Skip to main content

Batch processing

To clean up an existing list (customer base, prospect file, newsletter subscribers), import it into Batch processing in the dashboard. Addresses are verified with the same checks and the same verdict rules as the API.

Available from the dashboard

Batch processing is run from app.yeswecheck.fr. Its HTTP routes only accept a dashboard session and no API key: they are not part of the public API. To verify addresses as they come in from a server, use POST /v2/email/validate.

Prepare the file​

ConstraintValue
FormatsCSV or text (.csv, .txt), one address per line for a text file
CSV separatorDetected automatically (,, ;, tab)
Maximum size15 MB
Rows500,000 at most
Concurrent jobsOne active batch at a time per organisation

Export an Excel workbook to CSV before importing it. Only the address column is checked. To match the results to your database, choose at import the column that holds your identifier (customer number, CRM ID…): its value is returned as is in the client_id column of the export.

Run a batch​

  1. Import: a preview shows the first rows, and the detected address column and identifier column. Correct them if needed, including afterwards from the summary, without re-importing the file.
  2. Choose the options (table below).
  3. Track: progress, counters per verdict and estimated time remaining update live. You can leave the page, pause, resume or cancel.
  4. Export the results and the PDF report.
OptionDefaultEffect
SMTP checkenabledConfirms that mailboxes exist; 1 credit per conclusive address
DeduplicationenabledAn address that appears several times is verified and billed only once; the export keeps the first row and its identifier. Disable it if every identifier must come out
Second passdisabledRe-verifies later the addresses that stayed unknown (greylisting, slow servers); makes the batch longer
Reference countrynoneCountry used for the presumed age when the list is single-country on generic domains
Audit modedisabledQuality report on a sample, without row-by-row results (see below)
Label—Name of the batch in the history

How it works​

PhaseWhat happens
Pre-analysisSyntax, DNS, disposable addresses, typos, trap domains… for every row. Fast.
SMTPDialogue with the mail servers, spread by domain and paced to respect each provider. This is the longest phase.
Second passIf enabled: a new attempt on the unknown addresses, after a delay.

The duration depends mainly on the share of large providers (Gmail, Microsoft, Orange…) that impose a pace. Expect anything from a few minutes for a small list without SMTP to several hours for hundreds of thousands of addresses with SMTP.

Credits​

Same rules as the API: 1 credit per address verified with SMTP whose final verdict is not unknown. unknown addresses are refunded when the batch closes. If the balance runs out along the way, the batch continues without SMTP for the remaining rows instead of failing, and the report says so. See Credits.

Audit mode​

Audit mode answers "what is my list worth?" before you pay for full processing: pre-analysis covers the whole list, SMTP a representative sample, and the report estimates deliverability, with a margin of error, by provider.

Row-by-row results stay locked. You can unlock full processing within 30 days of the audit: the credits are then debited and the results become exportable, for the retention period chosen by your organisation (see Result retention).

Export​

You configure the export before generating it:

SettingChoices
FormatCSV, Excel (.xlsx), text (one address per line), JSON, NDJSON
ContentFull (all columns), simplified (row_id, client_id, email, result, notes, correction_typo) or custom
VerdictsAll rows, or only certain verdicts (for example valid only for a campaign)
SplitOne file, or one file per verdict
Custom columnsChoice, order and renaming of columns (row_id, email and result are always included)
CSV optionsSeparator (comma, semicolon, tab), UTF-8 encoding with or without BOM, true/false or 1/0 booleans, ISO or local date format

Large exports are prepared in the background: a download link, valid for 48 hours, is sent to you by email. The PDF report summarises the quality of the list: verdict breakdown, rejection causes, providers, recommendations.

Full export columns​

ColumnContent
row_idRow identifier specific to the batch: <job>_row_<N> (N = position of the row in your file) without an identifier column, otherwise a non-reversible hash of your identifier
client_idYour identifier, as it appears in the column chosen at import. The key for merging back into your database. Absent if no identifier column was chosen
emailAddress as imported
resultVerdict: valid, risky, invalid, unknown
scoreScore from 0 to 100
notesReadable reason for the verdict, with the typo or the trap domain as additional information
smtp_resultSMTP result: valid, invalid, unknown
smtp_reasonSMTP reason code (MAILBOX_NOT_FOUND, ACCEPT_ALL_DOMAIN…)
syntax_valid · dns_validCorrect syntax · reachable domain
disposableDisposable address
malicious_domainMalicious domain
role_accountGeneric account
profanityOffensive term
random_stringRandom local part
is_aliasSub-addressing
typo_domainLikely typo in the domain
typo_suggestionSuggested domain (empty if the intended domain no longer receives mail)
correction_typoCorrected address verified as deliverable (with SMTP only)
typosquatTrap domain
free_provider · business_emailConsumer provider · business address
whitelisted · blacklistedOn your allowlist · blocklist
first_name · last_name · genderInferred first name, last name, gender
presumed_age · age_rangePresumed age, age range
validated_atVerification date (ISO 8601)
processing_msProcessing time for the row
aliasing_noteWarning about heavy alias use on the same mailbox

Result retention​

Detailed results are kept for 7 days after the batch ends, then deleted. To keep them longer, choose 30 days in Settings → Preferences → Batch processing retention (restricted to the organisation's owner and administrators).

SettingResults downloadableEmail reminders before deletion
7 days (default)7 days after the batch ends2 days and 1 hour before
30 days30 days after the batch ends7 days, 2 days and 1 hour before

The setting applies to the whole organisation and to batches that finish after the change; a batch that has already finished keeps its deletion date, shown in the history. An audit that has not been unlocked is kept for 30 days in both cases; once unlocked, it follows the organisation's setting.

Export what you need within this period. Export downloads can be protected by a download PIN in Settings → Security.