Data rarely arrives in the shape the next app wants. A form sends "Jane Smith" when your CRM wants a first and a last name; a store sends 11/05/2025 when your calendar wants 2025-05-11; a payment app sends a product ID when your team chat should show the product's name. In Zapier, the fix is a Formatter step between the trigger and the action. This page explains what it does and walks through the transforms that solve most of these problems, based on Zapier's Formatter help as we read it on 2026-11-07.
What Formatter is
Zapier's own description: "Formatter is Zapier's built-in utility tool for transforming text, numbers, and other data into the format you need." Its help lists when to add one:
- data from one app "does not match the format another app expects (e.g., phone numbers, dates, number types)";
- you need to "split, extract, or combine text fields";
- you need to "translate values between apps (e.g., product IDs to friendly names)";
- you need "a default value for empty or missing data so later steps do not fail".
Two facts worth knowing before you add many of them: Zapier says "Formatter steps does not count toward your task usage", and it lists Formatter on the Free, Professional, Team and Enterprise plans. You can also describe what you need in plain language and let Zapier draft the Formatter step for you.
The four transforms you will use most
1. Dates and times
Date/Time > Format converts a date from one format to another. Zapier's example: a source app sends 11/05/2025 (DD/MM/YYYY) and the destination expects 2025-05-11 (YYYY-MM-DD). Two tips from the help:
- "Always set the From Format field when converting dates to avoid incorrect conversions."
11/05/2025is 11 May in most of the world and 5 November in the US; without the From Format, Formatter has to guess. - If your app returns UTC timestamps, use the same transform to convert to your local time zone — otherwise an evening event can land on the next day.
Related transforms add or subtract time (a reminder three days before) and compare two dates.
2. Splitting and combining text
Text > Split Text turns one field into several. Zapier's example: a trigger sends a single "Full Name" like "Jane Smith"; split on a space and take First for the first name and Last for the last name. The same idea splits a comma-separated list of tags or an address line.
Other text transforms in the same family change letter case, count characters or words, find and replace, remove extra text, extract an email or URL from a longer text, convert to HTML, Markdown or plain text, and match regular expressions.
3. Numbers and phone numbers
Numbers > Format Phone Number standardises a number such as 623-234-9012 into E.164, +16232349012, which many CRMs and ad platforms require. Format Number turns a string like "26.24" into a real number for apps such as Notion and Zoho that reject text in number fields. There are also math operations, random numbers and spreadsheet-style formulas for text and numbers.
4. Lookup tables
Utilities > Lookup Table maps one value to another. Zapier's example: Stripe sends a product ID and your Slack message should name the product — list each ID with its name, and the step outputs the name. Lookup tables also turn status codes into readable labels, country codes into country names, or plan IDs into prices to display.
The same Utilities group creates and converts line items, picks an item from a list, and imports CSV files.
The mistake that undoes it all
Zapier repeats one tip in two places: after a Formatter step, "map the Formatter step's Output field in your later actions—not the original trigger field." If you add a Formatter step and the next app still receives the raw value, the action is almost certainly still mapped to the trigger. Open the action, remove the old field, and insert the Output of the Formatter step instead.
Other common problems Zapier's Common Problems with Formatter page and its error articles cover: formatted data that comes out wrong in later steps, errors importing a CSV file, "Invalid Syntax: Only numeric values are allowed", and "413 Request Entity Too Large" when a step receives too much data.
A tidy pattern for any Zap
- Trigger — the raw data arrives.
- Formatter steps — one job each: format the date, split the name, map the ID, set a default for an empty field.
- Action — mapped only to Formatter outputs (and to trigger fields that needed no change).
One transform per step is easier to debug than one clever step: when something is wrong, the step's test output shows exactly which transform produced it. And since Formatter steps do not use tasks, splitting them up costs nothing.
When the next step makes an image
Some Zaps end in an image: a welcome card when someone joins, an event card when a session is added to a calendar, a quote card when a review comes in. The image is only as good as the text sent to it — a name in capitals, an unformatted date or an empty field shows on the picture for everyone to see. Formatter is where you fix that before the image is made.
We are the image step in that chain, and nothing more: Shareglint has no app in Zapier. A Webhooks by Zapier action sends the Formatter outputs to our API as JSON and gets back the URL of the finished image, which the next action can post, email or save. The Zapier recipe shows the request, the field mapping and a template to start with. Map the template's fields to Formatter outputs — the first name from Split Text, the date from Date/Time > Format — and the images come out clean.
Checklist
- Set From Format on every date conversion.
- Convert UTC to the audience's time zone before showing a time.
- Give empty fields a default value so later steps — and images — never show a blank.
- Map later steps to Formatter Output fields.
- Keep one transform per step; they cost no tasks.
