Export confirmed subscribers to Kit, Mailchimp or Sendy
By Flavio
Export the confirmed addresses from a Waiting Lists list as a CSV ready for Kit, Mailchimp or Sendy, and keep the consent archive that proves each signup.
Your waiting list did its job. People signed up, confirmed their address, and now launch day is close. You want to send them an email, and that means moving the list into the newsletter tool you already use: Kit, Mailchimp or Sendy.
The usual way is a CSV export, then a fight with the import screen. Wrong column names, an extra column the tool doesn’t understand, or a file full of people who never confirmed.
Waiting Lists has an export preset for each of those three tools, plus a plain email-only file and a full consent archive. Only confirmed addresses go in. By the end of this, you’ll have the right file for your tool, a consent record you keep for yourself, and a routine for exporting again as more people join. If you haven’t set up the service yet, start with building a waiting list with double opt-in on Cloudflare.
What you need
Waiting Lists runs on Node.js 22.12+ and Cloudflare Workers, D1 and Email Service. It’s source code you deploy in your own Cloudflare account, not a hosted service. After purchase, download the source zip from your Polar receipt and unzip it. The receipt always has the latest version.
You also need:
- A running installation, either deployed to your Cloudflare account or started locally with
npm run dev. The export reads from whichever one the CLI points at. - At least one list with confirmed subscribers. Pending addresses never appear in an export.
- An API key, created by the administrator at
/admin/agent. It’s shown once, and the app stores only its hash. - The CLI configured with that key. Pipe it in through standard input, so it doesn’t end up in your shell history:
pbpaste | npm run waitinglists -- configure --key-stdin --api-url https://lists.gardenplanner.app/api/v1
The key is saved in ~/.config/waitinglists/credentials.json with owner-only permissions. Leave out --api-url when you’re exporting from a local copy, since it defaults to http://localhost:4321/api/v1.
The preflight check: run lists from the unzipped source folder. If it prints your lists with a confirmed count, the key and URL are right.
npm run waitinglists -- lists
3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 Garden Planner 3 confirmed https://lists.gardenplanner.app/api/subscribe/garden-planner
The list, host and addresses in this tutorial are examples. Each line has the list ID, its name, the confirmed count and the public endpoint, separated by tabs.
Pick the right export format
Before exporting, decide what each file is for. The export command takes a --format with five values, and they don’t contain the same data. Picking the wrong one means either an import that fails, or a file that carries more personal data than the tool needs. Your newsletter tool only needs the address, and in Mailchimp’s case the two timestamps. You, on the other hand, need the full record of when and how each person agreed. For each subscriber, Waiting Lists stores the email, its status, the consent version, the request and confirmation times, an approximate country and the delivery status of the confirmation email. No names, no IP addresses.
--format | Dashboard label | Header row | Use it for |
|---|---|---|---|
archive (default) | Full archive | email,country,confirmed_at,consent_at,consent_version | Your own consent record |
universal | Email only (universal) | email | Any tool that takes one column |
sendy | Sendy | Name,Email | Sendy imports |
kit | Kit | Email address | Kit imports |
mailchimp | Mailchimp | Email Address,OPTIN_TIME,CONFIRM_TIME | Mailchimp imports |
Here’s how I decide:
Two details are worth knowing. Waiting Lists never collects names, so the Sendy file has a Name column that’s always empty. It’s there only to give Sendy the two columns it expects. The Mailchimp file is the only preset that carries timestamps: OPTIN_TIME is when the person submitted the form, and CONFIRM_TIME is when they pressed the confirmation button.
Export the consent archive first
Start with the archive, even if your tool only wants addresses. It’s the one file that shows when each person asked to join, which consent wording they saw and when they confirmed. If someone ever asks why they’re getting your emails, this is your answer. It also outlives the database: the included cleanup job deletes confirmed records after 24 months, so once a row is gone, the archive is the only copy. Without it, your newsletter tool holds addresses with no proof of how they got there.
Example 1: the main path, archive then Mailchimp. Run both exports from the source folder, passing the list ID from lists:
npm run waitinglists -- export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --output garden-planner-subscribers.csv
> waiting-lists@1.0.0 waitinglists
> node ./cli/waitinglists.mjs export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --output garden-planner-subscribers.csv
Export saved to /Users/flavio/dev/waiting-lists/garden-planner-subscribers.csv
The first two lines come from npm. The last one is the CLI, with the full path of the new file. No --format means archive. This is the shape of the file, with an example row:
email,country,confirmed_at,consent_at,consent_version
"marta@greenrow.net","Denmark","2026-10-02 18:00:00","2026-10-02 17:55:00","2026-09-garden-v1"
Every value is quoted, and lines end with \r\n. The country is the full English name, stored at confirmation without keeping the IP address. The archive also starts with a UTF-8 byte-order mark, three invisible bytes before email. A script that splits the first line by hand will see a header called email, so strip it first.
