Category id 134b9dc1deae69ae (derived from Scout's category name (no id in Scout's data)). 34 alternatives, shown in Scout's order.
Our own measured runs
Scout: own SMTP probe vs MillionVerifier on control100 (2026-11-03) (internal: alternatives/research/smtp-probe-control100-2026-11-03.md (Scout's; not published))
Alternatives
Findymail
commercialnot runnable on gaents yet
Email finder API keyed on (name + company_domain) or LinkedIn URL. Returns verified emails with a refund-on-bounce SLA that few competitors match. (Scout's description)
71% coverage in the Anymail Finder 14-tool study (mid-pack) — but 92% accuracy on EU domains in the b2b-email-finder benchmark and top-3 in every independent study it appears in. US-hosted; use as root on bigbrain, not on vpsm default path.
Lock-in risk
low
Credited to Scout (2026-11-09, Scout last_verified). gaents does not score items.
Scout's notes
TUNABLES (wire in /find-email wrapper): default_waterfall_position=root(bigbrain); default_confidence_threshold="valid" (their scale); default_catch_all_policy=skip_unless_second_tool_confirms; default_rate_limit_rps=10 (docs); default_gdpr_basis=LI+opt-out; why_root: highest independent EU accuracy at lowest £/1k verified in this batch. [2026-10-19T23:00Z cloud refresh: 2026-10-19 session 2: pricing still UNVERIFIED. findymail.com/pricing returns a Cloudflare 'Just a moment...' challenge (HTTP 403) to Chrome, Firefox, iPhone, curl and wget user agents; /pricing.md, /llms.txt (404), help.findymail.com, the blog and the sitemap are also blocked. The API docs (app.findymail.com/docs, last updated 2026-07-31) are readable: 'all endpoints have a concurrent rate limit of 300 simultaneous requests, unless stated otherwise'. This differs from our stored tunable 'default_rate_limit_rps=10 (docs)'. Verifier calls use 'one verifier credit on all attempted verification'.] [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: 2026-10-20: stored '$99/mo Starter (5k credits)' replaced: the $99 plan is now 'Individual' and the page no longer quotes a credit count.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: hosted service (API docs base URL https://app.findymail.com; pricing page says 'Hosted in the EU' and lists no self-hosted plan). licence_class left out (terms URLs tried returned 404). data_export left out (no doc).]
Dropcontact
commercialnot runnable on gaents yet
French email finder + enricher. 100% EU-server processing, CNIL-audited. LinkedIn URL supported on Growth+. Re-credit on miss. (Scout's description)
69% coverage / 93% accuracy in AMF benchmark. Growth tier needed for LinkedIn input. Best-in-class GDPR posture, which is why it's the root on vpsm even though a US-hosted competitor may be cheaper.
Lock-in risk
low
Credited to Scout (2026-11-09, Scout last_verified). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=root(vpsm); default_confidence_threshold="verified"; default_catch_all_policy=skip (their default); default_rate_limit_rps=5; default_gdpr_basis=EU-native (CNIL); why_root_vpsm: only tool where tenant counsel review is trivial — DPA is boilerplate. [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: api_or_webhooks from developer.dropcontact.com; licence_class left out (terms URL tried returned 404); self_hostable left out; data_export n/a (enrichment API).] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): licence_class proprietary from Dropcontact terms section 15 (software protected; no decompilation). self_hostable left out (no statement). data_export n/a (enrichment API).]
FullEnrich
commercialnot runnable on gaents yet
Waterfall-inside-a-waterfall: FullEnrich itself calls 25+ providers under the hood. Top coverage in the AMF study. (Scout's description)
CONFIRMED 2026-10-19: Pro shown as $55/mo for 1,000 credits ($0.055 per credit), monthly or annual, credit slider from 500 to 100K+; 'Waterfall Enrichment — 25+ Sources'; unlimited users; MCP Server and API included; credit rollover 3 months (monthly) / 12 months (yearly). Free trial: $0, 50 credits, no card. Enterprise custom. Credit costs: work email 1, mobile phone 10, personal email 3. as of 2026-10-19
Scout, alternatives.json v3.72026-10-19date written in Scout's pricing text
87% coverage / 96% accuracy — top of AMF study. Highest £/1k verified in the top-6 (~£47–55/1k). Right role: fill layer after Findymail/Dropcontact miss, not root (too expensive per-hit).
Lock-in risk
medium
Credited to Scout (2026-11-09, Scout last_verified). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=fill(both hosts); default_confidence_threshold=provider-composed; default_catch_all_policy=provider-composed (their waterfall handles it); default_rate_limit_rps=5; default_gdpr_basis=GDPR+CCPA+SOC2-TypeII; why_fill_not_root: highest £/1k in the top-6, only spend when cheaper root misses. [2026-10-19T22:08Z cloud refresh: 2026-10-19 price check: the page served to us shows USD ($55, JSON-LD priceCurrency USD), not the stored €55. The page may localise currency by visitor region; we did not test other regions. Plan name Pro and 1k credits unchanged. two_year_cost_gbp unchanged (null).] [2026-10-19T22:08Z cloud refresh: Q4-2026 review: price re-read; currency on the page served to us is USD. Verdict unchanged. Pricing for this entry is carried by patches/02-price-refresh.json (fuller text, same facts); apply 02 before 03.] [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (no recurring free tier: free trial 50 credits (one-off); rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from terms (no access to source code); api_or_webhooks from API docs. self_hostable and data_export left out (no doc).] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: FullEnrich terms say the Services and Software are accessible in SaaS mode through the Internet. data_export left out (pricing shows CSV export and CRM push of results, but no all-data export doc).]
