Capture Social Dms with Purchase Intent Using Fanbase, Apify and Google Sheets

The workflow JSON

Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →

{
  "id": "gOySWYooKXWbepYL",
  "name": "Turn Social Engagement into Qualified Leads with FanBase MCP, Apify and Sheets",
  "tags": [],
  "nodes": [
    {
      "id": "22167bc8-dcba-499e-9665-49cd4bb24b5f",
      "name": "Every 15 Minutes Trigger",
      "type": "n8n-nodes-base.scheduleTrigger",
      "position": [
        -816,
        640
      ],
      "parameters": {
        "rule": {
          "interval": [
            {
              "field": "minutes",
              "minutesInterval": 15
            }
          ]
        }
      },
      "typeVersion": 1.3
    },
    {
      "id": "79dfccc2-496e-4e42-adb3-d5f3cdfa3561",
      "name": "Retrieve Watermark Data",
      "type": "n8n-nodes-base.dataTable",
      "position": [
        -608,
        640
      ],
      "parameters": {
        "filters": {
          "conditions": [
            {
              "keyName": "key",
              "keyValue": "funnel_last_ts"
            }
          ]
        },
        "matchType": "allConditions",
        "operation": "get",
        "returnAll": true,
        "dataTableId": {
          "__rl": true,
          "mode": "list",
          "value": "REPLACE_WITH_YOUR_DATA_TABLE",
          "cachedResultName": "time_stamp"
        }
      },
      "typeVersion": 1.1,
      "alwaysOutputData": true
    },
    {
      "id": "1f11c635-dfaa-4d16-b9e0-b4e332705756",
      "name": "Fetch Direct Messages",
      "type": "@n8n/n8n-nodes-langchain.mcpClient",
      "position": [
        -432,
        640
      ],
      "parameters": {
        "tool": {
          "__rl": true,
          "mode": "id",
          "value": "query_activity"
        },
        "options": {},
        "inputMode": "json",
        "jsonInput": "={\n  \"after\": \"{{ $json.value ? new Date(Number($json.value) * 1000).toISOString() : new Date(Date.now() - 86400000).toISOString() }}\",\n  \"limit\": 100,\n  \"sortDirection\": \"desc\"\n}",
        "endpointUrl": "https://api.copilot.fanbase.gg/mcp",
        "authentication": "mcpOAuth2Api"
      },
      "typeVersion": 1.1
    },
    {
      "id": "d0929153-6a0a-443b-981a-9d50db46a1a0",
      "name": "Identify Intent in DMs",
      "type": "n8n-nodes-base.code",
      "position": [
        -240,
        640
      ],
      "parameters": {
        "jsCode": "\n// DMs with purchase intent since the last poll. Emits ONE item: { watermark, leads[] }.\n// DMs only \u2014 likes, comments, follows and mentions are ignored.\nconst raw = $input.first().json;\nlet payload = (raw && raw.content && raw.content[0]) ? raw.content[0].text : raw;\nif (typeof payload === 'string') { try { payload = JSON.parse(payload); } catch (e) { payload = null; } }\nconst events = (payload && Array.isArray(payload.events)) ? payload.events : [];\nconst apiTotal = (payload && typeof payload.total === 'number') ? payload.total : events.length;\nconst organization_id = (payload && payload.organizationId) ? String(payload.organizationId) : '';\n\nconst prevTs = Number(($('Retrieve Watermark Data').first().json || {}).value || 0) || 0;\n\nconst DM_TYPES = ['message', 'dm', 'direct_message'];\n\n// word-boundary matcher \u2014 stops \"order\" matching \"border\", \"cost\" matching \"costume\"\nfunction boundary(kw) {\n  const esc = String(kw).replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\$&');\n  try { return new RegExp('(?<![\\\\p{L}\\\\p{N}])' + esc + '(?![\\\\p{L}\\\\p{N}])', 'iu'); }\n  catch (e) { return new RegExp(esc, 'i'); }\n}\nconst INTENT = ['buy','buying','purchase','purchasing','order','preorder','pre-order','price','pricing',\n  'cost','checkout','payment','how much','where can i buy','where can i get','in stock','restock',\n  'sold out','discount','promo code','sign up','subscribe','upgrade','interested','want one','get one'];\nconst intentRes = INTENT.map(boundary);\n\nfunction igThreadId(messageId) {\n  if (!messageId) return null;\n  const s = String(messageId);\n  const cands = s.indexOf('ZA') === -1 ? [s] : [s, s.replace(/ZA/g, 'Z')];\n  for (const c of cands) for (let p = 0; p < 4; p++) {\n    let dec = '';\n    try { dec = Buffer.from(c + '='.repeat(p), 'base64').toString('utf8'); } catch (e) { continue; }\n    const m = dec.match(/^ig_dm_item:\\d+:IGMessageID:\\d+:(\\d+):\\d+/);\n    if (m) return m[1];\n  }\n  return null;\n}\nfunction replyUrl(ev) {\n  const inbox = organization_id ? 