Own-brand demand is one string on several sources. The string is the name customers type, not the aisle. POST https://api.trendsapi.ai/api with mode get_growth and source google search, tiktok, youtube. Plan limits stay on pricing. The search property notes stay on the Google Trends guide. Skip plotting until body is decoded from its string form.
The name, not the aisle
POST https://api.trendsapi.ai/api
Authorization: Bearer <api_key>
Content-Type: application/json
{"mode":"get_growth","source":"google search, tiktok, youtube","keyword":"nike","percent_growth":["1M","3M","12M"]}
A weekly brand packet that swaps in running shoes is a category packet. Keep that row on the ecommerce job. This page is the logo.
Nike search held. TikTok's score did not.
Own-name packet dated 27 Aug 2026 for nike:
| Source | 1M | 3M | 12M | Recent |
|---|---|---|---|---|
google search |
+3.17% | −17.72% | −2.99% | 65 on 22 Aug 2026, volume 26,200,000 |
tiktok |
−100% on score, −9.53% on volume | date_out_of_range |
date_out_of_range |
score 0.0 on 25 Aug 2026, volume 9,648,141 |
youtube |
−4.55% | −4.55% | −4.55% | 42 on 22 Aug 2026 |
YouTube printed the same −4.55% on every window because each baseline was 44. That is a flat tape, not three independent crashes. TikTok only held 27 Jul 2026 to 25 Aug 2026. A brand scorecard that requires 12M on TikTok will fail every week until that source lengthens.
volume_weighted_growth was −0.25. Coverage was 2 of 3 sources. Google search was 73.09% of the weighted volume. TikTok was 26.91%. YouTube dropped out. A brand lead who only sees −0.25 will miss that search is slightly up on the month and that TikTok's 0-100 hit the floor.
Store the legal name, the ticker-style short name, and one common misspelling as three keywords. Do not average them. Plot them. TikTok notes: the TikTok guide.