Kaspr
freemiumnot runnable on gaents yet
LinkedIn-sourced contact enrichment (mobile + email). Access is via Chrome extension on LinkedIn, CSV of LinkedIn URLs, or an API whose own page states: "All you need is a LinkedIn profile URL." There is no name+domain lookup path. (Scout's description)
CONFIRMED 2026-10-19: per user per month, shown in three currencies with two figures each (billing toggle; lower figure presumed annual, basis UNVERIFIED): Starter €45/€59, £39/£51, $49/$65; Business €79/€99, £69/£85, $79/$99; Free €0. Starter: unlimited B2B email, 1,200 phone credits/yr; Business 2,400/yr. Organization: annual only, min 5 users. as of 2026-10-19unverified in Scout's data
Scout, alternatives.json v3.72026-10-19date written in Scout's pricing text
DISQUALIFYING FOR OUR CORPUS: every access path requires a LinkedIn profile URL as input (verified on kaspr.io 2026-09-16). Tal's FH730 list is person_name + company_number + domain with NO LinkedIn URLs, and Talent records LinkedIn as HTTP-999-blocked for us. Kaspr cannot process our data without first solving the step that is already blocked. SEPARATELY: CNIL fined Kaspr EUR 240,000 on 2024-12-05 (SAN-2024-020) for collecting LinkedIn contact details of users who had limited their visibility — GDPR Arts. 5-1-e), 6, 12, 14, 15. Compliance deadline 2025-06-18; injunction since closed as satisfied. The prior entry cited 'French, GDPR-native' as a REASON TO PREFER it for tenant work, which inverts the evidence. Same structural LinkedIn-seat dependency as ContactOut and Wiza, both already `skip` on this shelf.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: n/a — skipped. WHY SKIP (2026-09-16, S39): (1) LinkedIn URL is a hard input requirement on every path, and our corpus has none; (2) CNIL SAN-2024-020, EUR 240,000, for exactly the collection practice we would be buying. CITATION TRAP: the EDPB hosts two pages on this sanction, headed EUR 240,000 and EUR 200,000. The 200,000 figure is WRONG — its own linked French source and Legifrance both say 240,000. Cite CNIL or Legifrance SAN-2024-020, not the EDPB summary. REVERSIBLE IF: Kaspr ships a name+domain lookup that does not need a LinkedIn URL. Then re-score — the mobile coverage was the attraction and that part was never disproved. SAME COMPANY AS COGNISM — never score these as two independent vendors or as a cheap-vs-expensive pair; they are one supplier. [2026-10-19T23:00Z cloud refresh: 2026-10-19: the page does show EUR prices (€45 Starter, €79 Business), so the 2026-09-16 note 'NOT EUR 45/79' is contradicted; the site has a currency selector. GBP prices also listed.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class proprietary from Kaspr Terms (licence to use the Solution and the Database); api_or_webhooks from kaspr.io/api. data_export left out: export of contact details is metered by Export Credits, not an all-data export. self_hostable left out.]
MillionVerifier
commercialnot runnable on gaents yet
Standalone email verifier used as the leaf in the waterfall — takes any candidate address and returns valid/invalid/risky/catch-all with a bounce-rate guarantee. (Scout's description)
Not a finder — verify-only. Doesn't help if the finder returned no candidate. US-hosted, so we use it on bigbrain leaf; Neverbounce for vpsm.
Lock-in risk
low
Credited to Scout (2026-11-09, Scout last_verified). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=verify-leaf(bigbrain); default_confidence_threshold=map "good"→pass, "catch-all"→never_send, "risky"→drop; default_rate_limit_rps=10; default_gdpr_basis=LI+opt-out (US); why_leaf_not_root: does not find, only verifies. [2026-10-19T22:08Z cloud refresh: TUNABLES: default_waterfall_position=verify-leaf(bigbrain); default_confidence_threshold=map "good"→pass, "catch-all"→never_send, "risky"→drop; default_rate_limit_rps=10; default_gdpr_basis=LI+opt-out (US); why_leaf_not_root: does not find, only verifies. [2026-10-19 Scout T2 price/licence check: Stored '~$0.005/verify' was too high: official packs run from $0.0039/verify (10K for $39) down to $0.00045 (1M for $449) and lower. /pricing now 404s; prices read from the home page and the pack table in its own /assets/js/main.js.]] [2026-10-19T22:08Z cloud refresh: Q4-2026 review: pricing read from the official homepage and its price script; per-verify cost is $0.0039 at 10k down to $0.00045 at 1M, below the stored ~$0.005. Verdict unchanged. Pricing for this entry is carried by patches/02-price-refresh.json (fuller text, same facts); apply 02 before 03.] [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from terms; api_or_webhooks from API docs; data_export n/a (verification service); self_hostable left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: hosted API (developer docs give api.millionverifier.com and bulkapi.millionverifier.com; the home page plans list no self-hosted option, agent-checked; /pricing returned 404). data_export n/a as in pass 1.]
Neverbounce
commercialnot runnable on gaents yet
Verifier used as vpsm leaf. ZoomInfo-owned; higher price than MillionVerifier but broader tenant familiarity in the EU market. (Scout's description)
CONFIRMED 2026-10-19: Pay as you go $8 per 1,000 credits ($0.008 per credit at 1,000), credits expire 12 months after purchase; Growth $49/month for up to 10,000 emails/month; Enterprise custom ('Typically used for 250,000+ emails per month'). as of 2026-10-19
Scout, alternatives.json v3.72026-10-19date written in Scout's pricing text
SMTP RCPT check MUST run from bigbrain (residential-adjacent) — never from vpsm shared IP (Spamhaus PBL risk to co-tenants). Vendor-specific verifiers for Yahoo/M365/Hotmail need community-contributed plugins for reasonable accuracy.
Lock-in risk
low
Credited to Scout (2026-11-09, Scout last_verified). gaents does not score items.
Scout tested it (full, 2026-09-08): Verify OSS SMTP-based email verifier against real UK-SME corpus (24 rows, 15 domains from Tal's FH-730 PRIME slice).
