Mindket

Outlier YouTube videos turned into title and outline ideas

Newversion 1 · updated

Overview

An n8n workflow that finds YouTube videos on your topic with more views than their channel has subscribers. Apify fetches each transcript, and OpenAI writes a new title, thumbnail text and outline into your Google Sheet.

Outlier YouTube videos turned into title and outline ideas is an n8n workflow in the Marketing & SEO category, listed by Mindket. It is free. This is version 1, last updated on . Download it on this page or copy its JSON, then import it into an n8n canvas. Nothing runs until you activate it in n8n.

At a glance

Runs in
  • n8n Cloud: import the file into a new workflow there
  • Your own n8n server: the same file and steps. A run with many outliers takes a while
  • Google Sheets: one row for each outlier, its details first and the new ideas filled in after
  • Your writing tool: copy an outline from newOutline into the document you script in

Platform

Connects 3 apps

How it works

Trigger

Topic form

It starts when someone sends the YouTube Topic Research form with:

  • A keyword or topic

Actions

  1. An Apify Actor searches YouTube for up to 100 videos from this month, 4 to 20 minutes long
  2. Keeps those with over 1,000 views and more views than subscribers, and drops any already in the sheet
  3. Saves each one to the Step 1 tab
  4. OpenAI describes its thumbnail, and a second Apify Actor fetches its transcript
  5. OpenAI writes a new title, thumbnail text and outline into the same row

Result

A sheet of outlier videos on your topic, each with a new title, thumbnail text and outline for your own take

Integrations

What you need before you start.

Apify

YouTube scraping

Runs one Actor that searches YouTube by your keyword and a second that returns each video’s transcript.

NeedsApify API token, typed into two headers, and credit for the runs

OpenAI

Vision and language models

Describes the thumbnail with gpt-4o-mini, writes the title and thumbnail text with gpt-4.1-mini and the outline with gpt-4.1.

NeedsOpenAI API key

Google Sheets

Idea sheet

Checks each video against the rows already there, adds the new ones and fills in their ideas.

NeedsA Google account to sign in with, and a sheet with 17 headers

Workflow check

Checked by Mindket on 13 Nov 2026

Steps
11 steps, starting on a form
AI
3 steps call an AI model: Analyze Thumbnail, YouTube Title Generator and Outline Generator
Sign-ins
Signs in to: Google Sheets, OpenAI and Apify
API keys
No API key is left in the file; 1 blank in it for your own key
Step types
Mindket’s checker reads all 6 of its step types

Read from the workflow’s file, not from a run.

About

What it is

This n8n workflow finds YouTube videos on a topic that did better than their channel’s size suggests, then plans your own take on each. A form takes a keyword. An Apify Actor searches YouTube for up to 100 recent videos of 4 to 20 minutes. It keeps those with over 1,000 views and more views than subscribers, skipping any already in your sheet. For each one, OpenAI describes the thumbnail, reads the transcript Apify fetched and writes a new title, thumbnail text and outline. Eleven nodes run it.

Who it suits

YouTubers and content teams who plan videos from what already works in their niche. A small channel’s breakout video says more about an idea than a big channel’s ordinary one, and that is the test it applies. Skip it if you want a script. It stops at an outline.

Who uses it, and how
IndustryWho uses itHow they use itExample
EducationTeachers who post lessonsSearch a subject each week for videos that broke outA maths tutor searches algebra tricks and plans two videos from the outlines.
SoftwareDeveloper advocatesSearch a tool’s name and read how the outliers frame itAn advocate finds a tutorial with 40,000 views on a channel of 3,000 subscribers.
Personal financeSolo creatorsKeep the sheet as a running list of ideasA finance creator runs five keywords on Monday and films the best idea that week.
Marketing agenciesContent strategistsBuild idea lists for client channelsA strategist sends a client ten outlines with the videos behind them.

You need some n8n and three accounts: Apify, OpenAI and Google. Its author puts setup at about 26.5 minutes.