The consent_version column is the value of CONSENT_VERSION in wrangler.jsonc when the person signed up. If you never changed it, every row says replace-with-your-consent-version, and that proves nothing. Set it to a name for your form’s wording before you collect real addresses, and change it whenever the wording changes.
The timestamps come from the database’s CURRENT_TIMESTAMP, which SQLite defines as UTC in the YYYY-MM-DD HH:MM:SS format. There’s no time zone in the text, so remember they’re UTC when you read them.
Export the import file for your tool
Now the file you’ll upload. Each preset matches its tool’s column names, so the import doesn’t stop to ask what a column is or fail on extra data. The preset files also carry less personal data than the archive: no country, no consent version. That’s the point. Your newsletter tool doesn’t need them, and the less you copy around, the less you have to protect. Run one export per tool, each to its own file name, because the CLI won’t reuse a name that’s already taken.
Continuing Example 1, here’s the Mailchimp file:
npm run waitinglists -- export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --format mailchimp --output garden-planner-mailchimp.csv
Email Address,OPTIN_TIME,CONFIRM_TIME
"marta@greenrow.net","2026-10-02 17:55:00","2026-10-02 18:00:00"
The CLI prints Export saved to and the path, as before. The block above is the file’s content, with the same example subscriber. Upload it in Mailchimp’s contact import and match each column to the field with the same name.
Example 2: Kit and Sendy, same subscribers, different files. Change only the format and the file name:
npm run waitinglists -- export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --format kit --output garden-planner-kit.csv
npm run waitinglists -- export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --format sendy --output garden-planner-sendy.csv
The Kit file has one column:
Email address
"marta@greenrow.net"
The Sendy file has two, with an empty name:
Name,Email
"","marta@greenrow.net"
Neither file has timestamps. If your tool can store the signup date and you want it there, keep the archive next to the import and add it by hand. If you use a tool that isn’t on the list, --format universal gives you a single email column without the byte-order mark.
How the tool treats these addresses is up to the tool. Check its import options, since some can send their own confirmation email to imported contacts. Your subscribers already confirmed once, through Waiting Lists.
Run the export from the right folder
The CLI isn’t a global command. npm run waitinglists runs cli/waitinglists.mjs from the source folder, and npm always runs scripts from the folder that has package.json. That matters twice. A relative --output path lands in the source folder, even if you typed the command from a subfolder. And without a list ID, the CLI looks for .waitinglists.json in that same folder. That file is written by waitinglists init and holds the non-secret list ID, slug and endpoint. If you skip this, you’ll export to a place you don’t expect or get a “not linked” error.
To export into another project, call the script directly from there. Here we’re in a website folder that init linked to its list, so the ID can go:
cd /Users/flavio/dev/garden-planner-site
node /Users/flavio/dev/waiting-lists/cli/waitinglists.mjs export --format kit --output garden-planner-kit.csv
In a folder without .waitinglists.json, and without an ID, you get:
waitinglists: No list ID supplied and this directory is not linked. Run `waitinglists init`.
Pass the ID, or cd into the linked folder. Don’t run init to fix it unless you want a brand-new list, since init creates one.
There’s also no way to print the export to the terminal. Leave out --output and the CLI refuses:
npm run waitinglists -- export --format kit
waitinglists: Provide --output FILE. Subscriber data is never printed to stdout.
That’s deliberate, and it’s why you can hand exports to a coding agent. The agent runs the command and sees only the path, never the addresses, so they don’t end up in its chat transcript.
Download the same files from the dashboard
Sometimes you don’t have the CLI set up on the machine you’re on, or you’d rather click. The dashboard has the same five exports, built by the same code, so the files are identical. The difference is in where they go. The CLI writes a file with owner-only permissions and refuses to overwrite one. A browser download follows your browser’s rules: it lands in your downloads folder with the usual permissions, and the browser decides what happens to an existing file with the same name.
