The fourteen checks
Reputation
Wikipedia, news, app stores, podcasts, employer sites, trademarks, forums: what the public record already says about you, gathered in one place.
Check 06 of 14.
What happens
What the open, public web says about you, from sources that need no account at all — plus whether you are running ads, which directories your industry lists you on, and which third-party sites your whole CATEGORY keeps appearing on whether or not they mention you. At the deeper tier: whether your Google Business listing is still posting, and a snapshot of open-web sentiment on your own disambiguated name.
Calls it makes
Wikipedia and Wikidata — identity, founding, employee-count claimsGoogle News RSS · Reddit search.json · the Bluesky public API · TwitchiTunes Search for podcasts · the Apple App Store and Google Play listingsGlassdoor and Indeed (link-outs, plus a best-effort Indeed rating) · a USPTO trademark search linkTrustpilot's public rating · the platforms' own ad-transparency centersScrapeCreators Facebook ad-library search — ~2 creditsDataForSEO serp/google/organic/live/advanced — up to nine site: queries for the industry-directory check, and up to fifteen CATEGORY queries for the recurring-domains sweep, $0.002 eachSerpApi engine=google_trends — search interest for your name over 12 months, and its regional split when there is any interest to splitSerpApi engine=google_ai_mode — what Google's AI tells someone in your city about your category, with the sources it drew onSerpApi engine=google_ads_transparency_center — whether you, and the sites recurring in your category, are buying Google adsSerpApi engine=youtube then engine=youtube_video_transcript — who talks about you on video, and the sentence they saidDEEPEST SETTING ONLY, CITY-GATED — SerpApi engine=google_maps then engine=google_maps_reviews: your own Google Maps listing's verbatim reviews, synchronously, with the owner's repliesDEEPEST SETTING ONLY, CITY- AND CATEGORY-GATED — SerpApi engine=google_maps (a second, city-level search) then engine=google_local_services: who holds the Google Guaranteed box in your city, for the home-services/legal/healthcare categories Google's own Local Services list coversDEEPEST SETTING ONLY, OPPORTUNISTIC — SerpApi engine=google_ai_overview: spent ONLY when a free page_token already exists from an engine=google search elsewhere in the pull; as shipped this lane's bulk SERP rides DataForSEO, so no such call exists yet and this reports a named skip, never a purchaseGDELT DOC 2.0 (mode=ToneChart) — news mention volume WITH a tone score over a rolling 3 months, $0.00, entity-anchor gated (a generic name gets the anchor's disambiguated phrase, never the bare word)USPTO TSDR casestatus — real trademark status for a KNOWN serial/registration number, $0.00; honestly idle without one, since this pipeline has no name→number lookup stepScrapeCreators GET v1/reddit/subreddit/details — tier >= 2, situational: only when the free Reddit search above already surfaced a relevant subreddit, ~1 creditDEEPER SETTINGS ONLY — DataForSEO POST business_data/google/my_business_updates/task_post then task_get: whether the business is posting to its Google listing, or an honest 'nothing found'DEEPER SETTINGS ONLY — DataForSEO POST content_analysis/{search,summary,sentiment_analysis}/live: brand citations across news, blogs and forums with a sentiment read, entity-gated on your own disambiguated name (never a bare generic word)www.sec.gov/files/company_tickers.json — a CIK lookup for public companies, $0.00, entity-gated (free-tail round, 2026-08-19); when no current name matches, one efts.sec.gov full-text search for annual reports that name the domain finds a renamed filer (release 072); renders as a small 'public filings' line in the Trust section, not hereAlgolia Hacker News Search API — whether the brand was ever discussed on HN, $0.00, entity-gated (free-tail round, 2026-08-19); collected and wired, no report card yet — see this row's own caveat on upgrade-proposals.mjs
The facts a reader needs
| Language model | None. |
| Cost | ≈ $0.004 per pull for the free-web sources and the Facebook ad search · ≤ $0.018 for the directory check · ≤ $0.03 for the category sweep · $0.02 of SerpApi at tier 1, ≤ $0.09 at tier 2, ≤ $0.19 at tier 3 (adds ≤ $0.04 for maps_reviews + local_services, both city-gated, plus ai_overview's opportunistic $0.01 that this lane never actually spends today). Every SerpApi engine costs the same one search, $0.01 on this account's grandfathered plan. GDELT, the TSDR status check, the SEC EDGAR lookup and the Hacker News search are all $0.00, and still receipted on the leg's own ledger. Reddit subreddit details is $0-$0.0015, tier >= 2, only when a subreddit was already found for free. The deeper tier adds ≈$0.00225 for the Google-posting check and ≈$0.072 for the open-web sentiment snapshot (three DataForSEO calls) — both carry their own inline cost_usd rather than rolling into this leg's ScrapeCreators-only ledger above. |