Scout's notes
TUNABLES (read from verifier.go/smtp.go): smtpCheckEnabled=false-by-default (README: 'most ISPs block port 25') — set true on bigbrain, false on vpsm; helloName='localhost'-default → set to bigbrain FQDN; fromEmail='user@example.org'-default → set to outreach@ei-outreach.com; connectTimeout=10s → 12s cold / 6s parallel; proxyURI=none → REQUIRED for vpsm (SOCKS5 to residential); gravatarCheckEnabled=false → true for high-value; AddDisposableDomains: auto-refresh from disposable-email-domains repo daily 06:00 UTC; why_adopt: only actively-shipping MIT verifier that covers the full surface. [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from LICENSE (MIT); self_hostable and api_or_webhooks from README (Go library plus self-hosted API server); data_export n/a (stateless library).]
JosieBot26/prospector-mcp-email-finder
open-sourcenot runnable on gaents yet
MCP-native email finder + verifier loop: website scrape → name→pattern generation → SMTP verify → catch-all detection. ~50/day self-throttled default. Plugs directly into Claude Code's MCP stack. (Scout's description)
Bus factor 1 (<20 stars, single-maintainer). Published 2026-03-05 — young. Worth an evaluation because MCP-native saves a wrapper layer for agents; production adoption should wait until we've run 500+ prospects through it.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout tested it (full, 2026-09-08): Test MCP-native find+verify loop against real UK-SME corpus (11/24 rows completed at handback + interim pattern established).
Scout's notes
TUNABLES: batch_size=25-default → 10 (30s MCP-tool budget); daily_cap=50-default → env-patchable; contact_name: prefer first+last from LinkedIn, fall back to first_initial+last; pattern top_n=5 client-side cap (5th-pattern hit rate <10%); why_evaluate_not_adopt: single-maintainer risk; MCP-native is compelling but not yet proven at our volume. · 2026-09-18: evaluate->skip. 11 stars/stale/single-maintainer; channel_pattern already covers its function. [2026-10-23T13:15Z cloud refresh: Sweep 2026-10-20: last commit 2026-03-05; licence unchanged (MIT).]
disposable-email-domains
open-sourcenot runnable on gaents yet
Daily-updated blocklist of disposable/throwaway email domains. Injectable into AfterShip email-verifier via AddDisposableDomains(). (Scout's description)
Requires a refresh cron; static bundled lists in verifiers age out within weeks.
Lock-in risk
low
Credited to Scout (2026-11-09, Scout last_verified). gaents does not score items.
Scout tested it (partial, 2026-09-08): Verify input list is refreshable and injectable into AfterShip email-verifier.
Scout's notes
TUNABLES: refresh_cadence: 'git pull cron 06:00 UTC daily' (upstream pushes 2-4x/week); mode=additive (keep AfterShip baked-in list as fallback); why_adopt: only reliable, current disposable-domain data source; trivial to wire. [2026-10-19T22:08Z cloud refresh: Q4-2026 review: licence field set from the repo LICENSE.txt (CC0 1.0). Repo active (2026-10-19). Verdict unchanged.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from LICENSE.txt (CC0-1.0, public domain, treated as permissive); self_hostable: plain list in the repo; api_or_webhooks left out; data_export n/a.]
Hunter.io
freemiumnot runnable on gaents yet
Domain-first email finder + verifier. Most familiar brand in the category; Starter tier is the cheapest way to bootstrap under 300 prospects/mo. (Scout's description)
58% coverage / 86% accuracy in AMF benchmark — mid-pack, and 14% bounce is dangerous. Fine as bootstrap tier (Phase 1) or verify-leaf (score ≥90 gate); NOT as production root once volume justifies the waterfall.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=verify-leaf-only (or Phase-1 sole root); default_confidence_threshold=≥90/100; default_catch_all_policy=reject if accept_all=true; default_rate_limit_rps=15 (500/min); default_gdpr_basis=LI+opt-out; why_consider_not_adopt: acceptable bootstrap when <300/mo, but 14% bounce rate makes it dangerous as production root. · TUNABLES: default_hunter_tier=free (recommend growth-annual for API+10K credits/mo). [2026-10-19T23:00Z cloud refresh: 2026-10-19 price refresh: page showed USD to this reader (stored figures were EUR 34-209; same numbers, currency may depend on location). Monthly-billing prices are now listed ($49/$149/$299).] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class proprietary from Hunter Terms (Hunter does not transfer any IP; 'Use License' section); api_or_webhooks and data_export from API docs (RESTful API to create, read, update and delete your Hunter resources, including saved leads). self_hostable left out.]
Anymail Finder
commercialnot runnable on gaents yet
Email finder with a claimed 98.9% accuracy in their own 14-tool benchmark. Strong dark-horse candidate as verify-leaf if we don't want to run MillionVerifier separately. (Scout's description)
Vendor-run benchmark self-selects — accuracy claim is directional, not gospel. Coverage in their own study was 86.4% (competitive) but our real-world hit-rate needs measurement.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=alternate-verify-leaf; default_confidence_threshold="verified"; default_rate_limit_rps=~5 (hit-based); default_gdpr_basis=GDPR+CCPA; why_evaluate_not_adopt: promising cheap verify-leaf; needs a 500-prospect A/B against MillionVerifier to justify the swap. [2026-10-19T22:08Z cloud refresh: 2026-10-19 freshness check: stored '€26/mo entry' is out of date. Official page now prices in USD and the entry plan is $29/month for 400 credits. two_year_cost_gbp: agent-chose basis = entry plan $29/month x 24 = $696, GBP at £0.7502 per $1 (shared brief rate, 2026-10-19) = £522.14; the yearly entry plan ($228/year) would be £342.09.] [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from terms; api_or_webhooks from API docs (also lists webhooks and bulk download). self_hostable and data_export left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: hosted API (api.anymailfinder.com; anymailfinder.com/pricing lists only hosted plans, agent-checked). data_export left out (pass-1 notes a bulk download feature, but no all-data export doc).]
