Bulk send CSV column mapping

Bulk send is a list job, not a mail merge. You choose a workspace template, then give Inkpacket names and emails — a recipient CSV upload or a paste. The product has to know which column is the address and which is the display name. Guessed columns, the unique-email rule, and the preview count are the whole flow. Get them wrong and you either skip people you meant to send or spend envelopes on rows you did not intend.

When a list is the right tool is bulk send for HR and sales envelopes. The other fork — you have the emails, or strangers start themselves — is bulk send vs Web Forms. Desk help is Docs. This guide is the CSV itself: headers, guessed columns, skipped rows, retries, and the cap.

What the file needs

Each row needs an email Inkpacket can recognize. A name is optional. If the name cell is empty, the desk derives a display name from the local part of the address. Quoted commas, tabs, and semicolons are accepted as column breaks. Angle-bracket addresses such as Jane Lee <jane@example.com> still parse. Empty lines are ignored.

A header row is optional. If the first line looks named — email, e-mail, signer email, name, full name, signer, recipient — and does not itself contain an address, Inkpacket treats it as headers and guesses the mapping. If the first line already has an email, there is no header. You can still remap on the desk before you start.

Guessed columns

• Email is the column whose header includes email, or the first cell that looks like an address.
• Name is the column whose header is name, full name, signer, or recipient — and is not the email column.
• If no name column is found, the other non-empty cell on that row becomes the name, or the local part of the email.

Confirm the guess. A sheet exported with Email Address in column C and Legal Name in column A will still map if the headers are honest. A sheet with no headers and the address in the second column will not guess correctly — point the email column at that index before you preview.

Unique emails and skipped rows

Each unique email becomes one envelope. Duplicate addresses in the same job are sent once; later copies are skipped. Rows without a usable email are skipped. The preview count is the number of envelopes you are about to create, not the number of lines in the file. Read that number. A 40-line paste with two duplicates and a blank is 37 envelopes, not 40.

A job accepts 100 unique emails at a time. Split a larger cohort into another job from the same template. That is two jobs, not one mega-list. Retries skip rows that already sent, so a second start on the same job does not mint a twin envelope for a person who already has a link.

Not a shared envelope

Each row is its own envelope with its own signing link, audit trail, and status. This is not sequential or parallel routing on one envelope. If two people must sign the same PDF together, send one envelope with two recipients — when parallel beats sequential routing and sequential vs parallel signing. A decline on one bulk row does not stop the rest of the job.

The click-path

1. Save a workspace template first. Bulk send will not invent a layout from a one-off draft.
2. Open Bulk send at /desk/bulk. On a phone it is in the desk menu.
3. Choose the template.
4. Paste the list or upload the CSV.
5. Map the email column and, if you have one, the name column.
6. Preview the unique count. That is the envelope spend.
7. Set reminder interval, expiration, and an optional access code — or keep workspace defaults.
8. Start the job.

Shared cadence and one optional access code apply to every row in that job. The PIN is not in the invite email. Share it on a second channel if you set one. When to require an access code is that lock. Templates versus one-off envelopes is why the template comes first.

After the job starts

If start fails after create, the desk says Job created, but sending did not finish. Open that job and retry. A sent row means the envelope exists. If invite email was not sent, open that envelope and copy the signing link — the same honest skip as a one-off send. Text-message (SMS) is coming soon on the workspace; the bulk job does not invent a phone column. The envelope and the URL still exist.

Void, correct, or resend a single row the same way you would a one-off. Do not clone a second job because one person lost the mail. When to void, correct, or resend stays on that envelope id. Completed envelopes cannot be voided. File each stamped PDF on its own envelope — completed envelope, stamped PDF, and audit.

Caps and what is not open yet

Each unique email that actually sends is an envelope. Free allows 5 packets a month. Free at 5 packets a month is the live cap desk send, bulk rows, Web Form fills, and API send share. A row that would exceed it returns HTTP 402 ENVELOPE_LIMIT. Starter is 20 envelopes — more than Free — plus 3 published Web Forms. Seat counts are planning numbers; invite does not refuse when a workspace is full.

Billing is published, not charging. Checkout is not open. Text-message (SMS) is coming soon. Drive missing: Drive not configured. Stripe off: Payments not live. Mapping columns does not wait on those keys. The FAQ and bulk send on Features use the same job sentences.

Map the columns. Read the preview. Start once. Stay on the envelopes the job created. Open a workspace on Free — see published pricing and the Free plan 5-packet cap if you want the live 5 next to a 40-row file — and run one short CSV so the guess and the remap are familiar before a real cohort.