'https://copilot.fanbase.gg/o/' + organization_id + '/inbox' : '';\n  const p = String(ev.platform || '').toLowerCase();\n  const meta = ev.metadata || {};\n  if (p === 'instagram') {\n    const tid = igThreadId(meta.messageId);\n    if (tid) return 'https://www.instagram.com/direct/t/' + tid + '/';\n  }\n  return inbox;\n}\nconst secs = (ev) => {\n  const t = new Date((ev.metadata && ev.metadata.timestamp) || ev.createdAt || '').getTime();\n  return Number.isFinite(t) ? Math.floor(t / 1000) : 0;\n};\n\nlet newest = prevTs;\nconst leads = [];\nconst seen = new Set();\n\nfor (const ev of events.slice().sort((a, b) => secs(b) - secs(a))) {   // newest first\n  const ts = secs(ev);\n  if (ts > newest) newest = ts;\n  if (ts && ts <= prevTs) continue;                                    // already handled\n  if (DM_TYPES.indexOf(String(ev.type || '').toLowerCase()) === -1) continue;   // DMs only\n\n  const text = (ev.metadata && ev.metadata.text) ? String(ev.metadata.text) : '';\n  if (!intentRes.some(re => re.test(text))) continue;\n\n  const fanId = (ev.fan && ev.fan.id) ? String(ev.fan.id) : '';\n  if (!fanId || seen.has(fanId)) continue;                             // one lead per fan per poll\n  seen.add(fanId);\n\n  leads.push({\n    fan_id: fanId,\n    fan_name: (ev.fan && ev.fan.name) || '',\n    platform: ev.platform || '',\n    intent_text: text.slice(0, 500),\n    intent_at: (ev.metadata && ev.metadata.timestamp) || ev.createdAt || '',\n    reply_url: replyUrl(ev),\n  });\n}\n\nif (!newest) newest = Math.floor(Date.now() / 1000);\nreturn [{ json: { watermark: String(newest), leads, lead_count: leads.length,\n                  scanned: events.length, api_truncated: apiTotal > events.length } }];\n"
      },
      "typeVersion": 2
    },
    {
      "id": "9521dc06-a6e2-48fe-8c5b-1a8dfcf0e989",
      "name": "Store Watermark Data",
      "type": "n8n-nodes-base.dataTable",
      "position": [
        80,
        320
      ],
      "parameters": {
        "columns": {
          "value": {
            "key": "funnel_last_ts",
            "value": "={{ $json.watermark }}"
          },
          "schema": [
            {
              "id": "key",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "key",
              "defaultMatch": false
            },
            {
              "id": "value",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "value",
              "defaultMatch": false
            }
          ],
          "mappingMode": "defineBelow",
          "matchingColumns": [],
          "attemptToConvertTypes": false,
          "convertFieldsToString": false
        },
        "filters": {
          "conditions": [
            {
              "keyName": "key",
              "keyValue": "funnel_last_ts"
            }
          ]
        },
        "options": {},
        "matchType": "allConditions",
        "operation": "upsert",
        "dataTableId": {
          "__rl": true,
          "mode": "list",
          "value": "REPLACE_WITH_YOUR_DATA_TABLE",
          "cachedResultName": "time_stamp"
        }
      },
      "typeVersion": 1.1
    },
    {
      "id": "b0c8055a-4fa1-41b1-801c-641eb2c0f363",
      "name": "Process CRM Page 1",
      "type": "@n8n/n8n-nodes-langchain.mcpClient",
      "position": [
        32,
        576
      ],
      "parameters": {
        "tool": {
          "__rl": true,
          "mode": "id",
          "value": "list_crm"
        },
        "options": {},
        "inputMode": "json",
        "jsonInput": "{\n  \"page\": 1,\n  \"limit\": 100,\n  \"sortBy\": \"engagement\"\n}",
        "endpointUrl": "https://api.copilot.fanbase.gg/mcp",
        "authentication": "mcpOAuth2Api"
      },
      "typeVersion": 1.1
    },
    {
      "id": "3fdf04cb-6641-4674-b416-b68f178045ea",
      "name": "Process CRM Page 2",
      "type": "@n8n/n8n-nodes-langchain.mcpClient",
      "position": [
        32,
        752
      ],
      "parameters": {
        "tool": {
          "__rl": true,
          "mode": "id",
          "value": "list_crm"
        },
        "options": {},
        "inputMode": "json",
