Skip to content
ConsultEvo

How to Use Formatter by Zapier to Clean and Map Data

Formatter by Zapier transforms data between steps in a Zap so a later app can use it. Add a Formatter action after a trigger or earlier action, configure the transformation, test its output, and map that output into the destination field.

Start with the destination field’s requirements rather than with a list of available transformations. Record the source field, required output, and what should happen when the value is blank or malformed. Formatter handles deterministic representation changes. It does not decide whether a value belongs to the intended customer, is allowed by policy, or is correct for a business process.

For example, a form might send customer_name_raw as “Ava Chen” while a CRM expects separate first-name and last-name fields. A Split Text step can produce those values when the delimiter and segment rule are reliable. If the source contains ambiguous or irregular data, route it for review instead of forcing a guess.

What Formatter by Zapier does and when to use it

Formatter is Zapier’s built-in utility for transforming text, numbers, dates, line items, and other data between workflow steps. Its documented action categories include Date/Time, Numbers, Text, and Utilities, although the available transforms can change.

The basic operating chain is:

Source field -> Formatter action and transform -> tested output -> destination field

Use Formatter when two apps represent the same information differently, such as a date in one format and a date field in another, or a known source label that must become an approved destination code. For stable delimiters, exact substitutions, known date formats, and approved mappings, a deterministic rule is preferable to AI. Consider AI or custom code only when the input requires semantic interpretation beyond a defined formatting rule.

Zapier’s current Formatter documentation lists Professional, Team, and Enterprise plans as available and Free as unavailable. Formatter steps do not count toward task usage in Zap workflows, according to Zapier’s task-usage documentation. In-line formulas are a separate Early Access feature, so confirm their current availability and billing treatment before relying on them. Check Zapier’s current pricing and plan details before publishing a workflow that depends on a plan entitlement.

Choose the transformation from the destination field contract

A field contract is a short definition of the acceptable input, required output, and exception behavior. For example, a proposed phone-field contract might specify customer_phone_raw as the source, an E.164 destination format, a United States country assumption, and an exception route when the country or number is unknown. The country assumption and failure rule belong to the process owner, not to Formatter.

Match the contract to a deterministic operation. Split text when the delimiter is stable. Use Date/Time when the source format and timezone are known. Use a lookup for an approved set of source-to-destination values. Convert Markdown and HTML when the receiving app expects a different representation. If the input varies in meaning rather than format, define a separate interpretation and validation step.

A transformation is useful only when its output matches a defined destination field contract. A formatted value is not automatically a valid business value.

Split text and extract values without assuming perfect input

For a stable delimiter, add a Formatter step, choose Text and Split Text, map the source into Input, and specify the separator. Select the segment you need, such as First, Second, Last, Second to Last, or All as line items. Zapier documents these setup options in its Split Text instructions. Some separators, including spaces, require the syntax shown in the current editor.

For a hypothetical source value of Ava Chen, splitting on a space and mapping the first and second segments can produce:

{
  "first_name": "Ava",
  "last_name": "Chen"
}

The example is suitable only when the source consistently contains two parts. A name such as “Maria de la Cruz” contains more segments, so a simple space split cannot reliably infer which parts belong in first-name and last-name fields. Test blank input, extra spaces, middle names, and compound surnames. If the value does not fit the contract, send it to the owner of the form or CRM field definition for correction.

Formatter’s built-in extraction transforms can find an email address, number, phone number, or URL in text, but they return the first matching value, not every match. Use Extract Pattern or another suitable parsing method when the input contains multiple candidates or needs a custom pattern. Test the pattern against actual examples, including punctuation and missing values, before mapping the result. Zapier documents extraction behavior and recognition limits in its data extraction guide.

Format dates and times with explicit assumptions

In a Date/Time Formatter action, map the source into Input, specify From Format when the source could be interpreted more than one way, select To Format, and set From Timezone and To Timezone when timezones matter. If From Timezone is blank, the documented default is UTC. Zapier recommends specifying the input format to reduce incorrect interpretation.

For example, 11/05/2025 could mean November 5 or May 11, depending on the source convention. Confirm that convention with the source-system owner before configuring the transform. For a local appointment time, set both the known source timezone and intended destination timezone, then compare the result with a known expected date and time. A successful test status does not prove that the date was interpreted correctly. See Zapier’s date and time formatting instructions.

01Define the field contractRecord the source field, required destination format, assumptions, and failure behavior. The process owner approves the rule.
02Configure the deterministic transformChoose the relevant Formatter action and provide the delimiter, pattern, format, timezone, or approved mapping.
03Test representative recordsInspect ordinary and edge-case inputs. Check the actual output against the contract, not only the step’s success message.
04Apply a validation gateCheck required values and allowed results before a consequential write. Route unresolved cases to a named operations or data owner.
05Map and verify the destination writeMap the tested Formatter output, not the original field, and confirm the resulting record in the destination system.

Map known values with a lookup table and a fallback

A Formatter Lookup Table translates a known input into its corresponding output, with an optional fallback for values that do not match. In an illustrative mapping, widget might become WIDGET_V2. In the later action, map the lookup output to the destination field, not the original source label. Zapier’s lookup-table instructions note that a blank lookup key produces no match and that Formatter Lookup Tables require Professional or higher.

