What's Trending

You give it your website; it suggests a "WordBank" of market terms and competitor profiles, then scans Reddit, Facebook, Instagram and X for public posts matching those terms. Each result — a Reddit thread, an X post, an Instagram caption — keeps its original URL, author handle, timestamp and whatever engagement metrics that platform exposes (comments, likes, shares, views). The product ranks topics by mentions, unique authors, freshness and engagement, and turns the strongest threads into a content brief: topic, why it matters now, two or three source links, and a suggested angle. It does not draft or publish content itself — the vendor is explicit about that in its own FAQ.

Two things stand out. First, evidence discipline: rather than a black-box "trending" score, every ranked topic links back to the actual post, and the vendor's own blog content (not just marketing copy) argues for retaining missing metrics as missing rather than zeroing them out and for counting unique authors separately from raw mention volume — a more careful methodology than most listening tools bother to state publicly. Second, the free tier is a real one-time full scan with no card and no crippled feature set, which lets a buyer see actual evidence from their own market before paying anything (https://whatstrending.cc/). Third, for a $50/month tool it ships a documented REST API and a working MCP server (verified live: openapi.json returns a real OpenAPI 3.1 spec, and the MCP discovery endpoint at api.whatstrending.cc/.well-known/mcp.json responds with three callable tools) — agent access at this price point on a tool this new is unusual (https://api.whatstrending.cc/openapi.json, https://api.whatstrending.cc/.well-known/mcp.json).

Three tiers, all publicly listed with real numbers, no "contact sales": Free ($0, one complete scan, no card); Weekly ($20/month with a 7-day free trial, one full scan per week); Pro ($50/month, unlimited scans up to hourly, plus the REST API and MCP agent access). There is no separate pricing page — the tiers live in a section on the homepage — and no annual-billing toggle, so the monthly figures are the only figures (https://whatstrending.cc/).

The product monitors four platforms — Reddit, Facebook, Instagram and X — and its own FAQ and terms both qualify this as "subject to each platform's public-data availability": engagement fields and post visibility vary per source, and the vendor does not claim access to private content or metrics a platform doesn't expose (https://whatstrending.cc/terms). That is an honest caveat rather than a hidden one, but it means the depth of Facebook and Instagram coverage in particular is not something a buyer can verify without running their own scan — public API access to those two platforms is generally the most restricted of the four.

What's Trending is one of at least six products (alongside SendHQ, AI for Database, AuraVMS, JobFinder AI and Compliance Radar) built by Wavicle, a self-described solo AI consulting studio that ships fixed-price client work in 4–8 week sprints and appears to build these tools as its own side portfolio (https://www.wavicle.tech/about). The whatstrending.cc domain was registered on 19 August 2026 per registry RDAP data, and the site's own terms and privacy policy carry the same date, putting the product at roughly one month old as of this listing (RDAP query, https://whatstrending.cc/terms). No independent reviews, press coverage or user discussion of the product were found via web and news search. This is not a red flag on its own — it is simply too new to have a track record, and a buyer should weigh that against the very low cost of testing it with the free scan.

The privacy policy states the product collects only publicly accessible posts, comments, handles and visible engagement metrics, and explicitly says it does not intentionally collect private messages or private-account content. Product analytics run through Mixpanel's EU-residency service (configured without IP enrichment), Microsoft Clarity, and Sentry for error diagnostics; Google Sign-In data is used only for authentication and is not used for ads or to train models (https://whatstrending.cc/privacy). There is no dedicated security/trust page (no SOC 2, DPA or sub-processor list published), which is normal for a product this young but means an enterprise buyer has nothing to point to beyond the privacy policy's prose.