Find indexed ecommerce and website records with available emails, phones and technology filters. 73.9M+ overlapping source rows checked 8 Nov 2026. Export CSV, JSON or Excel. $3/1K records; start fee extra, platform usage included.
Candidate — preserve additional warmed segments (25 October 2026)
Restore all standard source segments and retain 17 additional warmed segments.
The combined registry has 118 segments; source-record counts include overlap.
Require the matching 118-table server readiness generation before a new export.
Earlier 91-segment warmed cursors must restart at offset 0 on this registry.
Candidate — restored warmed coverage (25 October 2026)
Map relocated platform tables explicitly and import missing source segments.
Require the matching server coverage revision and all 101 source-table checks.
Keep warmed as the selected default and Standard explicitly available.
Changed warmed registry rejects old continuation tokens; restart at offset 0.
Candidate QA and owner review are required before changing latest.
Candidate 0.17.18 — fewer output-write requests (25 October 2026)
Write up to5,000 result rows per SDK call instead of500.
Keep accepted-row counts authoritative for partial budgets and continuation.
The cloud SDK splits payloads by byte size and applies PPE limits; no fee bypass.
Retain the owner-selected warmed default and explicit standard source option.
Candidate only, pending exact cloud QA and owner review.
Candidate 0.17.17 — warmed source by default (25 October 2026)
Owner-selected default: warmed62M+ source records, with standard70M+ still available explicitly.
Apply the same default in both the input schema and source selection code.
Explain the different coverage and explicit source selection for older saved inputs.
Retain stable pagination, budget-aware pages, eight-query bounds and unchanged prices.
Candidate only: publication still requires approval of the exact new build and QA card.
Candidate 0.17.16 — budget-aware source pages (25 October 2026)
Size each source page to the remaining paid output allowance and requested offset.
Keep at most 500 rows per source page, with the existing stable merge order.
Read at most eight source queries concurrently, with eight extra page buffers.
Preserve the existing page size when traffic filtering changes eligibility.
Candidate only; compare first and late-offset exports before claiming speed gains.
Candidate 0.17.15 — bounded page prefetch and timings (25 October 2026)
Read ahead source pages concurrently instead of serial refills at page boundaries.
Preserve the four-query limit and keep at most four extra page buffers.
Retain stable ordering, deduplication and accepted-row charging semantics.
Ignore failures of unused speculative pages and stop refilling after the row limit.
Add safe SUMMARY timings for initial fetch, SQL page work, replay and Dataset writes.
New candidate only; latest remains unchanged pending owner review.
Candidate 0.17.14 — stable default pagination (25 October 2026)
A 50,000-row cloud export found 31 repeated domains across resumed parts.
Default SQL now orders each source by Root Domain before LIMIT/OFFSET.
Cursor version 2 rejects older signed sessions before queries or paid rows.
Restart exports from offset 0 after this change; old offsets belong to the old order.
No claim of hundreds of thousands of unique sites in seconds. New owner review required.
Candidate 0.17 — warmed-copy preview (25 October 2026)
Optional server-selected warmed sources; standard coverage stays the default.
Signed source sessions pin continuation and migration to the same copy.
Missing optional source columns are projected as null; no invented enrichment.
Candidate only: latest changes require the owner review card.
Changelog
0.18 hotfix — 2026-09-28
Classify missing PostgreSQL tables as configuration errors that cannot be repaired by retrying.
Include safe segment identifiers in error diagnostics without revealing connection strings.
Prioritize partial coverage over the export row cap in SUMMARY.
Mark continuation unsafe when a source failed and provide original-input retry instructions: restored sources can change global sorting and offsets.
Prepared source-registry synchronization with the validated current table mappings after shard migration; deployment awaits authorization to upload the protected connection configuration.
0.18 — 2026-09-19
Added explicit Auto, Export and Count workflows for API and MCP clients.
Added Compact, Contacts, Sales, Traffic, Full and Custom output presets while preserving manual columns.
ERRORS is now always a stable array, including successful runs with no errors.
Reorganized the Console form into five numbered sections.
[0.17] — 2026-09-18
Added
Optional traffic-only filtering, minimum and maximum monthly-visit thresholds,
and highest/lowest monthly-visit ordering.
Traffic controls automatically enable enrichment for Console, API and MCP
callers.
SUMMARY.trafficEnrichment now reports matched rows, filtered rows, coverage,
active thresholds and lookup failures.
Fixed
Websites rejected by traffic filters are excluded before Dataset billing.
Offset and copy-ready continuation now page over eligible traffic-filtered
leads instead of raw source rows.
Invalid traffic ranges return structured diagnostics without paid rows.
[0.16] — 2026-09-18
Changed
Website traffic enrichment is now off by default so ordinary lead exports
remain fast.
When enabled, traffic is resolved once per 10,000 selected websites instead
of once per 500-row Dataset batch, while Dataset writes remain safely batched.
Long exports now report visible progress while traffic is added and rows are
written.
[0.15] — 2026-09-16
Added
Mixed website datasets can now be searched without assuming that every row
matches the technology named by its source table.
Additional compatible website datasets are discovered automatically.
Verified coverage increased to 69,954,034 source rows with no unavailable segments.
Fixed
A full database can be attached through a read-only logical mapping without
requiring schema changes in that database.