Choose a fallback that cannot be mistaken for a valid destination code, or route an unmatched value to an exception path. Formatter Lookup Tables suit step-local mappings, line-item processing, and fallback behavior. Consider Zapier Tables when the mapping is a shared data asset that needs broader maintenance. The product-data owner should approve changes to codes used by orders or inventory.

Decision point

An unmatched lookup is not a successful mapping. Check the lookup status before the CRM or product-system write, and send unresolved labels to the mapping owner instead of passing through the original value.

Convert Markdown and choose the right publishing control

Formatter supports text conversions including Markdown to HTML, HTML to Markdown, conversion to ASCII, and removal of HTML tags. Use the conversion required by the receiving field, then inspect the rendered result in the destination. The documentation establishes conversion, not sanitization or publishing safety.

Zapier’s public GitHub-to-Ghost template page identifies a GitHub New Issue trigger, a Formatter by Zapier text step, and a Ghost Create Post action. The public page does not establish every field mapping, post status, approval step, duplicate control, or failure route. Open and test the configuration in your account before relying on it.

A proposed controlled design is to filter for a designated ready-to-draft label, convert the issue body, create a Ghost draft, preserve the GitHub issue ID where the destination supports it, and assign a content owner to review and publish. Those gates are implementation choices, not controls verified on the public template page. Conversion and publication remain separate decisions.

Use an in-line formula or a separate Formatter step?

An in-line formula is entered directly in a mapped destination field. A separate Formatter step creates a visible intermediate output that later actions can map. An in-line formula can suit a simple, local, one-off change. Prefer a separate step when multiple actions need the result, when the transformation needs a descriptive label, or when you want to inspect it independently during troubleshooting.

Zapier describes in-line formulas as a separate Early Access feature with limited availability and active development. Check the current setup guidance and formula reference before building around one. Do not assume that formulas share Formatter’s documented task-usage treatment. Verify the live action output because formula testing may display the formula in the data view rather than showing the final transformed value.

Practical workflow patterns

Split a stable field before a CRM write

Trigger: a form provides customer_name_raw. Formatter: Text, Split Text, with a documented separator and segment selection. AI job: none, because the rule is deterministic. Validation: reject blanks, unexpected segment counts, and compound names that do not fit the contract. Destination: map the tested segments to the CRM’s first-name and last-name fields. Fallback: send the source record to the form owner for correction.

Convert a date only after defining its source convention

Trigger: an order or booking event supplies appointment_time_raw. Formatter: Date/Time Format with explicit From Format, From Timezone, To Format, and To Timezone. AI job: none, because AI cannot reliably resolve an undocumented date convention. Validation: compare the output with a known expected date and time. Destination: map the verified value to the scheduling or CRM field. Fallback: stop the write and send the event to the process owner when the source convention is unknown.

Map an approved product label to a destination code

Trigger: a sales or order event supplies product_label. Formatter: Lookup Table mapping widget to the illustrative code WIDGET_V2. AI job: none, because an approved lookup must be exact. Validation: confirm that the result is a recognized code and is not the fallback. Destination: map the lookup output to the inventory, order-management, or CRM field. Fallback: route an unmatched label to the product-data owner rather than writing the original label.

Convert a GitHub issue before a Ghost post action

Trigger: the verified public template identifies GitHub New Issue. Formatter: a Text conversion from Markdown to HTML, subject to testing in the configured Zap. AI job: none in the documented sequence. Validation: confirm the title and body mappings, post status, review gate, and duplicate controls in the account. Destination: Ghost Create Post, as shown on the public template page. Fallback: create a draft or route the issue to content operations if the configured workflow does not establish safe publication controls.

Test the workflow before mapping it into a live destination

Use representative records, inspect the Formatter output, test the destination mapping, and confirm the resulting record in the system of record. For a CRM write, preserving the source record ID and, when auditability matters, the original and transformed values can help trace a change. These are implementation recommendations, not automatic Formatter outputs.

Review before rollout
  • Test blank values, extra separators, malformed input, and multiple extraction candidates.
  • For dates, verify the source convention, input format, and both relevant timezones.
  • For lookups, distinguish an approved result from a fallback or no-match result.
  • Verify the destination mapping and inspect the created or updated record.
  • Assign an exception owner for unresolved or high-impact values.
  • Define a stable source-event key and destination uniqueness rule before enabling repeated writes.

A search-then-create sequence alone is not guaranteed to prevent duplicates when concurrent runs can occur. A Zap run is an execution grain, not the same thing as a source event or destination record. Where duplicate prevention matters, use a stable key such as source_system + source_record_id + transformation_version + destination and enforce uniqueness in the destination database, or use a transactional upsert where available. Treat this key as an illustrative design that must match the destination’s row grain. If no reliable destination control exists, route possible duplicates for review.

For teams defining CRM field contracts, ownership, and exception paths, CRM systems consulting may be relevant. For help designing or maintaining the workflow sequence, see Zapier automation services.

The practical rule is simple: automate predictable representation changes, validate allowed values before the destination write, and assign a person to ambiguous or high-impact exceptions.