LeadMagic
commercialnot runnable on gaents yet
B2B social→email (LinkedIn URL → email, 5 credits/hit). Mid-pack in WizLeads benchmark. Unmetered rate on higher tiers. (Scout's description)
CONFIRMED 2026-10-19: Plans (monthly / annual per month): Essential $99 / $83 (5k credits/mo); Growth $249 / $208 (20k); Professional $499 / $416 (50k, 5 RPS); Ultimate $849 / $708 (100k, 10 RPS). 'Failed lookups are free'. One credit pool for all endpoints, from $0.007 a credit. Team access on $499+ plans needs $49 per seat per month. as of 2026-10-19
Scout, alternatives.json v3.72026-10-19date written in Scout's pricing text
Mid-pack in the one benchmark that included it; probably not top-tier accuracy. Interesting as a fill-layer alternative if Kaspr rate limits bite.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=fill-alternate; default_confidence_threshold=1_credit=valid; default_catch_all_policy=zero-charge-on-fail; default_rate_limit_rps=10 unmetered Pro+; default_gdpr_basis=LI+opt-out (US-hosted); why_evaluate_not_adopt: mid-pack accuracy; only relevant if Kaspr rate limits become the bottleneck. [2026-10-19T22:08Z cloud refresh: 2026-10-19 check: entry tier is now Essential at $99/month; the stored $49.99 floor no longer appears. Rate limits now stated as 5 RPS on Professional and 10 RPS on Ultimate. two_year_cost_gbp unchanged (null).] [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from terms; api_or_webhooks from docs; data_export n/a (enrichment API); self_hostable left out.]
Apollo.io
freemiumnot runnable on gaents yet
Huge B2B contact DB (~275M contacts claimed). API accepts linkedin_url. Best-in-class for discovery (finding contacts you didn't know existed); weaker as a verifier. (Scout's description)
81% accuracy / 7.2% bounce (b2b study) — bounce rate is dangerous for cold outreach. Contact-scraping posture flagged by multiple analysts. NOT on vpsm default path — needs per-tenant DPA opt-in. Annual credit cap (48k/yr) is a hard ceiling.
Lock-in risk
medium
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=DISCOVERY-ONLY (never as verifier); default_confidence_threshold=email_status:'verified' only; default_catch_all_policy=re-verify with 2nd tool; default_rate_limit_rps=depends on plan; default_gdpr_basis=DPA-per-tenant-required; why_evaluate_not_default: 7.2% bounce and DPA overhead disqualify it as default; discovery use case only. · Scout 2026-09-19 head-to-head with Hunter for verify-at-scale (Judge-40 pattern): Hunter Growth wins on credit ceiling (10K/mo vs Apollo Pro 4K/mo shared) + existing find-contact wiring. Apollo stays DISCOVERY-only per S36; Apollo's 7.2% bounce rate is dangerous as verifier. · Scout 2026-09-19 head-to-head with Hunter for verify-at-scale (Judge-40 pattern): Hunter Growth wins on credit ceiling (10K/mo vs Apollo Pro 4K/mo shared) + existing find-contact wiring. Apollo stays DISCOVERY-only per S36; Apollo's 7.2% bounce rate is dangerous as verifier. · Scout 2026-09-19 head-to-head with Hunter for verify-at-scale (Judge-40 pattern): Hunter Growth wins on credit ceiling (10K/mo vs Apollo Pro 4K/mo shared) + existing find-contact wiring. Apollo stays DISCOVERY-only per S36; Apollo's 7.2% bounce rate is dangerous as verifier. [2026-10-19T23:00Z cloud refresh: 2026-10-19 session 2: Apollo's llms.txt says 'Prices are served live on the pricing page and change over time'. Only the monthly figures are in the page's structured data.] [2026-10-20T01:33Z cloud refresh: type commercial-saas -> freemium (recurring free allowance: Free $0 plan; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from Apollo terms; api_or_webhooks from docs.apollo.io; self_hostable and data_export left out (no export doc read).] [2026-10-23T13:15Z cloud refresh: 2026-10-20: annual prices read after scrolling the rendered page; 'Pro $79/mo annual' confirmed; Basic monthly is $65, not the $49 in the JSON-LD.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): data_export true from Apollo pricing FAQ (bulk-export everything you have saved; export limits apply per plan). self_hostable left out (no statement).]
**259 hits per 1k in WizLeads** — floor of the seven-tool pack. Marketing claims not backed by independent measurement. Skip as primary or fill.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: N/A — skipped. why_skip: 259/1k floor in the one independent benchmark that tested it; no path to justify over Findymail or Dropcontact. [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: 2026-10-20: '$99/mo Growth, 5k credits' confirmed; other tiers added.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: api_or_webhooks from REST API docs (enrich, bulk enrich, search endpoints, plus MCP server). licence_class left out: terms page is JS-rendered and redirected to the homepage in the headless browser. self_hostable and data_export left out.]
