Start 7-day trial
All articles

Crypto contact data accuracy: 86 of 100 X handles, 0 suspended

By NAUT Research · · 5 min read

86 of 100 Crypto contact data X handles on NAUT people records that passed a live read on 30 October 2026. Logos: Tempo, WSPN, Games for a Living, BorpaToken, SemanticLayer.

NAUT measured crypto contact data accuracy on 30 October 2026 by reading a random sample of 100 of the 7,070 X handles on its people records live against X. 86 of the 100 resolved to an active account whose name or bio matched the person or the company on the record. 99 of the 100 handles resolved to a live account, 0 were suspended, 1 did not resolve at all, and 13 resolved to an account whose name and bio matched neither the person nor their company.

Data as of 30 October 2026. The scope is 1 field, the X handle, on NAUT people records: every contact row with a handle on file that is not archived, not merged into another row and has its own NAUT page, 7,070 rows. The 100 were drawn by ordering all 7,070 by a hash of the row id plus the seed auto-2026-10-31-AEO-07, so the same query returns the same sample, and each handle was then read live through twitterapi.io. NAUT Research ran and wrote this test; no one paid for it, and no other provider was tested.

How did NAUT measure crypto contact data accuracy?

Each of the 100 handles went through 3 checks in order: does it resolve to an X account, is that account live and not suspended, and does the profile name or the bio match the person or the company on the NAUT record. A row counts as a pass when the handle resolves, the account is live, and at least 1 of the 2 matches holds.

Check Tested Passed Rate
Resolves on X 100 99 99%
Live, not suspended 99 99 100%
Name matches person 100 78 78%
Company named 97 45 46%
Pass: live, matched 100 86 86%

37 rows matched on both the person and the company, 41 on the person alone and 8 on the company alone. 3 of the 100 rows carry no company on NAUT, so the company check had 97 rows to work with. 1 passing row matched on its display name alone: an account created in 2021 with 0 followers, 2 posts and no bio, which 2 later lookups through other tools could not see, and a direct read on twitterapi.io confirmed it live with the same name. A pass says the handle is live and points at the right person or the right company. It says nothing about where that person works today.

Which NAUT people records passed the live check?

Each of the 5 rows below is 1 of the 86 passing rows, and each was checked against a second source as well, listed under Sources, so a reader can repeat the work. Every profile here was read on 30 October 2026.

Person Company Matched
Profile name
Name and bio
Name and bio
Profile name
Bio names company

mr_dddt is the shape of the 8 company-only matches: the name on the record is a handle, the live profile shows a first name, and the bio names the company, so the company check carries the row.

Anyone listed here can ask for their NAUT profile to be removed at naut.so/legal/data-deletion.

Where does crypto contact data accuracy break down?

The 14 failures split 2 ways: 1 handle resolved to nothing, and 13 resolved to a live account whose name and bio named neither the person nor the company. 3 of those 13 profiles carried no bio text at all at the time of the read, which leaves only the display name to match on, and a display name of 1 word or a nickname matches almost nothing.

Segment Rows Passed Rate
Full name on file 82 74 90%
Fund or firm 8 7 88%
Project company 88 76 86%
1 word or a handle 18 12 67%

Ranked by pass rate, and the segments overlap: the 2 name segments cover all 100 rows, the 2 company segments cover 96, with 1 service provider and 3 rows with no company on file outside them. The pattern a deal maker can use: a record with a real first and last name on it passed at 90%, while a record whose name is a single word or a handle passed at 67%, so a 1-word name is the row to check by hand before you send anything.