| Depends on | The two units on this leg run LAST, after the SEO and reviews legs, because they need what those produce: the category terms the sweep asks are the SEO leg's own industry-tier keywords, and the city that localises them comes from the reviews leg's domain-verified listing. A pull with no industry keywords reports the sweep as skipped with that reason rather than falling back to searching your brand name — searching your own name would measure who already knows you, which is the opposite of what a shelf map is for. GDELT and the trademark-status check run alongside them, gated on the same topped-up entity anchor. The Google-posting check and the open-web sentiment snapshot are both gated on DataForSEO credentials independently of the twelve free sources, and both are deeper-tier only. maps_reviews and local_services are BOTH city-gated on the same anchor — no city, no spend, never a guessed franchise — and local_services is additionally category-gated against Google's own closed Local Services list. |
| When it is absent | Reddit occasionally refuses an anonymous request. That is recorded as skipped, not as failed, and not as an absence of discussion. Without a SerpApi key the SerpApi units each report that by name and the category sweep still runs on DataForSEO alone — you lose the AI-answer surface, not the shelf map. And because SerpApi has no metered overage, an exhausted monthly block is refused OUT LOUD before anything is spent: the units say the block ran out, never that nothing was found. A BRAND OUTSIDE GOOGLE'S LOCAL SERVICES LIST (most of this product's brands — marketing agencies, restaurants, SaaS) gets local_services as an honest, named skip, never an invented query; the live probe confirmed an arbitrary category is REJECTED by the endpoint before it can even be billed. THE AI OVERVIEW FOLLOW-UP IS WIRED BUT DORMANT: it never buys a search purely to fish for a page_token, and as shipped nothing in this lane produces one for it to read for free — see the endpoint's own note above. GDELT is soft-throttled (~1 request/5s) and a pull that hits the shared limit reports skipped, not zero coverage. The trademark-status check reports a named skip — no number, or no USPTO_TSDR_API_KEY — never a fabricated 'not trademarked'; reputation.mjs's own tmsearch.uspto.gov search link is unchanged and stays the default for a bare name. And a name match is never treated as identity: every app, Wikipedia article, podcast, Bluesky handle and Twitch channel this leg finds is classified verified, lead, or off_entity before it renders — podcast verification requires a retained RSS-feed host tied to the brand domain even for a distinctive name; the other source types use their recorded domain, Wikidata or name checks, a lead is shown and labelled 'possible, unverified' but never counted as yours, and an off_entity hit that carried none of your own name at all is counted and named rather than silently dropped. When nothing survives as verified the leg says so directly — 'No app store listing of yours found', 'No Wikipedia page found for you', 'No podcast found for you' — instead of leaving the section blank. The Google-posting check's own task queue runs far slower than every other async unit in this codebase (measured ~650 seconds on one live pass) — expect it to answer 'pending' more often than a fast unit would, honestly, rather than waiting out the whole queue on every pull. The sentiment snapshot never fires on a brand name that is itself a common English word — it reports a named skip rather than a noisy, off-entity keyword search — and it is explicitly a SNAPSHOT at pull time, not ongoing monitoring, and a different measurement from ai_visibility's llm_mentions (that asks whether AI answers cite you; this asks whether the web writes about you at all). The SEC EDGAR lookup and the Hacker News search are both gated the same way GDELT is — a bare generic-word brand name is refused before the call runs — and a private company (the overwhelming default) or a brand with no HN history both get an honest, named absence rather than an empty section pretending to be a checked one. |
Who supplies it
- ScrapeCreators — Social profiles and posts across seven platforms, the brand's own video transcripts, organic mentions, and the ad-transparency archives.
- SerpApi — Four things nothing else in the stack can do, all inside the reputation check: Google's AI Mode answer for your city, whether your category's sites are buying ads, who talks about you on YouTube, and your search-interest curve. DEEPEST SETTING ONLY, AND CITY-GATED: your Google Maps listing's own verbatim reviews (with the owner's replies), and — for the home-services/legal/healthcare brands Google's Local Services program actually covers — who holds the Google Guaranteed box in your city. Without it the report still maps your category's shelves — that sweep runs on DataForSEO — but loses the AI-answer surface and the panels above.
- Meta ad library (Graph API) — Real Facebook and Instagram ad rows from Meta's own archive, free.
- Public web sources (no account, no key) — Everything below runs with no key at all, and always has.
- USPTO TSDR (Trademark Status & Document Retrieval) — Brand protection — real trademark status (live/dead/pending) for a KNOWN serial or registration number. This pipeline has no name -> number lookup step yet, so this call is real but usually idle; the deterministic tmsearch.uspto.gov search link stays the default path for a bare brand name.
SEO
Every keyword's real search volume, difficulty and cost-per-click — and where you actually rank for it — plus who overlaps you and the terms they win that you don't.
Ads
What you are running right now across four ad networks — creative, copy and start dates — from the archives the platforms are obliged to publish.