Where to use it

Any n8n that can reach Apify, OpenAI and Google. Everything lands in one tab of a Google Sheet.

  • n8n Cloud: import the file into a new workflow there
  • Your own n8n server: the same file and steps. A run with many outliers takes a while
  • Google Sheets: one row for each outlier, its details first and the new ideas filled in after
  • Your writing tool: copy an outline from newOutline into the document you script in

Setup

Get it saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Make the sheet first, since three nodes read or write it.

What you need first
NeedDetail
n8nOn n8n Cloud or on a server of your own
ApifyAn account and its API token, with credit for a YouTube search Actor and a transcript Actor
OpenAIAn API key for gpt-4o-mini, gpt-4.1-mini and gpt-4.1
GoogleAn account that can edit the sheet, for a Google Sheets OAuth2 credential
A sheetA tab named Step 1 with the 17 headers in the next table
The headers of Step 1
Filled byHeaders
Step 1 Resultsinput, id, title, url, fromYTUrl, thumbnailUrl, channelName, channelUrl, duration, likes, numberOfSubscribers, viewCount
Update RowsvideoTranscript, thumbnailDescription, thumbnailText, newTitle, newOutline
  1. In n8n, create a workflow, open its menu, choose Import from File and pick workflow.json.
  2. Make the sheet, then copy its ID: the long part of its address between /d/ and /edit.
  3. Add a Google Sheets OAuth2 credential. In Find Duplicate Entries, Step 1 Results and Update Rows, select it and replace YOUR_SHEET_ID with the ID. Each finds the tab by the name Step 1.
  4. Add an OpenAI credential and select it in Analyze Thumbnail, YouTube Title Generator and Outline Generator.
  5. In YouTube Video Scrape and HTTP Request, replace <YOUR_APIFY_TOKEN> in the Authorization header with your token. The brackets go too, so the header reads Bearer, a space and the token.
  6. Save, open the form’s test address and send one keyword.
  7. Press Publish once the rows look right.

How to use it

Send a keyword, such as a topic your channel covers. The form confirms at once. Apify’s search runs first, so the first row can take a few minutes. Each outlier then gets a row with its details, and the five idea columns fill in once its transcript and the three model calls are done. Send the same keyword next week and only new outliers are added.

What to change for a different result
To changeWhere
What counts as an outlierThe two tests in If: over 1,000 views, and more views than subscribers
What is searchedThe JSON body of YouTube Video Scrape: up to 100 videos of 4 to 20 minutes from this month, newest first
The new titleThe system message of YouTube Title Generator, which keeps the title under 70 characters and the thumbnail text to 3 to 5 words
The outlineThe system message of Outline Generator, which asks for a different angle in the usual YouTube sections
Modelsgpt-4o-mini for the thumbnail, gpt-4.1-mini for the title, gpt-4.1 for the outline

Example

You send home espresso.

Apify returns this month’s videos on home espresso. Say a channel with 2,400 subscribers posted one that reached 31,000 views. It passes both tests and is not in the sheet yet, so it becomes a row. The title step then answers in this shape:

{
  "newTitle": "…",
  "thumbnailText": "…"
}

The thumbnail description, the transcript and the outline fill the rest of the row. The outline runs from a hook and an introduction through the main points to a call to action.

For newcomers and experts

If you are new. n8n draws a workflow as boxes joined by lines, and each box, a node, does one job. The top line finds and filters videos. The bottom line works on each one that passed. Try a narrow keyword first, then open If under Executions to see how many videos passed its two tests.

For experts. Duplicates are dropped by a Merge, not a filter. If feeds the videos into its second input and Find Duplicate Entries feeds any matching rows into its first. Keeping the non-matches from the second input leaves only new videos.