        "jsonInput": "{\n  \"page\": 2,\n  \"limit\": 100,\n  \"sortBy\": \"engagement\"\n}",
        "endpointUrl": "https://api.copilot.fanbase.gg/mcp",
        "authentication": "mcpOAuth2Api"
      },
      "typeVersion": 1.1
    },
    {
      "id": "712db4c9-5d6f-4ef9-964e-f446fb502952",
      "name": "Merge CRM Page Data",
      "type": "n8n-nodes-base.merge",
      "position": [
        288,
        672
      ],
      "parameters": {},
      "typeVersion": 3.2
    },
    {
      "id": "409cd450-4c9f-4e52-9eb2-400080a3e389",
      "name": "Resolve Social Handles",
      "type": "n8n-nodes-base.code",
      "position": [
        512,
        672
      ],
      "parameters": {
        "jsCode": "\n// Joins each intent lead to its CRM record to get the handle, then emits ONE ITEM PER LEAD.\n// fan.id (query_activity) == clusterId (list_crm) \u2014 verified against live data.\n// list_crm caps at 100 rows per page and page 1 is all high-engagement twitter fans, so the\n// Instagram fans who actually DM only appear on later pages \u2014 hence the merged page inputs.\nconst leads = ($('Identify Intent in DMs').first().json.leads) || [];\nif (!leads.length) return [];\n\nconst fans = [];\nlet crmTotal = 0;\nfor (const item of $input.all()) {\n  let p = (item.json && item.json.content && item.json.content[0]) ? item.json.content[0].text : item.json;\n  if (typeof p === 'string') { try { p = JSON.parse(p); } catch (e) { continue; } }\n  if (!p) continue;\n  if (typeof p.total === 'number') crmTotal = Math.max(crmTotal, p.total);\n  for (const f of (p.fans || [])) fans.push(f);\n}\nconst byId = {};\nfor (const f of fans) if (f && f.clusterId) byId[String(f.clusterId)] = f;\n\nfunction profileUrl(platform, u) {\n  if (!u) return '';\n  const p = String(platform || '').toLowerCase();\n  if (p === 'instagram') return 'https://www.instagram.com/' + u;\n  if (p === 'twitter' || p === 'x') return 'https://x.com/' + u;\n  if (p === 'tiktok') return 'https://www.tiktok.com/@' + u;\n  return '';\n}\n\nreturn leads.map(lead => {\n  const fan = byId[String(lead.fan_id)] || null;\n  const profile = (fan && fan.profile) || {};\n  return { json: Object.assign({}, lead, {\n    handle: String(profile.username || ''),\n    profile_url: profileUrl(lead.platform, profile.username),\n    crm_truncated: crmTotal > fans.length,\n  }) };\n});\n"
      },
      "typeVersion": 2
    },
    {
      "id": "1fd8093c-0f31-4f59-a80f-c024fc266f40",
      "name": "Assemble CRM Data Row",
      "type": "n8n-nodes-base.code",
      "position": [
        1056,
        672
      ],
      "parameters": {
        "jsCode": "// Merges each Apify profile with its lead and produces ONE ROW PER LEAD.\n// The Apify node (\"Run an Actor and get dataset\") emits dataset items FLAT \u2014 there is no MCP\n// content[0].text envelope \u2014 and it emits NOTHING for a run that returns 0 items (bad/renamed\n// handle), so rows are driven off the leads and matched to profiles by username, never by\n// position. pairedItem (set by the Apify node to the input item index) is the fallback.\nconst leads = $('Resolve Social Handles').all().map(i => i.json);\nconst results = $input.all();\n\nfunction unwrap(item) {\n  const raw = (item && item.json) || {};\n  if (Array.isArray(raw.content)) {                 // MCP shape, if that node is ever swapped back in\n    for (const blk of raw.content) {\n      let o = blk && blk.text;\n      if (typeof o === 'string') { try { o = JSON.parse(o); } catch (e) { continue; } }\n      if (o && Array.isArray(o.items) && o.items.length) return o.items[0];\n      if (Array.isArray(o) && o.length && o[0] && o[0].username) return o[0];\n      if (o && (o.username || o.biography)) return o;\n    }\n    return null;\n  }\n  return (raw.username || raw.biography) ? raw : null;   // flat dataset item\n}\n\nconst norm = (u) => String(u || '').trim().replace(/^@/, '').toLowerCase();\nconst slot = (item, i) => {\n  const p = item && item.pairedItem;\n  if (typeof p === 'number') return p;\n  if (p && typeof p === 'object' && typeof p.item === 'number') return p.item;\n  if (Array.isArray(p) && p.length && typeof p[0].item === 'number') return p[0].item;\n  return i;\n};\n\nconst byHandle = {}, byIndex = {}, errByIndex = {};\nresults.forEach((item, i) => {\n  const idx = slot(item, i);\n  const profile = unwrap(item);\n  if (!profile) {\n    const err = item && item.json && item.json.error;\n    if (err && errByIndex[idx] === undefined) errByIndex[idx] = String(err.message || err);\n    return;\n  }\n  const key = norm(profile.username);\n  if (key && byHandle[key] === undefined) byHandle[key] = profile;\n  if (byIndex[idx] === undefined) byIndex[idx] = profile;\n});\n\nconst EMAIL = /[A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Za-z]{2,}/;\nfunction emailOf(profile, bio) {\n  if (!profile) return '';\n  // Instagram only exposes a contact email on some business accounts, so the bio is the main source.