get_top_trends for Google Trends with limit 25 returned 25 rows stamped 2026-11-17T09:01:42Z. outlook outage sat at rank 2 in that list. The same token was omitted from the 1d movers board, count 0, no baseline within 1 day of 2026-11-16, with a note that this feed updates roughly monthly. Weekly Google Search scored 100.0 versus 0.0 on 2026-11-14, growth 999999.0, matching series 100.0. Daily Search printed 100 on 2026-11-13, then 2 on 2026-11-16. Wikipedia, YouTube, and News boards stamped the same second as Google Trends. Yesterday the Trends board was still on 2026-11-04T07:01:36Z, a twelve-day lag. A helper that only stores 1d climbers then misses rank 2, the weekly 100.0 new high, and the board roll. Pulled on 2026-11-17.

What did live rank, 1d movers, and 14d movers print?

POST https://api.trendsapi.ai/api
Authorization: Bearer <api_key>
Content-Type: application/json

{"mode":"get_top_trends","type":"Google Trends","limit":25,"sort":"rank"}

The rank call returned count 25. Rank 1 faa is a related aviation token. Rank 2 is outlook outage. Rank 3 robert pattinson also sat at Wikipedia Trending rank 1. Rank 4 is microsoft outage, a sibling phrase of rank 2. Rank 5 ground stop sits next to rank 1. Grim tokens at ranks 15, 16, and 17 are skipped here. Rank 12 ao3 is skipped because the expanded title fails a first-person-plural word scan.

The 1d call used sort rank_change and window 1d:

{"mode":"get_top_trends","type":"Google Trends","limit":25,"sort":"rank_change","window":"1d"}

It returned count 0, data [], baseline_ts null, and a note: no baseline snapshot within 1 days of 2026-11-16, retry window 30d, this feed updates roughly monthly. The 3d call returned the same empty shape, pointing at 2026-11-14. The 7d call returned count 0, data [], no baseline within 3 days of 2026-11-10, and the same monthly-feed note. A helper that only stores 1d or 7d movers then has nothing to join onto rank 2.

The 14d call did return rows. count 3, baseline 2026-11-03T03:01:44Z. First row: snow, rank 37, previous 150, change +113. Second: chipotle, rank 77, previous 147, change +70. Third: lindsay clancy, rank 151, previous 167, change +16. All three sit off the live top 25. outlook outage was absent. A count of 3 is not a truncated top 25. The 30d call returned count 0, data [], baseline 2026-10-18T02:01:38Z, and a held-or-fell note: no item on the current board climbed versus that baseline. A helper that only stores 14d climbers then stores snow, chipotle, and a third off-board token, and misses rank 2.

Wikipedia Trending, YouTube Trending, and Google News Top News all stamped 2026-11-17T09:01:42Z, matching Google Trends. Yesterday Trends sat on 2026-11-04T07:01:36Z while the other boards had already moved to 2026-11-16T09:01:38Z. The twelve-day lag closed in one session. Wikipedia Trending omitted outlook outage (rank 1 Robert Pattinson, rank 4 Suki Waterhouse, rank 8 Jensen Huang, rank 10 The Body TV series). YouTube Trending omitted the token (rank 2 a GTA 6 marathon trailer, rank 4 XG DEPENDS, rank 7 Flunky x Jammin Otendewa, rank 10 a Christmas songs playlist). Top News omitted the token. Rank 3 paraphrased Peltola expanding a lead over Sullivan. Rank 10 paraphrased baby sea turtles reaching the ocean after a California nesting. Rank 14 paraphrased Pope Leo on a 12-day South America tour. Rank 15 paraphrased a campaign to bring Jensen Huang back as Nvidia CEO. Keyword News for the same string was already at 86.0, and Top News did not name it. Those are still different feeds.

Every successful response is {"statusCode": int, "body": string}. Parse body a second time or rank, rank_change, and recent_value stay inside a string. Envelope notes sit at https://trendsapi.ai/docs. Google Search board notes sit at https://trendsapi.ai/trends/google-trends.

What did weekly Search print against daily Search?

Weekly Google Search for outlook outage scored 100.0 versus 0.0 on 2026-11-14, growth 999999.0, matching series 100.0, data_quality low. Adjacent weeks were 0.0, then 0.0, then 0.0, then 0.0. 14D, 30D, 12M, and YTD each compared 100.0 versus 0.0 and printed growth 999999.0. 3M was 100.0 versus 1.0, up 9900.0%. Peak weekly was 100.0 on 2026-11-14, then 3.0 on 2026-09-05, 3.0 on 2026-01-24, 2.0 on 2026-05-02, 2.0 on 2025-07-12, and 2.0 on 2024-11-30. Tip 100.0 is a new all-time high off a zero floor. A 14D growth of 999999.0 off a 0.0 floor is not a 7D rise of 33.33%.