The model or service behind each step
StepModel or serviceWhat to know
YouTube Video ScrapeAn Apify YouTube search Actor, called by its IDA synchronous call, which Apify’s API reference ends after 300 seconds
HTTP RequestAn Apify transcript Actor, called by its IDOne run for each video, returning the transcript as one string in captions
Analyze Thumbnailgpt-4o-mini, at high detailDescribes the thumbnail’s layout, colours and text style in one paragraph
YouTube Title Generatorgpt-4.1-mini, with JSON outputReads the old title, the thumbnail description and the transcript
Outline Generatorgpt-4.1Reads the transcript alone

Good to know

Every outlier costs a transcript run on Apify and three model calls, and each outline starts from another creator’s work.

  • Billing: Apify bills the search and one transcript run for each video. OpenAI bills three calls for each, the outline the largest
  • Other people’s work: the transcript and the thumbnail come from someone else’s video. Treat the outline as a starting point and make the video your own
  • Sheet reads: Find Duplicate Entries reads the sheet once for each video that passed, so a big search makes many reads
  • Model dates: OpenAI’s deprecations page, read on 2 November 2026, lists GPT-4o mini, GPT-4.1 and GPT-4.1 mini for removal from the API on 14 January 2027, with gpt-6-luna as their replacement
  • Open form: once published, anyone with the form’s address can start a search on your Apify and OpenAI accounts
  • Your token: the Apify token sits in two headers, so take it out before you share the workflow
  • Credentials: the JSON holds none, and nothing from its author’s sheet

README and files

README

Outlier YouTube videos turned into title and outline ideas

An n8n workflow for planning videos. Send a keyword, and it finds this month's YouTube videos on it that have more
views than their channel has subscribers. For each new one, OpenAI describes the thumbnail and reads the transcript,
then writes a new title, thumbnail text and outline into a Google Sheet.

What is inside

  • workflow.json: the workflow as n8n exports it, eleven nodes and no sticky notes. It holds no credential.
  • README.md: these notes.

What we changed from the original copy

  • The Authorization headers of the two Apify calls held bare placeholders. They now read <YOUR_APIFY_TOKEN>.
  • The author's spreadsheet is gone. The three Google Sheets nodes ask for YOUR_SHEET_ID and find their tab by the
    name Step 1.

Before you start

  • An n8n instance, on n8n Cloud or your own server.
  • An Apify account and its API token, with credit for two Actors the file calls by their IDs: a YouTube search Actor
    (h7sDV53CddomktSi5) and a transcript Actor (1s7eXiaukVuOr4Ueg).
  • An OpenAI API key for gpt-4o-mini, gpt-4.1-mini and gpt-4.1.
  • A Google account, and a Google Sheet with a tab named Step 1 and the headers below.

The sheet

Write these 17 headers in row 1 of Step 1, spelled as shown:

| Filled by | Headers |
|---|---|
| Step 1 Results | input, id, title, url, fromYTUrl, thumbnailUrl, channelName, channelUrl, duration, likes, numberOfSubscribers, viewCount |
| Update Rows | videoTranscript, thumbnailDescription, thumbnailText, newTitle, newOutline |

id is the YouTube video's ID. Both Find Duplicate Entries and Update Rows find rows by it.

Set it up

  1. In n8n, create a workflow, open its menu and choose Import from File. Pick workflow.json.
  2. Make the sheet, then copy its ID: the long part of its address between /d/ and /edit.
  3. Add a Google Sheets OAuth2 credential. In Find Duplicate Entries, Step 1 Results and Update Rows, select it
    and replace YOUR_SHEET_ID with the ID.
  4. Add an OpenAI credential and select it in Analyze Thumbnail, YouTube Title Generator and Outline Generator.
  5. In YouTube Video Scrape and HTTP Request, replace <YOUR_APIFY_TOKEN> in the Authorization header with your
    token. The brackets go too, so the header reads Bearer, a space and the token.
  6. Save, open the form's test address and send one keyword.
  7. Press Publish once the rows look right.

