Three streams in, seven checks out.

Every record in the catalogue arrives through one of three sourcing streams, passes a compilation pipeline that removes far more than it keeps, and only then goes through seven verification checks run by third-party engines. Every record carries a source tag, so its origin can be traced after delivery.

verification runs monthly on the 9th, 11th or 13th

Sourcing streams

Verification checks

Records held

0 M

Verified, standard

0 %

Five checks remove a record. Two label it.

Verification runs through ZeroBounce, NeverBounce and equivalent providers rather than in-house scripts — they maintain continuously updated trap and blocklist intelligence, and the mailbox probing happens on their infrastructure rather than ours.

01 removed

02 removed

03 removed

04 removed

05 removed

06 flagged

07 flagged

Where records come from, and what happens before a single check runs.

Verification costs money per address, so the pool is cleaned first — there is no point probing a mailbox on a record that should never have been in the file.

Permission-based collection

Newsletter and subscription forms, gated content downloads, webinar and event registrations, preference-centre signups. Double opt-in where the collecting property supports it.

Public web and business directories

Company websites, published directories, corporate and trade registries, and professional listings where contact details are published for the purpose of being contacted.

Licensed data partners

Records acquired under written agreement from vetted providers. Each agreement requires the partner to warrant it holds the right to license the data onward.

Compilation and standardisation

Normalisation of casing and field mapping; exact dedupe on email then fuzzy matching on name plus domain; malformed and placeholder records dropped; completeness scoring; indexing by country, industry, company size and role seniority.

Records leave verification in one of four states.

Standard orders contain Valid records only. Accept-all and Role records are available on request, in clearly labelled segments, so you decide whether to mail them and at what volume. Invalid records are removed — not resold, not held back for a cheaper tier.

every record carries a source tag and a verification date

valid

All seven checks passed. What a standard order contains

accept-all

Domain accepts anything; mailbox unconfirmable. Labelled, on request

role

Generic mailbox. Segmented separately, on request

invalid

Removed. Not sold, not counted

Questions about how the data is built

Why outsource verification instead of building it?

Trap and blocklist intelligence has to be continuously updated to be worth anything, and mailbox probing at volume can damage the reputation of whoever runs it. Outsourcing both is the honest engineering choice, not a shortcut.

How much gets dropped in compilation?

A large proportion — duplicates across directories, malformed rows, and placeholder records. We do not publish a single percentage because it varies enormously by source, and a made-up figure would be worse than none.

Can you tell me a specific record’s source?

Yes. Every record carries a source tag; query a row and we will tell you which stream it came from.

Do you claim 100% accuracy?

No, and you should distrust anyone who does. Standard lists are up to 85% verified, with a 99% build available on request.

Five pages, no gates

Everything about how the data is built, measured, refreshed and removed. If a question here is not answered, ask on WhatsApp before you order rather than after.

Accuracy reporting

85% / 99%

What we measure, what we will not claim, and the four grades a record can carry.

Update frequency

9th · 11th · 13th

One verification pass a month, on one of three dates, plus a re-run before dispatch.

Compliance center

gdpr / can-spam

Who is responsible for what, and the obligations that stay with the sender.

Opt-out & removal

permanent

How to be removed, how suppression works, and what permanent actually means.

The pipeline is only worth as much as the rows it produces. Test 100.

Sales and service are reachable 24/7 on WhatsApp at +1 (315) 215-1563, or at [email protected].

Verified B2B and consumer contact data at a flat, one-time price. Every list dated, every record 11 fields deep.

Data products

Customer Care

Company

EMAIL	COMPANY	TITLE	CHECKED
j.whitcombe@•••••.com	Northloop	Founder	Updated
priya.n@•••••••.io	Cadence AI	CEO	Updated
+ 98 More Rows in Your Sample

EMAIL	COMPANY	TITLE	CHECKED
j.whitcombe@•••••.com	Northloop	Founder	Updated
priya.n@•••••••.io	Cadence AI	CEO	Updated
+ 98 More Rows in Your Sample

The B2B contact database, sold as files you own.

Every list, by country, industry, role and stack.

Buy data by what you are trying to do.

Account-level data, tied to the same contact rows.

Our workings, in public.

The B2B contact database, sold as files you own.

Every list, by country, industry, role and stack.

Buy data by what you are trying to do.

Account-level data, tied to the same contact rows.

Our workings, in public.

Three streams in, seven checks out.

Five checks remove a record. Two label it.

01

Syntax validation

removed

02

Domain and MX check

removed

03

Disposable domain screening

removed

04

Spam trap and honeypot screening

removed

05

SMTP mailbox verification

removed

06

Role account detection

flagged

07

Catch-all classification

flagged

Where records come from, and what happens before a single check runs.

Questions about how the data is built

Five pages, no gates

The pipeline is only worth as much as the rows it produces. Test 100.

The B2B contact database, sold as files you own.

Every list, by country, industry, role and stack.

Buy data by what you are trying to do.

Account-level data, tied to the same contact rows.

Our workings, in public.

The B2B contact database, sold as files you own.

Every list, by country, industry, role and stack.

Buy data by what you are trying to do.

Account-level data, tied to the same contact rows.

Our workings, in public.

Three streams in, seven checks out.

Five checks remove a record. Two label it.

01

Syntax validation

removed

02

Domain and MX check

removed

03

Disposable domain screening

removed

04

Spam trap and honeypot screening

removed

05

SMTP mailbox verification

removed

06

Role account detection

flagged

07

Catch-all classification

flagged

Where records come from, and what happens before a single check runs.

Questions about how the data is built

Five pages, no gates

The pipeline is only worth as much as the rows it produces. Test 100.

FREE 100-ROW SAMPLE

See 100 real rows before you spend anything.

EmailProLeads

100 verified rows, free, before you pay a thing.