\n  for (const v of [profile.businessEmail, profile.publicEmail, profile.email]) {\n    const m = String(v || '').match(EMAIL);\n    if (m) return m[0];\n  }\n  const inBio = String(bio || '').match(EMAIL);\n  if (inBio) return inBio[0];\n  for (const l of (Array.isArray(profile.externalUrls) ? profile.externalUrls : [])) {\n    const u = String((l && l.url) || '');\n    if (/^mailto:/i.test(u)) {\n      const m = u.slice(7).match(EMAIL);\n      if (m) return m[0];\n    }\n  }\n  return '';\n}\n\nreturn leads.map((lead, i) => {\n  const profile = byHandle[norm(lead.handle)] || byIndex[i] || null;\n  const bio = profile ? String(profile.biography || '') : '';\n\n  let note = '';\n  if (!lead.handle) note = 'no handle in CRM' + (lead.crm_truncated ? ' (CRM paging incomplete)' : '');\n  else if (!profile) note = errByIndex[i] ? 'scrape failed: ' + errByIndex[i] : 'scrape returned nothing';\n  else if (profile.private) note = 'private account';\n\n  return { json: {\n    captured_at: new Date().toISOString(),\n    fan_name: lead.fan_name || (profile ? profile.fullName : '') || '',\n    handle: lead.handle || (profile ? String(profile.username || '') : ''),\n    profile_url: lead.profile_url || (profile ? String(profile.url || '') : ''),\n    platform: lead.platform || '',\n    email: emailOf(profile, bio),\n    link_in_bio: profile ? String(profile.externalUrl || '') : '',\n    followers: profile && typeof profile.followersCount === 'number' ? profile.followersCount : '',\n    posts: profile && typeof profile.postsCount === 'number' ? profile.postsCount : '',\n    is_business: profile ? !!profile.isBusinessAccount : '',\n    bio: bio.replace(/\\s+/g, ' ').slice(0, 300),\n    intent_text: lead.intent_text || '',\n    intent_at: lead.intent_at || '',\n    reply_url: lead.reply_url || '',\n    fan_id: lead.fan_id || '',\n    status: note ? 'captured (' + note + ')' : 'captured',\n  } };\n});\n"
      },
      "typeVersion": 2
    },
    {
      "id": "4a815cd6-410a-4a96-9f7c-602e5dc184ab",
      "name": "Insert to CRM Sheet",
      "type": "n8n-nodes-base.googleSheets",
      "position": [
        1392,
        672
      ],
      "parameters": {
        "columns": {
          "value": {
            "bio": "={{ $json.bio }}",
            "email": "={{ $json.email }}",
            "posts": "={{ $json.posts }}",
            "fan_id": "={{ $json.fan_id }}",
            "handle": "={{ $json.handle }}",
            "status": "={{ $json.status }}",
            "fan_name": "={{ $json.fan_name }}",
            "platform": "={{ $json.platform }}",
            "followers": "={{ $json.followers }}",
            "intent_at": "={{ $json.intent_at }}",
            "reply_url": "={{ $json.reply_url }}",
            "captured_at": "={{ $json.captured_at }}",
            "intent_text": "={{ $json.intent_text }}",
            "is_business": "={{ $json.is_business }}",
            "link_in_bio": "={{ $json.link_in_bio }}",
            "profile_url": "={{ $json.profile_url }}"
          },
          "schema": [
            {
              "id": "captured_at",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "captured_at",
              "defaultMatch": false
            },
            {
              "id": "fan_name",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "fan_name",
              "defaultMatch": false
            },
            {
              "id": "handle",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "handle",
              "defaultMatch": false
            },
            {
              "id": "profile_url",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "profile_url",
              "defaultMatch": false
            },
            {
              "id": "platform",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "platform",
