SpyTrend: Ad Intelligence That Profiles Whole Competitors, Not Individual Ads

Most ad intelligence tools return a list of matching ads for a given search term. SpyTrend reframes the research unit: instead of asking 'what ads match this keyword?' it asks 'what is this advertiser running?' The platform consolidates every creative, landing domain, country target, and run duration for a single advertiser into one profile — so a research session starts with a competitor, not a search result. That architectural choice changes the workflow meaningfully: instead of piecing together strategy by examining ads one at a time, you see their entire advertising operation at once. The corpus that powers this is real in scale: 1.1B+ ads with approximately 230M new additions each month, auto-classified into 33 top-level categories and 343 subcategories. Classification is automatic — no manual tagging. Filtering by industry vertical, region, or ad format narrows to a relevant set without building the query from scratch.

Beyond ad discovery, SpyTrend surfaces signals that standalone ad libraries don't provide. Duration tracking shows how long a campaign has been running; paired with deduplication to flag repeated creatives, this identifies which ads a competitor has kept live for months — a more reliable performance signal than which ones launched recently. An ad that has run unchanged for 90 days is typically converting. An ad that launched last week may not have data behind it yet. Landing page uncloaking follows competitor redirect links to surface the actual destination URL. Many performance advertisers route traffic through cloaked links that obscure the final landing page; SpyTrend bypasses those redirects so the creative and its destination can be analyzed together — the full funnel in one view rather than a creative with a dead-end URL.

SpyTrend ships an official MCP server built to the OAuth 2.1 + PKCE standard, compatible with Claude, Claude Code, ChatGPT, Cursor, Codex, and VS Code. The server exposes 20 tools covering ads, advertisers, creatives, webmasters, hubs, shops, trends, and media — addressable in plain language without opening a dashboard. For AI agent workflows, a concrete example: instead of exporting CSVs and pasting data into Claude, you can ask 'show me every advertiser in health supplements running Meta ads for more than 90 days' directly from your AI client. The token model is transparent: Starter gets 500 lifetime tokens, Pro gets 40,000 per month. Meta data costs 1 token per row; TikTok data (Pro only) costs 100 tokens per row — at 40k tokens/month, TikTok-heavy research exhausts the allocation faster than Meta-only workflows. Media retrieval costs 1 token per ad or 10 per creative. Teams building LLM-powered research workflows can integrate SpyTrend's ad corpus without negotiating a custom data license.

The free Starter tier does not require a credit card and provides access to the Meta corpus with a 500-token lifetime allocation. At 1 token per Meta ad row, that is roughly 500 rows — enough to pull data on a real competitor and validate the data quality and interface before committing. That ceiling is by design: Starter is a sample gate, not a research budget. If the first competitor query lands, that is exactly what it was designed to tell you about whether Pro is worth it. Pro at $79/month adds TikTok data, unlimited searches, 40,000 monthly MCP tokens, webmaster grouping, Pixel & Landing Intelligence, deduplication, data export, and 90-day history. The Pro plan includes one AI category — a focused slice of the 33-category taxonomy; additional categories are available at $50/month each. Team pricing starts from $63.20/seat/month and scales with team size (4 to 20 members), with centralized billing and priority support. A 7-day money-back guarantee applies to first payments. Annual billing is available with a volume discount. Ten languages are supported — English, Russian, Ukrainian, Spanish, Portuguese, German, French, Chinese, Japanese, and Indonesian — for teams researching non-English advertising markets. No card needed to start: try the Starter tier at spytrend.com.

SpyTrend fits two distinct users well. The first is the performance marketer running Meta or TikTok campaigns who needs to understand a specific competitor's advertising operation: what they run, how long they run it, where traffic lands, and how creative reuse maps to campaign longevity. The advertiser profile model gives them the whole picture in one session rather than a filtered list of individual placements. The second is the AI or engineering team building research workflows: the MCP server, clean token model, and 20-tool surface make SpyTrend's ad corpus accessible to any MCP-compatible LLM without custom integrations. The platform's data coverage is currently Meta and TikTok; no other advertising channels are listed. Teams running cross-channel campaigns will need additional tools for those networks. Starter's 500-token lifetime allocation covers a validation session — serious ongoing research requires Pro.