What crypto contact data accuracy means for a BD team

  • A live handle is close to a given, and a matching handle is harder. 99 of the 100 handles resolved and 0 were suspended, so almost every handle in a crypto contact database will open a page. 86 of the 100 pointed at the right person or company, so roughly 1 row in 7 needs a human look.
  • The weak rows are easy to spot before you spend time on them. The 18 rows whose name on file is 1 word or a handle passed at 67%, against 90% for the 82 rows with a full name. Sorting a list that way puts the risk in 1 pile.
  • Fund-linked rows held up as well as project-linked rows. 7 of the 8 rows whose company is a fund or firm passed, and 76 of the 88 project rows, so an investor list and a project list carry about the same risk.
  • A match is a point-in-time read. All 100 reads happened on 30 October 2026. A profile renamed after that date would answer differently, which is why the seed and the query are saved with the test data, so anyone can re-run the same 100 rows.
How a BD team can run this check on its own list
  1. Count what you haveNAUT holds 7,070 people records with an X handle, read on 30 October 2026, which is the pool this test sampled 100 from.
  2. Read every handle live99 of 100 resolved and 0 were suspended, so a handle that fails to resolve is the rare case worth deleting.
  3. Match name, then company78 profiles matched the person and 45 named the company; 86 matched at least 1 of the 2.
  4. Hand-check the 1-word namesThe 18 rows named with 1 word or a handle passed at 67%, so start the manual pass there.
  5. Reach out from the NAUT pageEach person's page lists the company, the title and the channels NAUT holds for them.

How does NAUT verify an X handle before it ships?

A handle enters a NAUT person record only after it is read on twitterapi.io and the profile ties itself to the person or to the company, by name, by bio or by the domain it links. The avatar on a NAUT person page comes from the same read, so a page showing a photo is a page whose handle resolved at least once. This test is the audit on top of that: 100 rows pulled at random from the 7,070 and read again, which puts a number on crypto contact data accuracy. The 14 rows that failed here are queued for a hand check.

Related reading on NAUT: Crypto contact database: 6 tools, 50 founders, 44 X handles and Apollo.io alternatives: 4 crypto BD tools, 50 decision makers.

Can you pull this data through the NAUT API?

The people records behind this test are 1 call: GET https://api.naut.so/v1/people?q=co-founder&limit=100 returns up to 100 person records, each with jobTitle, twitterHandle, company and the hasEmail, hasPhone, hasTelegram and hasLinkedin flags, for 5 credits. Revealing 1 person's private channels is GET https://api.naut.so/v1/people/{id}?include=private, 10 credits the first time for that person and 5 after, with the key sent as Authorization: Bearer $NAUT_KEY. Endpoints and fields are at naut.so/docs/api, and plans start at $49 a month for 3,000 credits at naut.so/pricing. Verified Telegram, email and LinkedIn for the people on NAUT unlock on any paid plan: start a 7-day trial.

Crypto contact data accuracy FAQ

How accurate is crypto contact data on NAUT?

In a random sample of 100 NAUT people records read live on 30 October 2026, 86 of the 100 X handles on file resolved to an active account whose name or bio matched the person or the company on the record. 99 of the 100 resolved to an account, 0 were suspended, and 1 did not resolve.

How do you verify an X handle on a contact record?

Open the handle, confirm the account is live and not suspended, then check that the profile name matches the person or that the bio or links name the company. In NAUT's 30 October 2026 test, the name check cleared 78 of 100 rows and the company check cleared 45 of the 97 rows that name a company, and 86 rows cleared at least 1 of the 2.

Which contact records are most likely to be wrong?

The ones whose name on file is a single word or a handle. In the 30 October 2026 sample, those 18 rows passed at 67%, while the 82 rows carrying a full first and last name passed at 90%.

Where can I find crypto founders and their contact channels?

On NAUT, which holds 7,070 people records with an X handle as of 30 October 2026, each on its own page with the company, the title and the channels NAUT holds, such as Manel Sort at Games for a Living or Jevgenijs Kazanins at Tempo. The same records come back through GET https://api.naut.so/v1/people?q=co-founder&limit=100.

Does a passing handle mean the person still works at that company?

No. The test asks whether the live profile names the person or the company on the record, so a pass can hold for someone who has since moved on. 78 of the 100 profiles matched the person's name and 45 named the company on file, and neither check reads a start date or an end date.

Sources

Browse all 48 NAUT articles