Record linkage workbenchInvented data: every company, person, domain and file on this page is made up.

Five data files describe the same companies. Two of them copy one upstream file, so when they agree, that is one opinion, not two.

This run finds that pair from the files, merges companies and contacts across all five, picks the right value for each field against a hand-checked sample, measures how old each record is, and keeps the source and date of every value it outputs.

  • Files B and D likely share an upstream file. On the checked sample, 70% of their errors are the same wrong value in both files (next highest pair: 34%), and 72% of the companies they share carry the identical verified date.
  • 531 company records became 159 companies. 700 contact records were matched inside them. 13 unclear matches wait for review instead of merging.
  • Where the files disagree, the resolver picked the right value 78 times out of 94. One vote per file got 67. Scored on 50 hand-checked companies that did not set the weights.
  • File B is the oldest: half its records were last verified over 495 days ago. Intent topics go out of date fastest, about 5.6% a month.
  • All 1191 output fields carry a source file and a verified date.
ABCD
B8%A and B
C16%A and C14%B and C
D11%A and D70%B and D15%C and D
E7%A and E34%B and E17%C and E27%D and E
Of the checked fields where either file is wrong, the share where both are wrong with the same value. Only the pair B and D stands out.
Invented sample
17 of 17 checks pass on this sample

The five files

Each file holds companies (name, domain, city, size, industry, tech in use, intent topic) and contacts (name, title, sometimes an email), each with a last-verified date. Today is fixed at 2026-11-01.

A
File Afirmographic and contact file118 companies, 159 contacts
B
File Bcontact and intent file106 companies, 137 contacts
C
File Cfirmographic file with technographics127 companies, 167 contacts
D
File Dintent and contact file92 companies, 122 contacts
E
File Elow-cost bulk list88 companies, 115 contacts

1. Overlap, and the pair that shares an upstream file

Two files that buy the same upstream file agree with each other, and they are wrong together. Agreement is not evidence on its own, because two accurate files agree too. The run looks for what copying leaves behind: the same wrong values and the same verified dates.

PairCompanies in bothPlain agreementShared mistakesSame verified dateReading
B D81 (69% of either)90.5%70% of 116 errors72%Likely one upstream file: count once
B E55 (40% of either)60.8%34% of 98 errors0%Independent
D E46 (35% of either)59.4%27% of 64 errors0%Independent
C E64 (44% of either)62.3%17% of 83 errors0%Independent
A C93 (62% of either)89.8%16% of 49 errors0%Agree often, mistakes independent
C D71 (49% of either)73.3%15% of 75 errors0%Independent
B C84 (57% of either)63.8%14% of 121 errors0%Independent
A D69 (49% of either)72.7%11% of 72 errors1%Independent
A B81 (57% of either)62.8%8% of 120 errors0%Independent
A E58 (40% of either)60.9%7% of 82 errors0%Independent

Shared mistakes are counted on all 100 hand-checked companies. Rule: a pair is flagged when at least 40% of its checked errors are shared, that share is at least twice the median pair, and at least 30% of its shared companies carry the identical verified date.

2. One company and one contact, found across the files

Matching goes by domain first (after removing https, www, slashes and hyphens), then by name and city. A record with no domain merges only on an exact name in the same city; anything close waits for review. Two different domains never merge, even when the names look alike.

Orrinth Freight Co., as each file wrote it

FileName as writtenDomain as writtenVerifiedHow it matched
AOrrinth Freight Co.orrinth-freight.example2026-10-26first record: starts the company
BORRINTH FREIGHT, LLChttps://www.orrinthfreight.example2026-01-13same domain (orrinthfreight.example)
COrrinth Freightorrinthfreight.example2026-07-02same domain (orrinthfreight.example)
DOrrinth Freight, LLChttps://www.orrinthfreight.example2026-01-13same domain (orrinthfreight.example)
EOrrinth Freight Inc.www.orrinthfreight.example2026-01-24same domain (orrinthfreight.example)

One contact inside it

FileName as writtenEmailTitle as writtenVerified
AKenji DelacroixnoneRevOps Manager2026-10-26
BKenji DelacroixnoneRevOps Manager2026-01-13
CKenji Delacroixon fileRevOps Manager2026-07-02
DKenji DelacroixnoneRevOps Manager2026-01-13
EKenji DealcroixnoneRevOps Manager2026-01-24

Contacts merge on the same email, on the same name after nicknames are read as the full name (Bob as Robert), or on a close name with the same job title (VP Sales read as VP of Sales). Any other close name waits for review.

Waiting for review (5 companies, 8 contacts)

  • C “Quillon Group, LLC”, no domain, close to “Qiullon Group” (name 0.79, same city)
  • C “Sorvik Industreis Co.”, no domain, close to “Sorvik Industries, LLC” (name 0.83, same city)
  • E “Wexomor Works”, no domain, close to “Wexmoro Works, LLC” (name 0.86, same city)
  • E “Larkwlel Holdings”, no domain, close to “Larkwell Holdings Inc.” (name 0.89, same city)
  • E “Quillon Freigthline”, no domain, close to “Quillon Freightline” (name 0.85, same city)
  • C contact “Kenji Pterov”, close to an existing contact (name 0.77)

Look-alikes kept apart (4)

  • A “Tamberlyn Partnering, LLC” (tamberlynpartnering.example) is not “Tamberlyn Partners”: different domain
  • A “Quillon Freightline” (quillonfreightline.example) is not “Quillon Freight Co.”: different domain
  • B “OSTERY LABS” (osterylabs.example) is not “Ostery Lab”: different domain
  • D “Norvane Systems” (norvanesystems.example) is not “Corvane Systems Inc”: different domain

Against the generator’s answer key, which a real run would not have: 100.0% of automatic company merges are correct (702 of 702 merged pairs), and 100.0% of contact merges.