RocketReach
commercialnot runnable on gaents yet
Contact-database SaaS with API on the top tier only. (Scout's description)
10k lookups/year for $1,699 = $0.17/lookup — 17× the pay-per-hit cost of Findymail. Beaten on accuracy by Findymail in independent benchmarks. Rate limits undocumented publicly.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: N/A — skipped. why_skip: cost floor 17× competitors + beaten on accuracy + rate limits undisclosed. · 2026-09-18: WHY it's the de-facto AI-assistant contact source — RocketReach shipped an OFFICIAL first-party MCP (~2026-07-22): client-neutral (Claude/ChatGPT/Cursor), OAuth 2.1, one-command install, bills against existing credits with no extra subscription. NOT community adoption — every community rocketreach-mcp repo is 0-1 stars (verified gh search 2026-09-18). Verdict UNCHANGED skip: mechanism is Hunter-style (scrape+pattern+SMTP), API floor $1,699/yr, credit-metered, wrong fit for holding a corpus. The MCP explains why OTHERS reach for it; it does not change our build-vs-buy. [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: 2026-10-20: all prices changed since the 2026-10-02 correction (was $229 / $619 / $1,379 annual, $39 / $79 / $169 monthly). The earlier remark that API lookups use export credits came from a non-official source and was not re-checked.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from Terms 'Use License' (limited, revocable licence to access the Services); api_or_webhooks from API reference (Company and People lookup/search APIs). self_hostable and data_export left out.]
Cognism
commercialnot runnable on gaents yet
Enterprise B2B intelligence. UK-headquartered. Phone-verified 'Diamond Data' for mobile numbers. Compliance-first posture. (Scout's description)
$36k/yr MEDIAN real cost (corrected 2026-09-16; previously recorded as a ~$22.5k 5-seat floor, which understated it by ~60%). Platform fee + per-user. Contact-sales only — no public price. as of: no date recorded for this field
Scout, alternatives.json v3.72026-10-23T13:15ZScout patch_history.at; no date recorded for this field itself
Scout's judgement
Verdict
park
Limits for us
$22.5k/year floor is disproportionate to our need. Only justifies itself if a tenant needs **verified mobile numbers at scale** — Kaspr covers EU-GDPR emails at ~5% the cost. · 2026-09-16 CORRECTION: real median is $36k/yr not $22.5k. AND — Cognism and Kaspr are THE SAME COMPANY, so pairing them was never vendor diversity, it was one supplier twice. Kaspr is now `skip` (LinkedIn-URL mandatory + CNIL SAN-2024-020 EUR240k), which removes the cheap-substitute argument that made Cognism easy to defer: the choice is now Cognism at $36k or neither.
Lock-in risk
high
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: N/A until adopted. why_park: $22.5k/yr floor + Kaspr covers 80% of the value at 5% the cost. Revisit only if a specific tenant demands verified mobile. · 2026-09-16 (S39): cost corrected $22.5k->$36k median; same-company-as-Kaspr recorded so the two are never again scored as independent options. · 2026-09-18 MOBILE REFRAME (identity fork): a director-mobile number for a CH company DOES exist as a product — Experian/Equifax credit-header data via LexisNexis TraceIQ, with a CH-number->director-mobile path. BUT every holder's permitted-purpose list ends at debt collection and EXCLUDES prospecting. So the correct framing for D787/phone is 'everyone who has it forbids our use', NOT 'nobody has it'. Cognism's own published UK conversion is 10-15% (use that, not a guessed 20%, in any spend model). [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from Cognism terms of website use (covers applications hosted at cognism.com). api_or_webhooks left out (integrations page only; /api 404). self_hostable, data_export left out.]
ContactOut
commercialnot runnable on gaents yet
Chrome-extension email finder that runs on top of the user's own LinkedIn session. (Scout's description)
CONFIRMED 2026-10-20: Free $0 (5 emails, 5 phones, 5 exports per day); Email $49/month (shown as 'was $99', 'Limited time only'; unlimited emails, 300 exports/month); Email + Phone $99/month (was $199; 600 exports/month); Team / API: 'Contact us'. 1 user per company on self-serve plans. as of 2026-10-20
Scout, alternatives.json v3.72026-10-20date written in Scout's pricing text
Chrome-extension model = **operator's LinkedIn seat carries the ToS ban risk**, not the vendor. No vendor in this category indemnifies the operator against account loss. Categorically unsuitable for Adam's seat; burner-seat-only exception noted.
Lock-in risk
medium
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: burner_seat_only=true (never Adam's seat); why_skip: seat-ban risk on operator, not vendor; Adam's LinkedIn is his relationship graph — categorically bad trade. [2026-10-20T01:33Z cloud refresh: type commercial-saas -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: 2026-10-20: stored '$79-$399/mo tiers' replaced. Prices shown are a time-limited discount.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from ContactOut Terms. api_or_webhooks from api.contactout.com docs (API key on request via a meeting). self_hostable, data_export left out.]
Wiza
freemiumnot runnable on gaents yet
LinkedIn-native Chrome-extension finder. Top hits in WizLeads benchmark (633/1000). (Scout's description)
Same Chrome-ext seat-risk profile as ContactOut. Marketing strength doesn't offset the categorical risk to Adam's LI seat.
Lock-in risk
medium
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: burner_seat_only=true; why_skip: same as ContactOut — operator seat carries the ban, not the vendor. [2026-10-19T23:00Z cloud refresh: 2026-10-19 price refresh: top tier is $199/user/mo monthly (stored '$83-$249'). type 'commercial-saas' -> 'freemium'.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: api_or_webhooks from docs.wiza.co (Wiza API Documentation); data_export from home page (bulk export prospects to CSV or CRM). licence_class left out: terms page URLs tried return 404. self_hostable left out.]
reacherhq/check-if-email-exists
open-sourcenot runnable on gaents yet
Rust-based SMTP-probe email verifier. 9,535★ (most-starred in category). Vendor-specific verifiers for Yahoo/M365/Hotmail via headless-Playwright fallbacks. (Scout's description)
**AGPL-3.0-or-Commercial** — the AGPL network-copyleft clause triggers when we run it as a service that talks to our agents. Choice: buy the commercial licence, or AGPL-license every downstream agent that hits its HTTP endpoint. AfterShip/email-verifier (MIT) is 84% of the codebase quality without the licence gate.
