Call Trends API from MATLAB with webwrite

A MATLAB job talks to Trends API with one POST to https://api.trendsapi.ai/api, a Bearer key, and a JSON body that names a mode. There is no official MATLAB toolbox. webwrite sends the request. jsondecode parses the envelope, then parses the body string a second time. Pulled on 2026-10-08, Google Search for matlab scored 99.0 for the week of 2026-10-03, up 8.79% from 91.0. matlab programming scored 21.0, down 32.26% from 31.0. simulink scored 45.0, down 18.18% from 55.0. Do not copy 21.0 onto 99.0. Do not treat octave Search 73.0, down 17.05% from 88.0, as the GNU Octave token. The same helper that stores those rows also pulls a live Google Trends board; a rank-1 token that weekly Search printed at 25.0 versus 1.0 is documented in Gypsy Rose live rank versus weekly Search.

How does a MATLAB client call Trends API?

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

{"mode":"get_growth","source":"google search","keyword":"matlab","percent_growth":["7D","30D","3M","12M"]}

apiKey = getenv('TRENDSAPI_API_KEY');
url = 'https://api.trendsapi.ai/api';
payload = struct( ...
    'mode', 'get_growth', ...
    'source', 'google search', ...
    'keyword', 'matlab', ...
    'percent_growth', {{'7D','30D','3M','12M'}});
opts = weboptions( ...
    'HeaderFields', {'Authorization', ['Bearer ' apiKey]}, ...
    'MediaType', 'application/json', ...
    'ContentType', 'json');
envelope = webwrite(url, payload, opts);
% body is a JSON string; parse it a second time
inner = jsondecode(envelope.body);
disp(inner.results);

The first mistake most clients make is treating the body field as an object after one parse. It is still a JSON string. Parse the envelope, then parse body a second time. Library docs:

https://www.mathworks.com/help/matlab/ref/webwrite.html
https://www.mathworks.com/help/matlab/ref/jsondecode.html

Free tier allows 100 successful requests per month. Only HTTP 200 responses count against quota. Starter is 5,000, Pro is 25,000, Business is 100,000. There is no MATLAB-specific source on this API. npm covers npmjs package names. python covers PyPI project names. A watchlist that needs File Exchange download counts has to pull those from somewhere else. A Python sibling of this helper is on Python Trends API. An R sibling is on R Trends API.

What did get_growth return for MATLAB language tokens?

get_growth for matlab on google search tips on 2026-10-03. Store the exact strings. The product name, the programming phrase, the Simulink token, the musical octave token, and the GNU Octave phrase are different series.

get_time_series for matlab on Google Search closed at 99 on 2026-10-03, matching get_growth 99.0. Adjacent weeks: 91 on 2026-09-26, 92 on 2026-09-19, 95 on 2026-09-12, 87 on 2026-09-05. Peak in this pull is 99 on 2026-10-03, with 95 on 2026-09-12, 90 on 2026-08-01, and 88 on 2026-06-06. Tip 99 is a new series high. It is not 100. 12M printed +70.69% versus 58.0. YTD printed +65.0% versus 60.0. 7D printed +8.79% versus 91.0. A job that alerts on a one-week +8.79% while the series is already at a new high of 99 is reading a tip, not a start.

matlab programming closed at 21, matching growth 21.0, down 32.26% from 31.0. Adjacent weeks: 31 on 2026-09-26, 30 on 2026-09-19, 29 on 2026-09-12, 30 on 2026-09-05. Peak in this pull is 100 on 2026-06-06, with 87 on 2026-05-16, 74 on 2026-05-23, and 72 on 2026-05-30. 7D signs diverge: product name +8.79% versus programming phrase -32.26%. 12M signs also diverge: product +70.69% versus programming -50.0%. Do not copy 21.0 onto matlab 99.0. Do not alert language adoption off a programming-phrase 7D drop while the product token sits at 99.

simulink closed at 45.0, down 18.18% from 55.0. 30D was 45.0 versus 43.0 (+4.65%). 3M was 45.0 versus 75.0 (-40.0%). 12M was 45.0 versus 74.0 (-39.19%). A job that alerts "MATLAB is down" off Simulink 7D is reading the block-diagram series. octave closed at 73.0, down 17.05% from 88.0. gnu octave closed at 36.0, down 14.29% from 42.0. 12M signs diverge: octave -2.67% versus gnu octave +9.09%. YTD signs also diverge: octave +7.35% versus gnu octave +111.76% from 17.0. The short token is a music series. Do not copy 73.0 onto gnu octave 36.0.

