📌 Pinterest Trend & Reach Report

Search Pinterest by keyword, scrape structured per-pin data, and get a run-level Trend & Reach Report that aggregates the same scraped fields into keyword-level insight. This actor is a superset of a standard Pinterest search scraper: it keeps the exact per-pin output and adds a color-palette / creator-reach / posting-cadence / engagement / image-format report on top — computed only from data already fetched.

Keywords: pinterest scraper, pinterest search scraper, pinterest api, pinterest pins, pinterest trends, pinterest analytics, color palette analysis, creator reach, board analytics, engagement metrics, posting cadence, content freshness, image format distribution, pin data extraction, social media research.

What it does

For every search:

1. Per-pin scrape (default dataset). Each pin is pushed live to the dataset with the base per-pin schema: query, node_id, type, title, description, domain, link, dominant_color, created_at, board, reaction_counts, native_creator, pinner, image_url, image_width, image_height, plus scrapedAt.
2. Trend & Reach Report (key-value store → TREND_REPORT). The same records are aggregated into a single run-level object with five sections:
   • engagement — total / average / median / max reactions and the raw reaction-type-ID breakdown (IDs left unlabeled — Pinterest does not expose their names), plus the top pins by reactions.
   • colorPalette — hue-family shares (red/orange/yellow/green/cyan/blue/purple/pink + black/white/gray neutrals) from each pin's real dominant_color, top raw hex colors, average saturation/brightness, and average reactions per color family.
   • reach — unique pinners and boards, top pinners by follower count, top boards by pin count, and top linked domains.
   • cadence — oldest/newest pin, median age in days, pins in the last 30/90 days, and a per-month histogram from each pin's created_at.
   • imageFormats — portrait/landscape/square counts and average width/height/aspect-ratio.

Every value in the report is a count, mean, median, ratio, hue-bucket, or date-math of scraped fields. Nothing is fabricated, hardcoded, or modeled.

Input

Field	Type	Default	Description
query	string (required)	wallpaper	Search term.
filter	enum all | videos	videos	All pins or video pins only.
limit	integer 1–10000	10	Pins to scrape; also the report's sample size.
topN	integer 1–100	10	Entries per ranking in the report.
rawData	boolean	false	If true, only stream per-pin rows (skip the report).
proxyConfiguration	proxy	Residential	Residential proxy strongly recommended.

Output

• Default dataset: one row per pin (base schema + scrapedAt).
• Key-value store record TREND_REPORT: the aggregated report { query, filter, generatedAt, pinCount, analytics: { engagement, colorPalette, reach, cadence, imageFormats }, scrapedAt }. Not written when rawData=true.

Running

Run on the Apify platform or locally with the Apify CLI:

$apify run --purge

Provide input via the Console form or an INPUT.json. Read the report after a run from the key-value store record TREND_REPORT.

Notes & compliance

• Anti-bot: Pinterest blocks datacenter IPs. Without a residential proxy a run typically returns zero pins and an empty report. The actor retries with a fresh session up to 3 times and fails gracefully with a status message.
• Scrape only publicly available data and respect Pinterest's Terms of Service and applicable laws. Use collected data responsibly.