Feed pulls a recent RSS item.
- Confirm Fetch Video Transcript returns a transcript and Create Social Copy & Prompts outputs valid JSON.
- Verify that Airtable records are created and images are attached by checking the Store and Attach nodes’ outputs.
- If successful, toggle the workflow to Active so it runs automatically on new feed items.
Common Gotchas
- Airtable tokens and base permissions can block attachment updates. If images don’t appear, check the Airtable personal access token scopes and the base access first.
- If Dumpling AI is slow (or you’re generating multiple images), processing times vary. Increase any wait time you’ve added or rerun the failed image/patch step so the workflow isn’t patching an empty image.
- OpenAI outputs will feel generic if you don’t tell it what “you” sounds like. Add tone, formatting rules, and do-not-say phrases early, or you’ll be rewriting every caption anyway.
Frequently Asked Questions
About 45 minutes if your Airtable base is ready and you already have API keys.
No. You’ll mostly connect accounts, paste keys, and map a few Airtable fields.
Yes. n8n has a free self-hosted option and a free trial on n8n Cloud. Cloud plans start at $20/month for higher volume. You’ll also need to factor in OpenAI and Dumpling AI usage costs, which depend on how many videos you process.
Two options: n8n Cloud (managed, easiest setup) or self-hosting on a VPS. For self-hosting, Hostinger VPS is affordable and handles n8n well. Self-hosting gives you unlimited executions but requires basic server management.
Yes, but you’ll want to adjust both the “Create Social Copy & Prompts” step and your Airtable fields. Add another caption output in the OpenAI prompt, extend the “Map Platform Posts” mapping so the new platform has its own content field, then duplicate the create/update pattern used for the existing Airtable records and image attachments. Many teams also add a “Status” field (Draft, Reviewed, Scheduled) so edits and approvals don’t get lost.
Usually it’s an expired or under-scoped personal access token. Regenerate the token, confirm it can create records and update attachments in the right base, then paste it back into the Airtable credential in n8n. If record creation works but images don’t attach, double-check the Attachment field type and make sure the PATCH step is pointing at the correct record ID.
If you self-host n8n, there’s no fixed execution cap (it mainly depends on your server and your API limits). On n8n Cloud, your limit depends on your plan, and heavy usage comes more from image generation than from simple record creation. Practically, most small teams run this for a few videos a week to a few videos a day without issues, as long as their Dumpling AI credits are active and Airtable isn’t being rate-limited.
Often, yes, because this workflow has multiple AI calls, branching for three platforms, and record updates for attachments, which can get pricey and fiddly elsewhere. n8n handles multi-step logic cleanly, and self-hosting keeps you from paying per tiny action. Zapier or Make can still win for a very simple “trigger → one action” flow, especially if you never touch AI. If you’re unsure, write down how many posts you want per upload and how many tools are involved, then Talk to an automation expert and sanity-check it.
Once this is running, your YouTube uploads automatically turn into a tidy Airtable queue of captions and visuals. You’ll spend your time improving the message, not rebuilding the same draft from scratch.
Need Help Setting This Up?
Our automation experts can build and customize this workflow for your specific needs. Free 15-minute consultation—no commitment required.