Neighbors this session: python programming Search 21.0, down 38.24% from 34.0, 12M 21.0 versus 24.0 (-12.5%), YTD 21.0 versus 22.0 (-4.55%). r programming Search 36.0, down 42.86% from 63.0, 12M 36.0 versus 76.0 (-52.63%), YTD 36.0 versus 45.0 (-20.0%). julia Search 53.0, down 15.87% from 63.0, 12M 53.0 versus 60.0 (-11.67%). julia is also a given-name series. A Julia client that does the same double parse is on Julia Trends API. An Objective-C NSURLSession sibling is on Objective-C Trends API.

PyPI numpy on source python closed the week of 2026-09-28 at score 67.2 with 207,208,807 downloads versus 190,376,107 on 2026-09-21 (score +9.8%, volume +8.84%). 12M volume was 207,208,807 versus 125,224,131 (+65.47%). scipy closed at 65.7 with 73,907,238 downloads versus 70,415,679 (score +5.46%, volume +4.96%). 12M volume was 73,907,238 versus 48,407,871 (+52.68%). Compare those packages on recent_volume. Do not join 207 million downloads onto Search 99.0. There is no File Exchange source.

How do YouTube, News, and Wikipedia disagree with Search?

YouTube get_growth for matlab closed at 95.0 versus 95.0 (0.0%). get_time_series closed at 95, matching growth. Adjacent weeks: 95 on 2026-09-26, 91 on 2026-09-19, 98 on 2026-09-12, 93 on 2026-09-05. Peak in this pull is 98 on 2026-09-12, with 94 on 2025-09-13 and 93 on 2026-01-31 and 2026-09-05. Tip 95 is not a new YouTube series high, while Search is at a new high of 99. 12M YouTube was 95.0 versus 81.0 (+17.28%). YTD YouTube was 95.0 versus 79.0 (+20.25%). 7D signs: YouTube flat at 95.0 versus Search +8.79%. Do not fill YouTube from Search 99.0.

Google News get_growth closed at 46.0 versus 16.0 (+187.5%). get_time_series closed at 46 on 2026-10-03, matching growth. Adjacent weeks: 16 on 2026-09-26, 29 on 2026-09-19, 47 on 2026-09-12, 23 on 2026-09-05. Peak in this pull is 100 on 2022-09-24, with 93 on 2023-04-15 and 90 on 2023-07-22 and 2023-10-07. Tip 46 is not a new News high. YTD printed growth 999999 versus 0.0. Report 46.0 versus 0.0. 7D News +187.5% while Search +8.79%. Keyword News and the Google News Top News board are different feeds. Source docs for Search sit at Google Trends.

Wikipedia get_growth for matlab resolved to Matlab. 7D and 14D returned collapsed_range. September 2026 scored 1.5 with 365 pageviews versus August 0.0 with 324 pageviews (score growth 999999, volume +12.65%). Report 1.5 versus 0.0 and 365 versus 324. 3M was 1.5 / 365 versus June 0.6 / 341, volume +7.04%. 12M was 1.5 / 365 versus September 2025 10.2 / 607, volume -39.87%. YTD volume was 365 versus 607 (-39.87%). Peak in this pull is 100.0 / 3,090 in November 2025, with 53.6 / 1,807 in May 2022. Three hundred sixty-five September views is not the high-traffic product article. Wikipedia 12M volume falling 39.87% while Search 12M is +70.69%. Compare Wikipedia on volume. Do not join 365 pageviews onto Search 99.0. Source docs for Wikipedia sit at Wikipedia Trends.

How does a MATLAB job pull a live board?

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

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

board = struct('mode','get_top_trends','type','Google Trends','limit',25);
envelope = webwrite(url, board, opts);
inner = jsondecode(envelope.body);
disp(inner.data);

get_top_trends takes a feed type, not a keyword. Pair the live snapshot with sort=rank_change on 1d and 7d. On 2026-10-08T10:01:38Z the live Google Trends board ranked gypsy rose at 1. Weekly Search for that string closed at 25.0 versus 1.0. A helper that only stores 7D Search for matlab at 99.0 misses that board. Mode docs live at Docs.

What belongs in a MATLAB watchlist?

Store matlab, matlab programming, simulink, and gnu octave as separate Search series. Add YouTube matlab because 7D already disagrees in magnitude with Search (flat 95.0 versus +8.79%). Add Wikipedia volume for the resolved Matlab title, not the 0-100 score. Add numpy and scipy on source python and compare recent_volume. Skip a File Exchange column: that source does not exist. Do not store octave 73.0 as GNU Octave adoption. Do not store julia 53.0 as language adoption without a second, language-specific string. Six Search keywords plus one live Google Trends board is 7 calls. Four weekly runs fit the free 100-request cap. Starter is 5,000, Pro 25,000, Business 100,000. Only successful 200 responses count.