Sign in at /login, open the list, and use the export form at the top right:
Garden Planner
3 confirmed · 0 pending · Created Sep 28, 2026
Export for: [ Full archive ▾ ] [ Download CSV ↓ ]
Pick a format from Export for: Full archive, Email only (universal), Sendy, Kit or Mailchimp. Then click Download CSV ↓. The file is named after the list’s endpoint slug: garden-planner-subscribers.csv for the archive, and garden-planner-kit.csv, garden-planner-mailchimp.csv and so on for the others. The response is sent with Cache-Control: no-store, so the browser doesn’t keep a cached copy.
Keep the files private
An export is a list of people’s email addresses, and the archive adds when they agreed and roughly where they were. Treat every file like the database itself. Waiting Lists helps on its side: the CLI creates each file with owner-only permissions. But it can’t protect a file once you’ve copied it to a shared folder, attached it to a message or committed it. The easy one to miss is Git. The source folder’s .gitignore doesn’t exclude CSV files, and relative exports land right in that folder.
You can check the permissions:
ls -l garden-planner-kit.csv
-rw-------@ 1 flavio staff 37 Nov 13 07:16 garden-planner-kit.csv
Only your user can read or write it. To keep exports out of commits, add a line to .gitignore in the source folder:
printf '%s\n' '*.csv' >> .gitignore
Once the import worked, delete the preset file. You can always export it again. Keep the archive somewhere private, and make sure your retention process covers it. The docs ask you to preserve consent evidence and to use the addresses only for the purpose people accepted. If your form said “launch updates for Garden Planner,” a newsletter about something else isn’t covered.
One destructive action to know about: the Delete button on a subscriber row in the dashboard removes that record at once. The code has no confirmation prompt, and the page then shows “Subscriber data permanently deleted.” Export the archive before you clean up rows by hand.
Export again as more people confirm
An export is a snapshot. It doesn’t sync. People keep confirming after your first import, and some will use the removal link in their confirmation email. Neither change reaches your newsletter tool on its own. So plan a second export before launch, and treat removals as seriously as additions. If someone removed themselves from Waiting Lists, they shouldn’t get your launch email from Kit either.
Example 3: a second run that fails. Rerunning the exact command from earlier gives an error, because the CLI opens files in write-exclusive mode:
npm run waitinglists -- export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --format kit --output garden-planner-kit.csv
waitinglists: EEXIST: file already exists, open '/Users/flavio/dev/waiting-lists/garden-planner-kit.csv'
It never overwrites, so an older export can’t be replaced by mistake. The fix is a new name. I put the date in it:
npm run waitinglists -- export 3f6c2a10-8d4e-4b7a-9c21-5e0f7b9d1a42 --format kit --output garden-planner-kit-2026-11-13.csv
The rows come out newest first, ordered by when the person signed up. To find who’s new or who left, compare the two files with any diff tool, and import or remove only those addresses in your newsletter tool.
The export itself changes nothing in Waiting Lists. It only reads. To undo an export, delete the file. Undoing an import happens in your newsletter tool, with its own tools for removing contacts.
If something looks wrong
- If the CLI prints
waitinglists: Use archive, universal, sendy, kit, or mailchimp., the format name is wrong. A common one is--format convertkit. Usekit. - If it prints
waitinglists: EEXIST: file already exists, a file with that name is already there. Pick a new--outputname, or delete the old file if you no longer need it. - If it prints
waitinglists: No API key configured. Run `waitinglists configure --key-stdin` or set WAITINGLISTS_API_KEY., the credentials file is missing and the variable isn’t set. Configure the key again from/admin/agent. - If it prints
waitinglists: A valid API key with the required scope is required., the key is wrong or was revoked in/admin/agent. Create a new one and configure it. - If it prints
waitinglists: Waiting list not found., the list ID doesn’t exist on the server the CLI is talking to. Runlistsand check you’re not pointing at your local copy with a production ID. - If the export has fewer rows than you expected, the missing people haven’t confirmed yet. The list page shows the pending count and a Resend confirmation link for each pending address.
- If
consent_versionsaysreplace-with-your-consent-version, setCONSENT_VERSIONinwrangler.jsoncand deploy. Rows that already exist keep the value they were saved with. - If a script reads the archive’s first header as
email, that’s the byte-order mark. Strip it, or use--format universalwhen you only need addresses.
Get Waiting Lists
Collect confirmed subscribers with your own forms, dashboard and agent CLI. Includes full MIT source for deployment in your Cloudflare account.
See Waiting Lists → · Read the docs · $20. See the app page for tax details, with full source code.
Want me to talk about your product? You can sponsor this site.