Daily Google Search printed 81 on 2026-11-12, 100 on 2026-11-13, 5 on 2026-11-14, 1 on 2026-11-15, and 2 on 2026-11-16. The live board stamped 2026-11-17, after daily already decayed from 100 to 2. Daily get_growth recent was 2.0 versus 0.0 on 2026-11-16, growth 999999.0, data_quality low. 14D daily and 30D daily compared 2.0 versus 0.0 and printed growth 999999.0. A helper that only stores weekly Search 100.0 then misses daily 100 on Nov 13 and the decay to 2. A helper that only stores daily growth recent 2.0 then also misses 100. Daily decayed from 100 to 2 while the live board still ranked the token at 2.

microsoft outage, live rank 4, scored weekly Search 100.0 versus 0.0 on 2026-11-14, growth 999999.0. 3M compared 100.0 versus 0.0, growth 999999.0. 12M was 100.0 versus 1.0, up 9900.0%. Same 7D close as outlook outage, different live rank, different 3M baseline. Phrase rank 2 is not phrase rank 4. Do not copy outlook 100.0 onto a microsoft-outage alert that only watches rank.

php in the same session still sat on the 2026-11-07 Saturday at 25.0 versus 38.0, volume 556000. gleam, zig, hare, elm, and haskell programming also sat on 2026-11-07. outlook outage had already rolled to 2026-11-14. Two Saturday tips in one run. Do not join php 25.0 onto outlook outage 100.0 as the same week. A PureScript client that stores those language rows is at https://trendsapi.ai/purescript-trends-api.

Feed Close Versus Window Tip or stamp
Live rank 2 on top 25 rank 2026-11-17T09:01:42Z
1d movers omitted count 0, no baseline 1d 2026-11-17T09:01:42Z
7d movers omitted count 0, no baseline 7d 2026-11-17T09:01:42Z
14d movers omitted count 3, all off top 25 14d baseline 2026-11-03
30d movers omitted count 0, held or fell 30d baseline 2026-10-18
Weekly Search 100.0 0.0, 999999.0 7D 2026-11-14
Weekly Search peak 100.0 new high weekly series 2026-11-14
Daily Search peak 100 then 2 daily series 2026-11-13 then 2026-11-16
Daily growth recent 2.0 0.0, 999999.0 7D daily 2026-11-16
YouTube 100.0 0.0, 999999.0 7D 2026-11-14
News 86.0 0.0, 999999.0 7D 2026-11-14
Shopping 37.0 0.0, 999999.0 7D 2026-11-14

Did YouTube, News, Shopping, and Wikipedia agree?

YouTube for outlook outage scored 100.0 versus 0.0 on 2026-11-14, growth 999999.0, matching series 100.0, data_quality low. Adjacent weeks were 0.0, then 0.0, then 0.0, then 0.0. 14D, 30D, 3M, 12M, and YTD each compared 100.0 versus 0.0 and printed growth 999999.0. Peak was 100.0 on 2026-11-14, then 1.0 on 2026-09-05, and 1.0 on 2024-07-20. Tip 100.0 is a new high. Same 7D sign as weekly Search. YouTube Trending omitted the token.

Google News scored 86.0 versus 0.0 on 2026-11-14, growth 999999.0, matching series 86.0. Adjacent weeks were 0.0, then 41.0, then 55.0, then 53.0. 14D was 86.0 versus 41.0, up 109.76%. 30D was 86.0 versus 53.0, up 62.26%. 3M was 86.0 versus 50.0, up 72.0%. 12M and YTD compared 86.0 versus 0.0 and printed growth 999999.0. Peak was 100.0 on 2026-08-22, then 98.0 on 2026-09-05, 86.0 on 2026-11-14, 55.0 on 2026-10-24, 53.0 on 2026-10-17. Tip 86.0 is not a new high. Search 100.0 is a new high. News 86.0 is not. Top News omitted the token while keyword News sat at 86.0. A helper that only stores keyword News 86.0 then still needs the Top News miss and the Search 100.0 new high.

Google Shopping scored 37.0 versus 0.0 on 7D, growth 999999.0, data_quality low. 14D, 30D, and 3M compared 37.0 versus 0.0 and printed growth 999999.0. Shopping printed 37.0 the same week weekly Search printed 100.0. TikTok returned 404 no_data. Amazon returned 404 no_data. Wikipedia keyword outlook outage returned 404 no_data. Store the misses. Wikipedia Trending omitted the string. A helper that only stores Wikipedia Trending then misses Search 100.0, daily 100, and live rank 2.

What breaks if the helper stores only one of these fields?

Forty-fourth shape: live rank 2 sits on a 25-row snapshot dated 2026-11-17T09:01:42Z, matching Wikipedia, YouTube, and News at the same second, after the board rolled off a twelve-day-stale 2026-11-04 stamp. 1d, 3d, and 7d movers are empty with no baseline and a monthly-feed note. 14d movers return 3 off-board tokens. 30d movers are empty after a held-or-fell note versus 2026-10-18. Weekly Search 100.0 on 2026-11-14 is a new all-time high off a zero floor. Daily Search printed 100 on 2026-11-13, then 2 by 2026-11-16. YouTube 100.0 is a new high. News 86.0 is not a new high. Shopping printed 37.0. TikTok, Amazon, and Wikipedia keyword series 404. Top News omitted the token. Wikipedia Trending omitted the title. Inverse of the 2026-11-16 rank-14 omitted-1d shape on a stale board. The board rolled. The movers did not fill in.

A helper that only stores live top 25 then still needs weekly Search 100.0, daily 100, and the mover-window misses. A helper that only stores 1d or 7d climbers then gets an empty list. A helper that only stores 14d climbers then stores snow at rank 37 and misses rank 2. A helper that only stores weekly Search 100.0 then misses daily 100 and the decay to 2. A helper that only stores daily growth recent 2.0 then misses 100. A helper that only stores keyword News 86.0 then still needs the Top News miss and the Search new high. Pair board rank, every mover window, daily series, daily growth, and weekly growth. Store all timestamps.