              "defaultMatch": false
            },
            {
              "id": "email",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "email",
              "defaultMatch": false
            },
            {
              "id": "link_in_bio",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "link_in_bio",
              "defaultMatch": false
            },
            {
              "id": "followers",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "followers",
              "defaultMatch": false
            },
            {
              "id": "posts",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "posts",
              "defaultMatch": false
            },
            {
              "id": "is_business",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "is_business",
              "defaultMatch": false
            },
            {
              "id": "bio",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "bio",
              "defaultMatch": false
            },
            {
              "id": "intent_text",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "intent_text",
              "defaultMatch": false
            },
            {
              "id": "intent_at",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "intent_at",
              "defaultMatch": false
            },
            {
              "id": "reply_url",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "reply_url",
              "defaultMatch": false
            },
            {
              "id": "fan_id",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "fan_id",
              "defaultMatch": false
            },
            {
              "id": "status",
              "type": "string",
              "display": true,
              "removed": false,
              "readOnly": false,
              "required": false,
              "displayName": "status",
              "defaultMatch": false
            }
          ],
          "mappingMode": "defineBelow",
          "matchingColumns": [
            "fan_id"
          ],
          "attemptToConvertTypes": false,
          "convertFieldsToString": true
        },
        "options": {},
        "operation": "appendOrUpdate",
        "sheetName": {
          "__rl": true,
          "mode": "name",
          "value": "Leads"
        },
        "documentId": {
          "__rl": true,
          "mode": "url",
          "value": "https://docs.google.com/spreadsheets/d/REPLACE_WITH_YOUR_SHEET_URL/edit"
        }
      },
      "typeVersion": 4.5
    },
    {
      "id": "e4954212-0900-415a-9d18-0a8bf8c7c214",
      "name": "Execute Actor for Lead Data",
      "type": "@apify/n8n-nodes-apify.apify",
      "onError": "continueRegularOutput",
      "position": [
        784,
        672
      ],
      "parameters": {
        "actorId": {
          "__rl": true,
          "mode": "list",
          "value": "dSCLg0C3YEZ83HzYX",
          "cachedResultUrl": "https://console.apify.com/actors/dSCLg0C3YEZ83HzYX/input",
          "cachedResultName": "Instagram Profile Scraper (apify/instagram-profile-scraper)"
        },
        "operation": "Run actor and get dataset",
        "customBody": "={\n    \"usernames\": [\n        \"{{$json.handle}}\"\n    ]\n} "
      },
      "typeVersion": 1,
      "alwaysOutputData": true
    },
    {
      "id": "34fb08d5-1d02-44c1-969f-b9fac95cb373",
      "name": "Overview",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -1424,
        160
      ],
      "parameters": {
        "color": 5,
        "width": 480,
        "height": 1236,
        "content": "## GTM Engineering: Turn Social Engagement into Qualified Leads with FanBase MCP, Apify and Sheets\n\n### How it works\n\nEvery 15 minutes this workflow pulls the FanBase DMs received since the last saved watermark and keeps only the ones showing **purchase intent** \u2014 likes, comments, follows and mentions are dropped. A DM only carries a display name, so it looks the fan up in the CRM (`fan.id == clusterId`) to get their @handle, scrapes that public profile with Apify, and appends one enriched, CRM-ready row per lead to a Google Sheet. The watermark is then advanced so the next run only sees new engagement.