Lock-in risk
medium
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES (when/if commercial licence bought): RCPT_TIMEOUT=30000ms → 12000ms for cold-outreach batch; SMTP_HELLO_DOMAIN=(unset — fails) → set to bigbrain FQDN; PROXY_URL=none → SOCKS/HTTP for IP diversity; MICROSOFT365_API_URL=none → REQUIRED (M365 blocks SMTP RCPT); YAHOO_METHOD='headless' → keep (headless > api > smtp on accuracy); why_park: commercial licence unpriced-yet; AfterShip covers 84% of the surface under MIT. [2026-10-23T13:15Z cloud refresh: Sweep 2026-10-20: last commit 2026-10-16; licence changed: MIT.]
joeyism/linkedin_scraper
open-sourcenot runnable on gaents yet
Python LinkedIn profile / company scraper. Works; well-starred. (Scout's description)
GPL-3.0 is viral for commercial SaaS embedding. Adam's already-adopted **agent-browser** (Vercel, ADOPT-NOW) drives LinkedIn UI properly with no licence entanglement.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: N/A — skipped. why_skip: GPL viral for our commercial SaaS + agent-browser (already ADOPT) does the same job licence-clean. [2026-10-23T13:15Z cloud refresh: Sweep 2026-10-20: last commit 2026-04-09; licence unchanged (GPL-3.0).]
laramies/theHarvester
open-sourcenot runnable on gaents yet
OSINT domain harvester — collects emails/subdomains/hosts across a domain. Used in recon workflows. (Scout's description)
17,256★ but **no LICENSE file** — legal ambiguity treats as all-rights-reserved. Not production-safe. Also wrong tool: recon, not per-contact enrichment. · 2026-09-18 CORRECTION: was recorded NOASSERTION/skip on a WRONG licence fact. gh api: GPL-2.0 WITH a LICENSE file, 17,531 stars, pushed 2026-09-18 (today, most active in the space). GPL-2.0 is viral for DISTRIBUTION; we would wrap it as a CLI we call, not redistribute, so the licence is not a blocker for our use. Verdict skip->evaluate.
Lock-in risk
low
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: N/A — skipped. why_skip: missing LICENSE = legal ambiguity + wrong problem shape (broad recon vs targeted enrichment). · 2026-09-18: licence + stars corrected; skip was based on a licence fact that was wrong. [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from LICENSE (GPL-2.0); self_hostable from README Docker Compose; api_or_webhooks from README REST API; data_export from README (run evidence export).]
treg (tools-registry)
source-availablenot runnable on gaents yet
An aggregation layer over roughly 60 people-search and enrichment providers — 'withheld for agent tools'. One interface, one prepaid balance, pay-per-result at the provider's own rate with no markup. Self-hostable. (Scout's description)
tools-registry License — Apache-2.0 plus overriding Additional Terms (hosted-service restriction; contributor commercial-use assignment). Source-available, not OSI open source.
$0.0089 per verified work email found, metered at the underlying provider's rate from a prepaid balance, 0% markup, misses free, $1.00 free credit per new verified team. No subscription. as of: no date recorded for this field
Scout, alternatives.json v3.72026-10-23T13:15ZScout patch_history.at; no date recorded for this field itself
GitHub stars
8026 as of: no date recorded for this field
Scout, alternatives.json v3.72026-10-23T13:15ZScout patch_history.at; no date recorded for this field itself
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
HEADLINE CHECK — BOTH MARKETING CLAIMS FAIL. (1) 'Fully open source' is OVERSTATED: the LICENSE is a bespoke 'tools-registry License' = Apache-2.0 PLUS Additional Terms that take precedence, including a hosted-service restriction (no offering it as SaaS/managed/embedded to third parties without written authorisation) and a contributor term assigning commercial use of contributions to the licensor. Source-available, not OSI open source. GitHub is right to refuse to classify it. PRACTICALLY FOR US THIS IS FINE — self-hosting for our own use is expressly permitted and encouraged; we just should not repeat 'fully open source'. (2) '#1 on People Search Bench' is MISLEADING. PeopleSearchBench is a REAL third-party benchmark (github.com/LessieAI/people-search-bench, 137 stars, MIT, 119 multilingual queries, arXiv 2603.27476) and TREG DOES NOT APPEAR ON ITS LEADERBOARD — published rows are Lessie 65.2, Exa 55.0, Claude Code 46.0, Juicebox 45.8. The '#1' lives on treg's own blog and measures the same agent WITH vs WITHOUT the treg plugin: their harness, their write-up, not an adjudicated rank. This is the S31 'own prior version' cherry-pick, with a real benchmark's name borrowed for credibility. (3) The 'Claude for people-search' framing is the author's; treg's own site disclaims any Anthropic relationship. | THE $0.0089 FIGURE IS REAL and is the genuinely interesting part. | WHY EVALUATE, AND THE REASON IS ARCHITECTURAL NOT PRICE: this shelf already has 21 researched entries and four adopted providers (Findymail, Dropcontact, FullEnrich, Kaspr). treg's value is that it is the LAYER ABOVE them — the same relationship withheld has to the model providers we adopted in S34. If aggregation works it collapses four integrations into one; if it does not, we are out a $1 credit. | THE QUESTION FOR CHAD is narrower than 'should we use it': do we want a broker between us and four providers we have already contracted with? A broker adds a failure point and a data path. Against that: 0% markup, misses free, self-hostable. | TUNABLES: prepaid balance cap / unset / SET IT LOW for the trial / an aggregator with an open balance is an open bill. provider allow-list / all ~60 / PIN to the four we already adopted / otherwise the comparison is not apples-to-apples and we are silently buying from unvetted vendors. hosting / treg.to hosted / SELF-HOST / the hosted path routes our prospect list through a third party, which is precisely what choosing Findymail direct avoided. [2026-10-19T23:00Z cloud refresh: [2026-10-19 licence check] Repo LICENSE (HEAD 2026-10-19) is Apache-2.0 plus Additional Terms that 'take precedence over the Apache License'. Term 1 bars providing it as a hosted, managed or embedded service to third parties without written authorisation. Internal use is expressly permitted. With the extra restriction it is no longer an OSI licence, so type set to source-available under Scout ruling 1. CONFIRMED.] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: api_or_webhooks from README (HTTP call proxy); data_export left out.]
