How to Automate Content Briefs with n8n

1 Trigger on a new keyword row

The Google Sheets Trigger node in n8n polls a spreadsheet and starts the workflow whenever a row is added. Each new keyword becomes one workflow execution, so the brief appears without anyone clicking a button.

Add a Google Sheets Trigger node to a new n8n workflow and configure:

Node: Google Sheets Trigger
Event: Row Added
Document: By ID (your spreadsheet ID)
Sheet: Keywords
Poll Times: Every Minute
Options > Columns to Watch: keyword

Polling every minute is fine for a small sheet. Every poll counts toward Google API quota, so use Every 15 Minutes for busy documents in n8n.

{
  "row_number": 14,
  "keyword": "n8n error handling",
  "intent": "informational",
  "status": ""
}

2 Request SERP results

The HTTP Request node calls your SERP API. The parameter names differ by provider, but most accept a query, a location, and a result count. The example below uses a generic structure you can adapt.

Add an HTTP Request node to the n8n canvas and configure:

Node: HTTP Request
Method: GET
URL: https://api.your-serp-provider.com/search
Authentication: Generic Credential Type > Header Auth
Send Query Parameters: On
q: {{ $json.keyword }}
num: 10
gl: us
hl: en
Response Format: JSON

Store the key in an n8n Header Auth credential rather than typing it into the node, so it does not leak when you export the n8n workflow JSON.

{
  "organic_results": [
    { "position": 1, "title": "Error Handling in n8n", "link": "https://example.com/error-handling", "snippet": "Use the Error Trigger node..." },
    { "position": 2, "title": "Retry On Fail Guide", "link": "https://example.org/retry", "snippet": "Configure retries..." }
  ]
}

3 Fetch each result page

Snippets are too thin for analysis, so the n8n workflow needs the full pages. First add a Split Out node to turn the results array into one item per result, then another HTTP Request node to download each page.

Node: Split Out
Fields To Split Out: organic_results
Include: No Other Fields

Node: HTTP Request (page fetch)
Method: GET
URL: {{ $json.link }}
Response Format: Text
Put Output in Field: html
Options > Batching > Items per Batch: 2
Options > Batching > Batch Interval: 1500
Settings > On Error: Continue

Some pages will block the request or time out. On Error: Continue lets n8n carry on with whatever pages it did retrieve, and the Code node in the next step skips empty HTML.

4 Extract headings and word counts

An n8n Code node reads each page, pulls out H2 and H3 headings, and estimates the visible word count. Using regular expressions avoids installing extra packages on your n8n server.

Add a Code node and configure:

Node: Code
Mode: Run Once for Each Item
Language: JavaScript

const html = $json.html || '';
const body = html.replace(/<script[\s\S]*?<\/script>/gi, ' ')
                 .replace(/<style[\s\S]*?<\/style>/gi, ' ');
const headings = [];
const re = /<h([23])[^>]*>([\s\S]*?)<\/h\1>/gi;
let m;
while ((m = re.exec(body)) === null ? false : true) {
  const t = m[2].replace(/<[^>]+>/g, '').replace(/\s+/g, ' ').trim();
  if (t) headings.push({ level: Number(m[1]), text: t });
}
const words = body.replace(/<[^>]+>/g, ' ').split(/\s+/).filter(Boolean).length;
return { json: { url: $json.link, title: $json.title, headings, wordCount: words } };

The backreference \1 in the regular expression makes sure an opening h2 only pairs with a closing h2.

{
  "url": "https://example.com/error-handling",
  "title": "Error Handling in n8n",
  "headings": [
    { "level": 2, "text": "Using the Error Trigger node" },
    { "level": 2, "text": "Retry On Fail settings" }
  ],
  "wordCount": 1820
}

5 Find common topics and gaps

Now an aggregate pass in n8n. Another Code node in Run Once for All Items mode counts how many competing pages share each heading keyword. Topics that appear on at least three pages are must cover subjects. Topics that appear on only one page are gap opportunities.

Node: Code
Mode: Run Once for All Items

const stop = new Set(['the','a','and','to','of','in','for','how','with','is','your','what']);
const counts = {};
const pages = $input.all();
for (const p of pages) {
  const seen = new Set();
  for (const h of p.json.headings) {
    for (const w of h.text.toLowerCase().split(/[^a-z0-9]+/)) {
      if (w.length > 3 && stop.has(w) === false) seen.add(w);
    }
  }
  seen.forEach(w => { counts[w] = (counts[w] || 0) + 1; });
}
const common = Object.keys(counts).filter(w => counts[w] >= 3).sort((a, b) => counts[b] - counts[a]).slice(0, 12);
const gaps = Object.keys(counts).filter(w => counts[w] === 1).slice(0, 8);
const wc = pages.map(p => p.json.wordCount).filter(n => n > 200).sort((a, b) => a - b);
const median = wc[Math.floor(wc.length / 2)] || 0;
return [{ json: { common, gaps, median, urls: pages.map(p => p.json.url), headings: pages.flatMap(p => p.json.headings.map(h => h.text)).slice(0, 20) } }];

In n8n, counting a word once per page, using the seen set, prevents one long article from dominating the tally.

6 Structure the brief with Set

The Set node (called Edit Fields in recent n8n versions) shapes the final brief fields. Reference the keyword from the trigger by node name, because the data on the main path now comes from the Code node.

Add a Set node and configure:

Node: Set (Edit Fields)
Mode: Manual Mapping
targetKeyword: {{ $('Google Sheets Trigger').item.json.keyword }}
suggestedHeadings: {{ $json.headings.join(' | ') }}
mustCoverTopics: {{ $json.common.join(', ') }}
gapTopics: {{ $json.gaps.join(', ') }}
wordCountRange: {{ Math.round($json.median * 0.9) }} to {{ Math.round($json.median * 1.2) }}
competitorUrls: {{ $json.urls.join('\n') }}
createdAt: {{ $now.toISO() }}

{
  "targetKeyword": "n8n error handling",
  "mustCoverTopics": "error, trigger, retry, workflow",
  "gapTopics": "alerts, sentry",
  "wordCountRange": "1638 to 2184",
  "createdAt": "2026-11-04T09:30:00.000-08:00"
}

7 Write the brief to the output sheet

The final n8n Google Sheets node appends the brief. A follow up node can then set the original keyword row status to done.

Node: Google Sheets
Resource: Sheet Within Document
Operation: Append Row
Sheet: Briefs
Mapping Column Mode: Map Automatically

Because the Set node field names match the sheet headers, Map Automatically fills every column. Writers can open the Briefs tab fed by n8n and start drafting straight away.