\n\nEach row carries the DM that triggered it and a deep link straight back into the conversation, next to the enrichment \u2014 so you can reply while the fan is still in the chat.\n\n### Setup steps\n\n- Create a **Data Table** with two string columns, `key` and `value`, and select it on *Retrieve Watermark Data* and *Store Watermark Data*. Nothing to seed \u2014 the first run scans the last 24 h and writes the row itself.\n- Connect the **FanBase MCP** credential (`https://api.copilot.fanbase.gg/mcp`) on the three MCP nodes.\n- Add your **Apify** credential on *Execute Actor for Lead Data*. The actor (`apify/instagram-profile-scraper`) is already selected.\n- Create a Google Sheet with a **`Leads`** tab whose header row is exactly:\n`captured_at, fan_name, handle, profile_url, platform, email, link_in_bio, followers, posts, is_business, bio, intent_text, intent_at, reply_url, fan_id, status`\nthen connect **Google Sheets** and paste the sheet URL into *Insert to CRM Sheet*.\n- Adjust the 15-minute schedule to your DM volume, then run once manually before activating.\n\n### Customization\n\nEdit the keyword lexicon in *Identify Intent in DMs* to match how your audience asks to buy \u2014 or add partnership/collab terms to catch inbound deals too. Swap the Apify actor for another network, add CRM pages 3+ if you have more than 200 fans, and repoint the final node at HubSpot, Airtable or Notion: *Assemble CRM Data Row* emits a flat object, so nothing upstream changes."
      },
      "typeVersion": 1
    },
    {
      "id": "d1075926-8f1b-4098-b87e-8482ef310f7f",
      "name": "Section 1",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        -864,
        112
      ],
      "parameters": {
        "width": 784,
        "height": 426,
        "content": "## 1 \u00b7 Scheduled DM intake\n\n**Every 15 Minutes Trigger** \u2014 schedule. Nothing here needs to be realtime, and a wider interval keeps the API calls cheap.\n\n**Retrieve Watermark Data** \u2014 reads the `funnel_last_ts` row: the timestamp of the newest DM handled last poll, so no message is ever processed twice. *Always Output Data* is on, so a missing row on the first run can't stall the chain.\n\n**Fetch Direct Messages** \u2014 FanBase MCP `query_activity`, asking only for activity *after* that watermark (last 24 h on the first run).\n\n**Identify Intent in DMs** \u2014 Code. Keeps DMs only, matches a 26-keyword intent lexicon with word boundaries so \"border\" never matches \"order\", takes one lead per fan per poll, and rebuilds a deep link to the conversation from the message id. Emits one item: `{ watermark, leads[] }`."
      },
      "typeVersion": 1
    },
    {
      "id": "8e8131b6-1025-47f3-887c-42b43aa9205d",
      "name": "Section 1b",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        0,
        0
      ],
      "parameters": {
        "width": 524,
        "height": 286,
        "content": "## Advance the watermark\n\n**Store Watermark Data** \u2014 upserts the newest DM timestamp so the next poll starts where this one stopped.\n\nIt runs in parallel with the enrichment branch. Move it after *Insert to CRM Sheet* if you would rather re-scan leads whose row failed to write."
      },
      "typeVersion": 1
    },
    {
      "id": "191ab72a-ae83-47c4-b593-ed2d6a22484f",
      "name": "Section 2",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        0,
        912
      ],
      "parameters": {
        "width": 716,
        "height": 388,
        "content": "## 2 \u00b7 Turn the fan into a scrapeable handle\n\nA DM event carries only `fan: {id, name}` \u2014 a display name, no @handle. The handle lives in the CRM, keyed by `clusterId`, and `fan.id == clusterId`.\n\n**Process CRM Page 1 / Page 2** \u2014 FanBase MCP `list_crm`. The API caps at 100 rows per page and ignores a larger limit, and page 1 is dominated by your most-engaged fans, so a single page can resolve nobody. Add page 3+ for a larger CRM.\n\n**Merge CRM Page Data** \u2014 appends both pages into one stream. Two edges into one input port would *not* merge \u2014 n8n counts input ports \u2014 so each page gets its own port.\n\n**Resolve Social Handles** \u2014 Code. Indexes fans by `clusterId`, joins each lead to its `profile.username`, and builds the profile URL for that platform (Instagram, X or TikTok). A lead with no CRM match is kept and flagged, not dropped."