How a run goes

  1. On form submission takes Keyword or Topic.
  2. YouTube Video Scrape asks Apify for up to 100 videos on it: uploaded this month, 4 to 20 minutes long, newest
    first, with no Shorts or streams.
  3. If keeps a video when it has over 1,000 views and more views than its channel has subscribers.
  4. Find Duplicate Entries looks each one up in the sheet by id. Merge keeps only the videos it did not find.
  5. Step 1 Results adds each new video as a row.
  6. Analyze Thumbnail describes the thumbnail in one paragraph, at high detail.
  7. HTTP Request runs the transcript Actor for the video and gets the words back in captions.
  8. YouTube Title Generator writes newTitle, under 70 characters, and thumbnailText, 3 to 5 words.
  9. Outline Generator writes a new outline with a different angle on the same topic.
  10. Update Rows writes the five idea columns into the video's row.

Change the result

  • What counts as an outlier: the two tests in If. Raise 1,000 for a busy niche.
  • What is searched: the JSON body of YouTube Video Scrape, including the date filter, the length filter and
    maxResults.
  • The title and the thumbnail text: the system message of YouTube Title Generator.
  • The outline: the system message of Outline Generator, which asks for a hook, an introduction, main points and a
    call to action.
  • Models: gpt-4o-mini in Analyze Thumbnail, gpt-4.1-mini in YouTube Title Generator, gpt-4.1 in
    Outline Generator.

Good to know

  • Each outlier costs one transcript run on Apify and three OpenAI calls, on top of the search itself.
  • The transcript and the thumbnail come from someone else's video. Use the outline as a starting point and make the
    video your own.
  • Find Duplicate Entries reads the sheet once for each video that passed If, so a big search makes many reads.
  • Apify's API reference gives a synchronous call 300 seconds. A search that takes longer ends without its videos.
  • OpenAI's deprecations page, read on 2 November 2026, lists GPT-4o mini (gpt-4o-mini-2024-07-18), GPT-4.1
    (gpt-4.1-2025-04-14) and GPT-4.1 mini (gpt-4.1-mini-2025-04-14) for removal from the API on 14 January 2027,
    with gpt-6-luna as their replacement. Change all three models before then.
  • Once published, anyone with the form's address can start a search on your Apify and OpenAI accounts.
  • The Apify token sits in two headers, so take it out before you share the workflow.
Files
README.md  5 KB
workflow.json  31 KB

2 files, 35 KB, downloaded as a zip.

What’s new

  • Version 1 · 2026-11-03

Reviews

No ratings yet

Write a review

No reviews yet.

More from Mindket

See all from Mindket

Huashu Design

alchaincyf · Skill

Turns a one-line request in your AI agent into a finished design: a launch animation, a clickable app prototype, an editable slide deck, an infographic or a se…

New29.4k

Guizang PPT Skill

op7418 · Skill

Skill that makes single-file HTML slide decks, slide illustrations and platform covers in two visual styles, with a presenter mode for rehearsal. For speakers…

New29.8k

Claude Code Skills & Plugins

alirezarezvani · Skill

Adds 411 ready-made skills for Claude Code and other coding agents, grouped by role from engineering and product to marketing, finance and compliance. Installs…

New29.8k

Similar in Marketing & SEO

See all in Marketing & SEO

Pixel Art XL

nerijs · Kit

A pixel art style add-on for Stable Diffusion XL that turns text prompts into pixel-style images, with an optional LCM LoRA for faster runs. For game makers, d…

New

Licon MSR for LTX-2.5

LiconStudio · Kit

Keeps characters, clothing, objects and backgrounds consistent in generated video by feeding the model up to five reference images. A LoRA for LTX-2.5 in Comfy…

New

Flux.2-Klein-9B-MatchingPose

nhathoangfoto · Kit

A LoRA for FLUX.2 Klein 9B that puts a character into the pose shown by a mannequin reference while keeping the character's face and style. For artists and des…

New