Competitor Tracker

1. 01Read the brands and the names people type for them1.8 s wall timeNov 3, 2026, 18:01 UTC
   → callagnt_db_readcopy

   {
     "action": "select",
     "table": "mindshare_brands",
     "limit": 100
   }

   ← result

   {
     "rows": [
       {
         "id": 1,
         "brand": "Semrush",
         "variants": ["\\bsemrush\\b"],
         "active": true
       },
       {
         "id": 3,
         "brand": "Moz",
         "variants": ["\\bmoz\\b", "@moz\\b"],
         "active": true
       }
     ]
   }

   » Trimmed: Ahrefs and SE Ranking, the other two of 4

2. 02Semrush: the latest posts, first page2.5 s wall timeNov 3, 2026, 18:02 UTC
   → calldata_x_discovercopy

   {
     "action": "search",
     "query": "semrush",
     "section": "latest",
     "limit": 40,
     "response_format": "json"
   }

   ← result

   {
     "items": [
       {
         "created_at": "Tue Nov 03 18:00:11 +0000 2026",
         "author": "jetwriterjohn",
         "text": "The more things change, the more they stay the same.\n\nI've been combing over the blogs from the major SEO providers (Ahrefs, Semrush, Backli…",
         "likes": 0,
         "replies": 0
       },
       {
         "created_at": "Tue Nov 03 17:41:02 +0000 2026",
         "author": "VikasAnandHS",
         "text": "Semrush confirms 'poisoned' Google results are impacting SEO tools https://t.co/dZvFyzMeTS",
         "likes": 0,
         "replies": 0
       }
     ]
   }

   » Trimmed: 18 more posts of 20, the next-page token, and the fields this step doesn't read (views, media, follower counts)

3. 03Moz: the same search shape, which mostly finds other meanings of the word2.2 s wall timeNov 3, 2026, 18:03 UTC
   → calldata_x_discovercopy

   {
     "action": "search",
     "query": "moz",
     "section": "latest",
     "limit": 40,
     "response_format": "json"
   }

   ← result

   {
     "items": [
       {
         "created_at": "Tue Nov 03 18:00:29 +0000 2026",
         "author": "KalaMana",
         "text": "The Complete AI Research Workflow: From Prompt Discovery to Content Creation moz.\u200bcom/blog/ai-research-workflow-moz-pro https://t.co/X3zrlrW…",
         "likes": 0,
         "replies": 0
       },
       {
         "created_at": "Tue Nov 03 17:45:58 +0000 2026",
         "author": "Angel05337196",
         "text": "Mooch gets Mooched - Wake up with Moz! @MeganFoxWriter https://t.co/Cpho70361F via @YouTube",
         "likes": 1,
         "replies": 0
       }
     ]
   }

   » Trimmed: 18 more posts of 20: Mozambique, a German newspaper, a Japanese idol group and a Serbo-Croatian word among them. 8 of the 20 named "moz", and 1 was about the company

4. 04Moz again, tighter: the handle, the site, its products, or moz next to an SEO word3.5 s wall timeNov 3, 2026, 18:06 UTC
   → calldata_x_discovercopy

   {
     "action": "search",
     "query": "(moz (seo OR ahrefs OR semrush OR backlink OR backlinks OR \"domain authority\" OR keyword OR keywords)) OR @moz OR \"moz pro\" OR \"moz.com\" OR mozbar OR \"moz local\"",
     "section": "latest",
     "limit": 40,
     "response_format": "json"
   }

   ← result

   {
     "items": [
       {
         "created_at": "Tue Nov 03 18:00:29 +0000 2026",
         "author": "KalaMana",
         "text": "The Complete AI Research Workflow: From Prompt Discovery to Content Creation moz.\u200bcom/blog/ai-research-workflow-moz-pro https://t.co/X3zrlrW…",
         "likes": 0,
         "replies": 0
       },
       {
         "created_at": "Tue Nov 03 12:46:53 +0000 2026",
         "author": "HelixKingLabs",
         "text": "@CyrusShepard @Moz A checklist covering technical, security, and international in one place is rare. This has to have saved a lot of people …",
         "likes": 0,
         "replies": 0
       }
     ]
   }

   » Trimmed: 18 more posts of 20, reaching back to Nov 1, and the fields this step doesn't read

5. 05Log one row per brand, so tomorrow adds a point1.4 s wall timeNov 3, 2026, 18:08 UTC
   → callagnt_db_writecopy

   {
     "action": "insert",
     "table": "mindshare_daily",
     "return_rows": true,
     "rows": [
       {
         "brand_id": 1,
         "posts_per_hour": 7.13,
         "median_likes": 0,
         "sample_size": 19,
         "top_post_url": "https://x.com/semrush/status/2117661877253665257",
         "method_version": "latest-40-literal"
       },
       {
         "brand_id": 3,
         "posts_per_hour": 0.32,
         "median_likes": 0,
         "sample_size": 18,
         "top_post_url": "https://x.com/jabezreuben/status/2117213659382599893",
         "method_version": "latest-40-literal-tight"
       }
     ]
   }

   ← result

   {
     "rows": [
       {
         "id": 5,
         "brand_id": 1,
         "measured_at": "2026-11-03T18:08:05.280243+00:00",
         "posts_per_hour": 7.13,
         "sample_size": 19
       }
     ]
   }

   » Trimmed: the Ahrefs and SE Ranking rows (4 sent, ids 5 to 8 back) and each row's other columns

The table it wrote is below. Semrush led at 7.13 posts an hour, 1.6 times Ahrefs, just past the 1.5x line under which a gap is noise. Moz's row went in under its own method name, so a tighter query never mixes with the plain one in the trend.