Snov.io
commercialnot runnable on gaents yet
Email finder + verifier + drip-campaign SaaS with REST API from Starter tier upward. Waterfall-adjacent to Hunter/Apollo/Findymail for cold-outreach discovery. (Scout's description)
$39/mo Starter (1K credits, 500 API req/hr); $99 Pro S (5K); $189 Pro M (20K). Credits cover finder + verifier + drip. API from Starter. as of: no date recorded for this field
Scout, alternatives.json v3.72026-10-23T13:15ZScout patch_history.at; no date recorded for this field itself
evaluate-only: recommended NOT to adopt today because the head-to-head vs Hunter Growth is decisive on both credits and integration cost. Never freeze this verdict — revisit if the Hunter path fails.
Limits for us
No independent accuracy benchmark at our scale; loses head-to-head vs Hunter Growth on credit ceiling + established prior + existing find-contact wiring. New-adapter integration cost 3-4h.
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: default_waterfall_position=alt-verify-leaf (only when Hunter path breaks); default_api_rate_limit_starter=500req/hr. Trigger to promote: Hunter Growth quota shape changes unfavourably OR Judge-40 scales past Hunter's 10K/mo. [2026-10-20T01:33Z cloud refresh: type None -> commercial (proprietary; pricing paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: api_or_webhooks from snov.io/api. licence_class left out (terms speak only of trademarks, no software licence wording found). self_hostable, data_export left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): licence_class proprietary from Snov.io terms (no IP transferred; all rights stay with Snov.io). self_hostable false: hosted API (api.snov.io) and snov.io/pricing lists only hosted plans (agent-checked). data_export left out (no doc).]
GetProspect
freemiumnot runnable on gaents yet
Email finder (name + domain via POST /v2/email/find) + verifier. (Scout's description)
Free: 50 valid emails + 100 verifications/mo, recurring (unused credits roll over up to one month's allowance). $49/mo for 1,000 valid emails + 2,000 verifications. as of: no date recorded for this field
Scout, alternatives.json v3.72026-10-23T13:15ZScout patch_history.at; no date recorded for this field itself
adopt-as-is for a test; don't take its 'valid' as ground truth on catch-all domains.
Limits for us
UK SME coverage unmeasured. Whether signup needs a business email is UNVERIFIED (signup page returned 403).
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
TUNABLES: charges only for a valid email; not_found and accept_all results are refunded, duplicates free. API on all plans; not rate-limited beyond credits (402 when empty). Sources: getprospect.com/pricing, /api-docs/credits-and-limits.md. [2026-10-20T01:33Z cloud refresh: type None -> freemium (recurring free allowance: Free 50 valid emails + 100 verifications/mo; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: api_or_webhooks from API docs; licence_class left out (terms page had no licence-grant text I could read); data_export left out; self_hostable left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): data_export true from GetProspect pricing (leads and contact lists export to CSV/XLSX). licence_class left out (terms pages returned 403/410). self_hostable left out.]
Datagma
freemiumnot runnable on gaents yet
Person enrichment: email and phone finder via REST API. (Scout's description)
strip-first: measure precision on UK SMEs before weighting.
Limits for us
Its terms let it refuse personal and generic addresses at signup, and they ban creating accounts by automated means.
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
API on all plans (datagma.com/en/pricing). Sign up with a person-style company address. Free-plan rate limit UNVERIFIED. [2026-10-19T22:08Z cloud refresh: 2026-10-19 check: free tier is now 100 shared credits/month, not '90 emails + 3 phones'. The $49 plan is 3,000 credits shared between emails and phones, not 3,000 emails plus 100 phones. two_year_cost_gbp unchanged (null).] [2026-10-20T01:33Z cloud refresh: type None -> freemium (recurring free allowance: Free $0 forever: 100 credits/month; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from Datagma terms; api_or_webhooks from datagma.com/en/api; self_hostable false: hosted API gateway (same page); data_export n/a (enrichment lookups). doc.datagma.com was blocked by the proxy.]
Minelead
freemiumnot runnable on gaents yet
Domain and person email finder. (Scout's description)
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
No charge when no valid email is found; duplicate searches in the same month are free. Sources: tomba.io/pricing, tomba.io/api. [2026-10-20T01:33Z cloud refresh: type None -> freemium (recurring free allowance: Free 25 searches/mo recurring; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from Tomba terms; api_or_webhooks from docs.tomba.io; self_hostable and data_export left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: hosted API (pricing page gives base URL https://api.tomba.io/v1; plans listed there have no self-hosted or on-prem option, agent-checked). data_export true: pricing table shows bulk-job results can be exported on paid plans (not on the free plan).]
Enrow
commercialnot runnable on gaents yet
Email finder with a documented API (10 req/s per key; a bulk POST counts as one request). (Scout's description)
CONFIRMED 2026-10-19: 50 free credits. Start $17/month for 1,000 credits (1,000 emails, 25 phone numbers); Start 4,000 credits $47; Pro $87/month for 10,000 credits (credits roll over); Scale $397/month for 50,000 credits; Custom 200,000+. Prices from JSON-LD in USD. UNVERIFIED: EUR prices (page has a EUR/USD toggle rendered by JavaScript). as of 2026-10-19unverified in Scout's data
Scout, alternatives.json v3.72026-10-19date written in Scout's pricing text
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
Useful as the paid fallback when free pools run dry. Sources: enrow.io/pricing, docs.enrow.io/rate-limits. [2026-10-19T22:08Z cloud refresh: 2026-10-19 freshness check: stored 'EUR 15/mo for 1,000' could not be confirmed. Official page lists the cheapest paid plan as 'Start — 1,000 credits' at USD 17. Plans now include phone credits and an MCP server. two_year_cost_gbp: agent-chose basis = Start 1,000 credits $17/month x 24 = $408, GBP at £0.7502 per $1 (shared brief rate, 2026-10-19) = £306.08.] [2026-10-20T01:33Z cloud refresh: type None -> commercial (no recurring free tier: 50 free credits (one-off); rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from terms; api_or_webhooks from API page (REST + webhook delivery). self_hostable, data_export left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: hosted API (api.enrow.io; enrow.io pricing lists only hosted plans incl. Custom, agent-checked). data_export left out (no doc).]