[0.14] — 2026-09-16
Added
Coverage for newly provisioned data segments across supported platforms.
Fixed
Empty data segments no longer create user-facing query failures while awaiting data.
Incomplete segment discovery is reported through structured run diagnostics.
[0.13] — 2026-09-14
Added
Optional domain-level enrichment from the 40M+ traffic dataset.
Monthly visits, global traffic rank, traffic growth, bounce rate, pages per
visit, average visit duration, Search/Direct/Social/Ads/AI traffic shares and
the traffic-data timestamp.
Traffic enrichment coverage is reported in the structured run summary.
[0.12] — 2026-09-11
Changed
All 59 output fields are now visibly selected by default in new or reset
Actor input forms, so every dataset view is populated without extra setup.
Callers can still deselect fields for compact exports; Root Domain and
_platform remain mandatory identifiers.
[0.11.1] — 2026-09-11
Fixed
Single-platform exports now always include _platform, matching the Overview
and Freshness dataset-view schemas.
Input help now explains that dataset tabs only display fields saved by the
run; leave Output columns empty to populate all views with all 59 fields.
[0.11] — 2026-09-11
Added
Explicit machine-readable output and dataset schemas for hosted Apify MCP and
AI-agent structured-output discovery.
Stable SUMMARY and ERRORS records with completion state, retry guidance,
failed-segment diagnostics and paid-row counts.
Formal item schema for advanced filters and an AI/MCP contract in the README.
Changed
Default export size is 50 rows so first runs and autonomous agent samples are
inexpensive. Large exports remain available up to 250,000 rows.
Invalid platform input and runtime diagnostics no longer create paid dataset rows.
[0.10] — 2026-09-10
Added
Cross-link to the dedicated Tech Stack Leads & Lookup Actor for reverse
technology prospecting, AND/OR/NOT combinations and domain enrichment.
Security
Hardened local release tooling so authentication material is never included
in Actor source files or build images.
[0.9] — 2026-09-10
Added
Every capped export now saves EXPORT_CONTINUATION in the Key-Value Store with
the exact nextOffset and a copy-ready resumeInput for the next run.
Fixed
countOnly no longer creates paid dataset items. Counts are saved to the
Key-Value Store as COUNT_SUMMARY.
Continuation offsets now account for source rows removed by domain
deduplication, preventing repeated scanning on the next run.
[0.8] — 2026-09-07
Changed
Per-run limit raised 100K → 250K rows. There is no total cap — use Offset
to page through unlimited results across runs without paying twice. When a run
returns the maximum, the status message now shows the exact Offset to use next.
[0.7] — 2026-09-07
Added
Expanded WordPress coverage: 12.0M → 20.6M sites (+8.6M). Total database
grew from ~52M to ~60M sites across the same 14 platforms.
[0.6] — 2026-08-21
Performance
Bulk exports are significantly faster, especially for large result sets.
Paginated exports and resumed runs now skip previously processed records faster.
Count-only previews complete faster across multiple selected platforms.
Fixed
When a run reaches its spending limit, the final status now reports the exact
number of rows actually delivered and explains how to request more.
[0.5] — 2026-08-18
Fixed
Long-running exports preserve progress and domain deduplication more reliably,
preventing already-delivered records from being emitted again.
[0.4] — 2026-08-01
Added
Six Output tabs for faster scanning, with icons: 🏢 Overview,
📧 Contacts, 💰 Firmographics, 📊 Traffic & Authority, 🛒 Tech Stack,
📍 Location. No data changes — just cleaner views over the 59 fields.
[0.3] — 2026-08-01
Added
New platform: Angular — Angular-built sites are now selectable in the
Platforms input.
Fixed
Selecting a temporarily unavailable platform no longer crashes the run.
Unavailable platforms are skipped with a warning; if none of the selected
platforms are available the run ends cleanly with a clear message (no charge).
[0.2] — 2026-07-21
Added
New platform: WordPress — ~12M WordPress CMS sites.
Expanded coverage: database grew from 36.2M to ~52M sites.
Fuller coverage for Shopify, WooCommerce, WooCommerce Checkout, and Squarespace.
Continuous enrichment — new sites are added and existing records receive
refreshed contact data, social profiles, and technology information. Check
Last Found / Last Indexed for record-level freshness.
[0.1] — 2026-07-10
Added
Initial release with 36.8M sites across 12 ecommerce/CMS platforms.
59 enriched fields per site: verified emails, phones, social profiles, firmographics, tech stack, geo, rankings.
Flexible filtering: country (multi-select), keyword, phone code, has-email/has-phone toggles, and universal JSON filters on any column (9 operators, AND logic).
Output column selection — leave empty for full enrichment (all 59 fields).
Sort by key metrics (Overall Score, Tranco, Sales Revenue, Employees…) to get top leads first.
Deduplicate by domain, count-only preview mode, max 100K rows per run.
Platform multi-select with "All platforms" default.
Phone numbers normalized to a clean international format.
_platform field added only when multiple platforms are queried.
Dataset Overview view with Root Domain first.
Input form grouped into 3 logical sections (Data source, Filters, Options).
Automatic retry and recovery for intermittent service connections.
Input validation and safe filtering for all supported columns and operators.
Moved source selection to an isolated service. Limited a fresh input to ten rows and resolved absent optional columns as null. Explicit source selection, continuation, prices and database contents are preserved.
