Google Trends rank_change climbers via get_top_trends

Google Trends rank_change on Trends MCP is get_top_trends with type Google Trends, sort rank_change, and a window such as 1d. From the August 26, 2026 pull, the board stamped 2026-08-26T12:01:40 UTC. The 1d window used baseline 2026-08-25T02:01:58 UTC and returned count 28. jake lang sat at rank 43, prev_rank 130, rank_change 87. The 7d window returned count 3 against baseline 2026-08-19T02:01:41 UTC. The 30d window returned count 0. google search for jake lang sat at 26.0 on 2026-08-22, up 271.43% over 7 days from 7.0. Free tier access is 100 requests per month.

What Google Trends rank_change data shows right now

From the Trends MCP scan of August 26, 2026, Google Trends sort rank_change stamped 2026-08-26T12:01:40 UTC and filled 28 climber rows on window 1d, 21 on 3d, 3 on 7d, 2 on 14d, and 0 on 30d, so a default sort rank pull at limit 25 never lists these mover fields.

Shared as_of 2026-08-26T12:01:40 UTC:

Google Trends 1d climbers (first 12 of 28):

The 1d list also reprinted jake lang later as rank 43, prev_rank 44, rank_change 1, a second object for the same current rank. The 3d list did the same: jake lang at rank 43 with prev_rank 143 / rank_change 100, then again with prev_rank 108 / rank_change 65. Quote both rows; do not silently drop one.

Growth follow-ups on the same pull day used source google search and youtube (weekly points, recent_date 2026-08-22, 261 points):

alyshia rushing sat at Google Trends rank 64 on the 1d climber board. google search sat at 16.0 against a 0.0 baseline on 7D and 30D, flagged data_quality low because the series is over 90% zeros. Quote 16.0 versus 0.0, not 999999.0 percent. trump approval sat at Google Trends rank 94 and was the number-two 7d climber (prev_rank 132, rank_change 38) with google search 41.0, down 4.65% over 7 days from 43.0 and down 59.0% over 30 days from 100.0. For the current-rank board, see the Google Trends trending now API.

Exact API calls for climber boards

sort rank_change plus window is a separate get_top_trends request from default sort rank, so a 1d climber pull and a current-rank pull on August 26, 2026 are two requests, and each extra window is another request because window is per call.

{
  "mode": "get_top_trends",
  "type": "Google Trends",
  "sort": "rank_change",
  "window": "1d",
  "limit": 50
}

{
  "mode": "get_top_trends",
  "type": "Google Trends",
  "sort": "rank_change",
  "window": "7d",
  "limit": 50
}

{
  "mode": "get_growth",
  "source": "google search, youtube",
  "keyword": "jake lang",
  "percent_growth": ["7D", "30D"]
}

The 1d baseline on this stamp was 2026-08-25T02:01:58 UTC against as_of 2026-08-26T12:01:40 UTC, about 34 hours, not a rolling 24-hour clock. Offset and limit still apply; details live on get_top_trends offset and limit.

Window length changes who counts as a climber

A longer rank_change window on Google Trends can return fewer rows, not more, because only titles still on the current 200-row board that also sat on the older snapshot can climb, and a 30d baseline can produce count 0 with an explanatory note.

The 1d call returned 28 movers. The 3d call returned 21, led by dolly parton at rank 1 from prev_rank 191 (rank_change 190), then hawk fire at rank 14 from 146 (rank_change 132). hawk fire google search sat at 2.0 on 2026-08-22, up 100.0% over 7 days from 1.0. The 7d call returned three rows: jake lang +108, trump approval +38, carnival breeze sanitation program results +9 (rank 137 from 146). The 14d call returned jelly roll at rank 28 from 116 (rank_change 88) and police at rank 109 from 190 (rank_change 81). jelly roll google search sat at 6.0, flat over 7 days from 6.0 and down 45.45% over 30 days from 11.0.