QuickEmailVerification
freemiumnot runnable on gaents yet
Email verifier (SMTP etc. from their IP pool) with API and sandbox. (Scout's description)
Benchmark against Reoon on known-good and known-bad M365 addresses before trusting its verdicts.
Limits for us
Catch-all status field not confirmed; rate limit not published.
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
Largest recurring free verifier. Signup asks for a work email + company; no card or phone seen. API on free per the vendor's search snippet (the /plans page wasn't explicit). [2026-10-20T01:33Z cloud refresh: type None -> freemium (recurring free allowance: Free 100 verifications/day recurring; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from terms (IP in all software and source code kept by vendor); api_or_webhooks from API page; data_export n/a (verification service).] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: QuickEmailVerification terms say it is an online service with no software delivered. data_export n/a (verification service).]
Verifalia
freemiumnot runnable on gaents yet
Email verifier with a mature API and open-source SDKs; `ServerIsCatchAll` status. (Scout's description)
A free-plan terms update from Dec 2025 was not read.
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
API on free confirmed. 18 req/s per IP, burst 45. Signup: work email, name, country; no phone or card. [2026-10-20T01:33Z cloud refresh: type None -> freemium (recurring free allowance: Free 25 verifications/day recurring; paid plans exist; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from Verifalia terms; api_or_webhooks from developer docs; self_hostable false: hosted RESTful API (same page); data_export from API docs section 'Exporting entries' (job results export).]
CONFIRMED 2026-10-19: Free: '100 free credits daily', no card. Pay-as-you-go (credits never expire, minimum 1,000): 1,000 at $0.0040/email ($4); 2,000 at $0.0025; 5,000 at $0.0018; 10,000 at $0.0015 ($15); 25,000 at $0.00116; lower per-email rates at higher volumes. Subscriptions: minimum 5,000 credits/month. as of 2026-10-19
Scout, alternatives.json v3.72026-10-19date written in Scout's pricing text
Whether the API works on the free tier is UNVERIFIED: the pricing page says API is for paid plans, the homepage says registered users get it.
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
30/min on the single-email API. [2026-10-19T22:08Z cloud refresh: Price changed: stored '$12 per 10k'; the official page's price table now gives $0.00150 per email at 10,000, which is $15 per 10k. Prices were read from the tier table embedded in the page script (the page fills prices with JavaScript). two_year_cost_gbp: agent-chose basis = 10,000 PAYG verifications per month x 24 months at $15 = $360, at £0.7502 per $1 = £270.07.] [2026-10-20T01:33Z cloud refresh: type None -> freemium (proprietary; pricing has free tier + paid; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from site footer (all rights reserved, closed service); api_or_webhooks from homepage; self_hostable left out; data_export left out (results download as CSV, but the site says lists are deleted after download, so not a stored dataset).] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: hosted API (base URL https://client.myemailverifier.com; pricing page lists only hosted credit plans, agent-checked). data_export left out as in pass 1.]
ZeroBounce
freemiumnot runnable on gaents yet
Email verifier with a strong API and a catch_all_domain flag; also gives 10 finder credits a month. (Scout's description)
ToS: one free account per person or business. Buying credits ends the free status.
Credited to Scout (2026-10-23T13:15Z, Scout patch_history.at). gaents does not score items.
Scout's notes
Never charges for 'unknown'. Source: zerobounce.net/docs/getting-started/freemium-terms-and-conditions. [2026-10-20T01:33Z cloud refresh: type None -> freemium (recurring free allowance: Free 100/mo recurring; rule set agent-chose 2026-10-20, Adam to rule)] [2026-10-23T13:15Z cloud refresh: Fact fields 2026-10-20: licence_class from ZeroBounce terms; api_or_webhooks and data_export (results download) from ZeroBounce docs; self_hostable left out.] [2026-10-23T13:15Z cloud refresh: Fact fields pass 2 (2026-10-20): self_hostable false: ZeroBounce terms define the product as the ZeroBounce Hosted Service, accessed via the site.]
Lusha
freemiumnot runnable on gaents yet
Contact and company data with browser extension and CRM sync. (Scout's description)
Cost basis: Starter annual x2. two_year_cost_gbp basis: one acquired company, stated seats/plan, list price x 24 months, ex VAT, USD x0.75 and EUR x0.87 where no GBP shown; staff time not counted. Judgement fields (integration_hours, lock_in_risk, feature_overlap, verdict) are Scout judgements for a 10-250 staff UK B2B maker/distributor. Playbook part 4 (research/playbook-2026-10-23/04-tools-sales-marketing.md).
ZoomInfo
commercialnot runnable on gaents yet
US-centric sales intelligence; London EMEA HQ. (Scout's description)
two_year_cost_gbp basis: one acquired company, stated seats/plan, list price x 24 months, ex VAT, USD x0.75 and EUR x0.87 where no GBP shown; staff time not counted. Judgement fields (integration_hours, lock_in_risk, feature_overlap, verdict) are Scout judgements for a 10-250 staff UK B2B maker/distributor. Playbook part 4 (research/playbook-2026-10-23/04-tools-sales-marketing.md).