Newsletters

A newsletter is a written brief generated from fresh dataset rows on a schedule and emailed to recipients. Ask Jason for one, review issues, and send.

A newsletter turns changing data into something people read: a morning brief on price moves, a weekly digest of new listings, a monthly summary for a client. Each issue is written from the datasets at the time it is generated, previewed on the newsletter page, and emailed to recipients by hand or automatically.

Creating a newsletter

There is no New Newsletter button. Describe the newsletter to Jason in chat and it is created, linked to the right datasets, and given its first issue. It then appears on the Analytics page with a Newsletter badge.

Include four things in the brief: who it is for, which datasets it should read, what it should lead with, and how often it should go out.

Create a Monday-morning newsletter for the merchandising team from the Live product offers and Daily market rollups datasets. Lead with the biggest price drops since last week, then any products that went out of stock, then new products discovered. Keep it under 400 words, end with a three-item watchlist, and send it at 07:00 London time.

Jason creates the newsletter, generates a first issue so you can see the format, and asks who the recipients should be. Add them on the newsletter page; the schedule is already set.

New workspaces start with an example newsletter marked Synthetic. It shows what an issue looks like but cannot generate further issues or be scheduled; ask Jason for a real one.

The newsletter page

A newsletter page with the envelope card showing To, Schedule and Subject above the rendered issue preview
The newsletter page. The envelope is editable; the preview below it is the issue as it will be emailed.
  • Title. Click to rename.
  • Sources. A subtitle listing the linked datasets; click it to open Settings.
  • Version chip. v3 with a clock icon. Every issue is a version; open the chip to see them all with their status (draft, sent, failed) and when they were made. Choosing an older issue shows it with an amber “Viewing v2” banner and a Back to current link.
  • Settings. Opens the settings dialog below.
  • Send. Emails the issue you are viewing to the recipients, after a confirmation. Disabled with a reason when there are no recipients (“Add recipients in Settings first”), when the issue was already sent, or when there is no issue yet. Its menu has Send test to me.
  • Generate now. Writes a new issue from the current data. Shown only on newsletters set up for recurring generation.

The envelope

Field Values Default What it does
To Email addresses as chips Empty Who receives Send and scheduled sends. Type an address and press Enter or comma; paste several separated by spaces, commas or semicolons. Backspace on an empty box removes the last chip. Saves immediately. Invalid addresses are rejected with “Not a valid email”.
Schedule No schedule, Hourly, Daily, Weekly; a time for daily and weekly; day toggles M–S for weekly No schedule (time 09:00, Monday) When new issues are generated. Saves immediately. See Schedule and automatic sending.
Subject Text The issue’s generated title The email subject line for the issue you are viewing. Enter or clicking away saves and re-renders the issue. Disabled on sent issues (“Sent issues can’t be edited”).

A green Sent to N pill appears on an issue after it has gone out.

Preview

The issue rendered exactly as it will be emailed, in an isolated frame beneath the envelope. Nothing in the preview is clickable except links inside the issue itself. Empty state: “No issue generated yet” with a Generate Issue button on newsletters that can generate.

Settings

Press Settings.

The Newsletter Settings dialog with data source checkboxes, a recipients box, schedule pills and the Send automatically option
Newsletter Settings. Sources, recipients and the schedule in one place.
Option Values Default What it does
Data sources Checkboxes for every ready dataset Set by Jason Which datasets each issue reads.
Recipients One email per line (commas and semicolons also work) Empty The same list as To on the envelope, in a larger box.
Schedule Off, Hourly, Daily, Weekly Off When new issues are generated as drafts. Daily and weekly add a time (default 09:00); weekly adds MonSun toggles (default Mon). Times are in the newsletter’s time zone, UTC unless Jason set one.
Send automatically On / off Off Email each scheduled issue to the recipients the moment it is generated, without review.

Save Settings applies everything at once.

With Send automatically on, an amber warning reads “Generated issues will go to N external recipients unreviewed. Make sure you’ve sent yourself a test first.” Turn it on only after you have read a few generated issues and are happy with the tone and the numbers.

The Schedule and Send automatically controls appear only on newsletters set up for recurring generation. An example newsletter shows the note “This is a fixed synthetic example. Recurring generation and scheduling are unavailable.” instead.

Generating an issue

Press Generate now (or Generate Issue in the empty state). The toast reads “Generating a new issue…”, the page checks every few seconds, and the new draft opens when it is ready, usually within a minute. Issues read the active version of each source dataset at generation time, so the numbers match what the dataset page shows at that moment.

A generation that fails is recorded as a failed issue in the version chip, so a gap in a schedule is visible rather than silent. If generation keeps failing, open the chat: Jason can read the failure and adjust the newsletter.

Schedule and automatic sending

  • Off. Issues are only generated when you press Generate now.
  • Hourly, Daily, Weekly. New issues are generated on the schedule as drafts. If Send automatically is off, the newsletter’s owner receives an email “Draft ready” and the draft waits on the page for review and a manual Send.
  • Send automatically on. Each scheduled issue is emailed to the recipients as soon as it is generated.

A manual Generate now never emails anyone, whatever the settings. Scheduled generation is skipped while a previous generation is still running, so a slow issue does not pile up duplicates.

Sending

  1. Send yourself a test. Open the Send menu and choose Send test to me. One copy goes to your own address with the subject prefixed “[Test]”. The issue’s status does not change.
  2. Check the subject. Edit it on the envelope if needed.
  3. Send. Press Send and confirm “Send this issue? Issue v3 will be emailed to 12 recipients. This can’t be undone.” The toast reads “Sent to 12 recipients” and the issue is marked sent.

Each recipient gets their own copy with a personal unsubscribe link in the footer. The email subject is the newsletter name followed by the issue title. An issue can be sent once; to send a corrected version, generate a new issue. If some addresses fail, the issue still counts as sent to the ones that received it.

Issues

Every generated issue is kept as a numbered version: draft until sent, sent with the time and recipient count, or failed. Use the version chip to open any past issue, read it, or send yourself a test copy of it. Editing the subject of an older draft re-renders that issue. Issues cannot be deleted individually; deleting the newsletter deletes them all.

Unsubscribe

Every email carries an Unsubscribe link that works without signing in. Clicking it removes that address from the recipient list and shows a confirmation page. The link is specific to the newsletter and address; it does not expire. You will notice the address gone from To. Test emails carry a working link too, so do not click it in your own test unless you mean it.

Sharing

A newsletter can be added to the workspace share; the public tab shows the latest issue. See Sharing.

Changing a newsletter

Ask Jason. “Add a section on promotions ending this week”, “make it shorter”, “switch it to Fridays” all work from chat, and the next generated issue reflects the change. Datasets, recipients and the schedule can also be changed in Settings.

What’s next

Connect your data assistant

Build datasets and work with your data in ChatGPT, Claude, Copilot or another assistant.

Connect in ChatGPT

  1. Open Settings → Security and login and enable Developer mode.
  2. Open Plugins and select + to create a connection. Name it Jsonify, add a short description, and paste the URL below.
  3. Use OAuth for authentication, select Create, and sign in to your Jsonify account when prompted.
  4. Start a new chat and select Jsonify from + → More, then describe your dataset.
Server URLhttps://factory.jsonify.com/mcp

If Developer mode is unavailable, your plan or workspace settings may restrict custom connections.

Official ChatGPT setup guide ↗

Then say: “build me a dataset of competitor product prices and availability, refreshed daily”. Full instructions per client on /connect.