The 30d call returned count 0 and the note "No items on the current 'Google Trends' board climbed versus the 30d baseline; everything held or fell." Baseline_ts was 2026-07-27T01:01:39 UTC. Empty is a real response. Do not retry with a larger limit to invent climbers. Category boards on the same Google pipeline are documented on Google Trends by category API.

Same 1d sort on other live feeds

sort rank_change with window 1d is valid on other daily feeds from the same Trends MCP account, and fill rates on this August 26, 2026 stamp ranged from 28 climbers on Google Trends and Wikipedia Trending down to 2 on YouTube Trending.

Wikipedia Trending 1d used the same as_of and the same 2026-08-25T02:01:58 UTC baseline. Count 28. The top mover was a file title at rank 66 from 174 (rank_change 108). Named article climbers included Donald Trump at rank 55 from 117 (+62), List of S&P 500 companies at 45 from 100 (+55), Killing of the Clancy children at 23 from 77 (+54), and ChatGPT at 22 from 49 (+27). Wikipedia 7d against 2026-08-19T02:01:41 UTC returned 25 rows at limit 25, led by Toxic (2026 film) at rank 14 from 171 (rank_change 157).

TikTok Trending Searches 1d returned count 22, led by Otterbox and Best Buy Sell iPhone 18 Pro Case at rank 15 from 68 (rank_change 53), then Jumper Dance TikTok at 14 from 60 (+46). YouTube Trending 1d returned count 2: Heaven Baby at rank 42 from 67 (+25) and a Jarahn audio title at rank 10 from 13 (+3). Spotify Top Podcasts 1d returned 25 rows at limit 25, led by 2 Bears, 1 Cave with Tom Segura and Bert Kreischer at rank 57 from 110 (rank_change 53). For the Wikipedia current-rank board, see the Wikipedia Trending API.

Duplicate jake lang rows on one snapshot

Google Trends rank_change can emit two objects that share the same current rank and keyword while disagreeing on prev_rank, so a unique-keyword count of the August 26 1d list under-counts the payload and a unique-rank count drops a real row.

On window 1d, jake lang appeared as rank 43 / prev_rank 130 / rank_change 87 and as rank 43 / prev_rank 44 / rank_change 1. On window 3d the pair was prev_rank 143 / rank_change 100 and prev_rank 108 / rank_change 65, still both at current rank 43. The 7d list had a single jake lang row (prev_rank 151, rank_change 108). Store the raw objects. google search sat at 26.0 on 2026-08-22, up 271.43% over 7 days from 7.0 and up 550.0% over 30 days from 4.0. youtube sat at 34.0, up 277.78% over 7 days from 9.0.

google docs sat once, at rank 27 from 47 (rank_change 20), with google search 52.0, up 1.96% over 7 days from 51.0. insidious 6 sat at rank 10 from 12 (rank_change 2) with google search 84.0, up 300.0% over 7 days from 21.0 and up 1580.0% over 30 days from 5.0. A small rank_change can sit on a large weekly index move. Rank change is board position versus a snapshot, not search-index growth.

Limits of sort rank_change

rank_change only lists titles that remain on the current board and that also existed on the chosen baseline, so new entries with no prev_rank never appear as climbers, and a 30d window can return zero rows even when 1d returns 28.

The 1d baseline is a prior feed snapshot (here 02:01 UTC the previous calendar morning), not a precise 24-hour rolling window. Weekly google search points end on 2026-08-22, four days behind this live stamp. alyshia rushing google search is data_quality low with a 0.0 baseline; quote the two values. File: titles on Wikipedia climber boards are real rows in the payload and will break a naive "article title" parser.

Each window is its own get_top_trends request. Five Google Trends windows plus four other-feed 1d checks is nine live-board requests before any get_growth follow-up. The free tier is 100 requests per month. Starter is $19 for 1,000 requests. Check current pricing before locking a production budget.