YouTube niche research and video idea generator

What it is

This n8n workflow studies what already works on YouTube in your niche, then writes three title and thumbnail ideas for your own channel. Three triggers start it. A form takes three competitor channels and records each one’s best recent videos. A weekly and a daily schedule search your niche’s keyword and record the top videos of the week. After the daily search, the workflow reads the comments on your own channel’s newest videos and hands everything to an AI agent. Apify fetches the YouTube data, OpenAI reads titles, thumbnails and comments, and Google Sheets keeps every result.

Who it suits

YouTube creators, and the strategists and agencies who plan videos for them. The ideas come from what is already winning in a niche, so it fits a channel that knows its subject and wants stronger titles and thumbnails. Skip it for a one-off look at a rival: the spreadsheet and the Apify token take longer to set up than three channels take to read by hand.

Where to use it

Any n8n that can reach Apify, OpenAI and Google will run it. The results land in a spreadsheet, where a team already works.

• n8n Cloud: import the file into a new workflow there, and publish it so the schedules run
• Your own n8n server: the same file and steps, with the schedules following the server’s time zone
• Google Sheets: every result lands in one of five tabs, ready to sort, chart or comment on
• A chat or a board: add a node after Append Ideas to send the three ideas on

Setup

Get it saves a zip with workflow.json and Mindket’s setup notes in your downloads folder. Make the spreadsheet first, since six nodes point at it. Its author puts setup at about 13 minutes.

1. In n8n, create a workflow, open its menu, choose Import from File and pick workflow.json.
2. Make the spreadsheet with the five tabs, then copy its ID: the long part of its address between /d/ and /edit.
3. Add a Google Sheets OAuth2 credential. In each of the six Google Sheets nodes, select it and replace YOUR_SHEET_ID with the ID. Each node finds its tab by name.
4. Add an OpenAI credential and select it in the five OpenAI Chat Model nodes and in Analyze Thumbnails, Analyze Thumbnails1 and Analyze Thumbnails2.
5. In the five HTTP Request nodes that call api.apify.com, 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. In Broad Niche and Niche, type the words to search for. The file ships with wordpress and n8n.
7. In Channel URL, enter your channel’s address. In Channel Description, replace the placeholder with a few sentences about the channel.
8. Check the times in the Sundays and 6 am schedule nodes, save, send the form once from its test address, then press Publish.

How to use it

Open the form from On form submission and enter three channels from your niche. All three fields are required. The form starts only the first phase. The weekly and daily phases need no one: once the workflow is published they start on their own at the hours set in their schedule nodes, and each saves to its own tab.

Example

You run a woodworking channel. You send the form three larger woodworking channels, set Niche to a search such as beginner woodworking, and fill in Channel URL and Channel Description. The next morning, Niche Daily holds five rows dated that day. Each row has a view count and one to three power words from the title, and a short note says why the thumbnail pulls the eye. Your viewers’ three lists are in Comment Analysis. The agent’s reply, saved in Ideation, has this shape:

{
  "title_1": "…",
  "thumbnail_1": "…",
  "title_2": "…",
  "thumbnail_2": "…",
  "title_3": "…",
  "thumbnail_3": "…"
}

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. This file has 48 of them, but five are sticky notes that only label the phases. A credential is a key n8n stores for you and hands to whichever node you select it in. Send the form first. Each node then shows what it received and sent, so a wrong key or a mistyped tab name shows up at once.

For experts. Two kinds of OpenAI model do the reading, and Apify does the fetching through two actors it names by ID in each request.

One actor lists videos for a search or a channel, and the other reads the comments of a video. Each request waits for its actor to finish, and the actor’s own page on Apify sets its price per result.

Good to know

Every phase spends money on Apify and OpenAI, and the daily branch repeats work.

• Billing: Apify bills the actor runs and OpenAI bills the model calls, for each run of each phase
• Repeated calls: Niche Daily passes its five rows on, so Scrape Channel runs once per row and Get Comments once per video found. Switch on Execute Once in the settings of Channel URL to run phases 4 and 5 a single time
• Open form: once the workflow is published, anyone with the form’s address can start phase 1 on your tokens, so turn on the form trigger’s authentication or keep the address private
• Older channels: Filter keeps only videos from the last six months, so a quiet channel adds fewer than ten rows
• Where data goes: channel addresses and search words go to Apify, and titles, thumbnail addresses and your viewers’ comments go to OpenAI and Google
• Credentials: the JSON holds none, and nothing from its author’s spreadsheet or channel