      },
      "typeVersion": 1
    },
    {
      "id": "ffe6ff6e-6f72-4a16-a00e-2c9407ffca51",
      "name": "Section 3",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        736,
        48
      ],
      "parameters": {
        "height": 572,
        "content": "## 3 \u00b7 Scrape lead profiles\n\n**Execute Actor for Lead Data** \u2014 Apify, running `apify/instagram-profile-scraper` with `{\"usernames\": [\"{{ $json.handle }}\"]}`. Runs once per lead, waits, and returns the dataset item flat: `biography`, `followersCount`, `postsCount`, `externalUrl`, `private`\u2026\n\n*On Error \u2192 Continue* so one bad or renamed handle can't kill the poll, and *Always Output Data* so rows are still written when every scrape comes back empty."
      },
      "typeVersion": 1
    },
    {
      "id": "1ba488f2-fca2-45b8-89f9-58161d7de097",
      "name": "Section 4",
      "type": "n8n-nodes-base.stickyNote",
      "position": [
        1040,
        48
      ],
      "parameters": {
        "width": 560,
        "height": 560,
        "content": "## 4 \u00b7 Build and save rows\n\n**Assemble CRM Data Row** \u2014 Code. One row per lead, matched to its scraped profile **by username, never by position** \u2014 the Apify node emits nothing at all for a profile it could not scrape, so index pairing would put one lead's data on another lead's row. The email is read from `businessEmail`/`publicEmail`, then the bio text, then a `mailto:` bio link \u2014 and left blank rather than invented. Every lead reaches the sheet, with a `status` note when the handle or the scrape is missing.\n\n**Insert to CRM Sheet** \u2014 Google Sheets *append or update* matching on `fan_id`, so a fan who DMs again updates their row instead of duplicating it."
      },
      "typeVersion": 1
    }
  ],
  "active": false,
  "settings": {
    "binaryMode": "separate",
    "executionOrder": "v1"
  },
  "versionId": "07437fb9-a3b1-4f3c-9b53-64ba230dba09",
  "nodeGroups": [],
  "connections": {
    "Process CRM Page 1": {
      "main": [
        [
          {
            "node": "Merge CRM Page Data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Process CRM Page 2": {
      "main": [
        [
          {
            "node": "Merge CRM Page Data",
            "type": "main",
            "index": 1
          }
        ]
      ]
    },
    "Merge CRM Page Data": {
      "main": [
        [
          {
            "node": "Resolve Social Handles",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Assemble CRM Data Row": {
      "main": [
        [
          {
            "node": "Insert to CRM Sheet",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Fetch Direct Messages": {
      "main": [
        [
          {
            "node": "Identify Intent in DMs",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Identify Intent in DMs": {
      "main": [
        [
          {
            "node": "Store Watermark Data",
            "type": "main",
            "index": 0
          },
          {
            "node": "Process CRM Page 1",
            "type": "main",
            "index": 0
          },
          {
            "node": "Process CRM Page 2",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Resolve Social Handles": {
      "main": [
        [
          {
            "node": "Execute Actor for Lead Data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Retrieve Watermark Data": {
      "main": [
        [
          {
            "node": "Fetch Direct Messages",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Every 15 Minutes Trigger": {
      "main": [
        [
          {
            "node": "Retrieve Watermark Data",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Execute Actor for Lead Data": {
      "main": [
        [
          {
            "node": "Assemble CRM Data Row",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}

About this workflow

This workflow runs every 15 minutes to scan your socials with FanBase Copilot DMs for purchase-intent messages, enriches matched fans by resolving their social handle via FanBase CRM and scraping their Instagram profile with Apify, then writes or updates a lead row in Google…

Source: https://n8n.io/workflows/18123/ — original creator credit. Request a take-down →

More Marketing & Ads workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.