3. Conflicts, settled against a hand-checked sample

100 merged companies form the hand-checked sample: a person checks size, industry, contact titles and intent against the company itself (here the generator plays that person). The first 50set each file’s accuracy. The other 50 only score the result.

Accuracy per file, on the first 50

FileCompany sizeIndustryJob titleIntent topicAll four
A100% of 4095% of 4090% of 5188% of 4093%
B75% of 3288% of 3250% of 3841% of 3263%
C90% of 4295% of 4278% of 5195% of 4289%
D93% of 2889% of 2874% of 3461% of 2879%
E76% of 2576% of 2560% of 3536% of 2562%

Right value on the other 50, where the files disagree

FieldDisagreementsOne vote per fileResolver
Company size2117 18
Industry99 9
Job title3321 26
Intent topic3120 25
All four9467 (71%)78 (83%)

The resolver weighs each file by its measured accuracy on that field, lowers the weight of old records by the decay measured below, and counts the flagged pair as one file. A close vote is marked for review in the output.

4. Freshness: how old each record is, and how fast it decays

FileMedian ageOlder than 12 monthsTitles and intent out of date now (estimated)In 6 months
A57 days0%8% 30%
B495 days76%50% 61%
C99 days0%13% 33%
D389 days53%36% 51%
E463 days61%46% 58%

Titles and intent topics still right, by record age

  • 0 to 3 months89% of 279
  • 3 to 6 months82% of 175
  • 6 to 12 months49% of 85
  • 12 to 24 months44% of 230
  • 24+ months25% of 40

Measured change per month (on the first 50)

  • Intent topic5.6%
  • Job title3.4%
  • Company size1.4%
  • Industry1.2%

Same dates, same staleness: files B and D age together, so a refresh of one does not refresh the other.

5. Provenance: every output field keeps its source and date

FieldValueFromVerifiedWhy this value
Company nameOrrinth Freight Co.A2026-10-26most recent record
Domainorrinthfreight.exampleA2026-10-26most recent record
Company size5000+A with D, C2026-10-26weighted vote
IndustryManufacturingA with C, D2026-10-26weighted vote
Tech in usecloud CRMA with C, E2026-10-26weighted vote
Intent topicdata enrichmentC with D, E2026-07-02weighted vote
Job title (Kenji Delacroix)RevOps ManagerA with C, D, E2026-10-26all 5 sources agree
Job title (Selin Achterberg)Head of MarketingA with C, E2026-10-26all 3 sources agree

When a value turns out wrong, its source and date point to the one file that supplied it.

Checks on this run

Recomputed in your browser on every sample. The same checks run at /api/sources-selftest.

  1. Pass: All five files load with rows 118 / 106 / 127 / 92 / 88
  2. Pass: Exactly one pair is flagged as sharing an upstream file 1 flagged
  3. Pass: The flagged pair is the pair the generator planted B-D
  4. Pass: Plain agreement alone does not single out that pair top raw agreement B-D 90.5%; B-D 90.5%; within 6 points: A-C
  5. Pass: Identical verified dates appear only in the flagged pair B-D 72%, max other 1%
  6. Pass: On the checked sample the pair shares more of its mistakes than any other pair 70% of its errors shared vs max other 34%
  7. Pass: Automatic company merges are at least 99% precise 100.00% of 702 merged pairs
  8. Pass: No look-alike company is merged into another 0 mixed entities
  9. Pass: Unclear company matches wait for review 5 in review
  10. Pass: Contact merges are at least 98% precise 100.00% of 797 pairs
  11. Pass: Accuracy is set on one half of the sample and scored on the other 50 + 50, overlap 0
  12. Pass: The resolver beats one-source-one-vote on the held-out half 78 vs 67 of 94
  13. Pass: Intent topics go out of date faster than titles, titles faster than industry per month: intent 0.056, title 0.034, industry 0.012
  14. Pass: Every output field carries a source and a date 1191 fields, 0 without
  15. Pass: Every cited source really holds that value on that date 0 of 1191 do not
  16. Pass: Every domain is an invented .example domain 481 domains
  17. Pass: The same sample gives the same result twice fca4331d

What this run does not show

  • The data is invented and the shared upstream file is planted. Real vendor files are larger and messier, and their sourcing is often not published.
  • The hand-checked sample is played by the generator. On real data a person checks those companies, and the sample size sets how sure each accuracy figure is.
  • Matching rules are tuned to these files. Real files need their own review of the unclear matches before the rules are trusted.
  • It runs on about 159 companies in a browser, not on millions of records.