Power Automate isFloat() Function: Check if a String Is a Number

The Power Automate isFloat() function checks whether a string can be read as a decimal number and returns true or false. Use it before float(), so a value like “N/A” or “$25.00” doesn’t break your flow with a conversion error.

I ran into this with a Contoso expense form: people typed amounts as “125.50”, “125,50” or “about 125”, and the flow that added them up kept failing. In this tutorial, I’ll show the syntax, what isFloat really returns for tricky inputs (I tested each one in a flow), how the locale parameter works, and a safe pattern to convert the text.

Quick answer: isFloat('<string>', '<locale>') returns true if the string is a valid number, including whole numbers such as “10” and negatives such as “-3.75”. Text with letters or currency symbols returns false, and so does an empty string. The optional locale, such as 'de-DE', tells it which decimal separator to expect.

isFloat() syntax

isFloat('<string>', '<locale>'?)

Parameter	Required	What it does
string	Yes	The text to check. It must be a string, so wrap numbers or nulls with string() first.
locale	No	An RFC 4646 code such as en-US, de-DE or fr-FR. Without it, the invariant culture is used (period for decimals, comma for thousands).

The function only validates. It doesn’t convert anything, so you still need float() to get a number you can add or compare. That’s covered in converting a string to a float.

What isFloat() returns: tested examples

I built a Compose action with eleven isFloat expressions and ran the flow. Here’s the expression list:

And these are the results from the run:

Expression	Result	Why
isFloat(‘10.5’)	true	A normal decimal
isFloat(’10’)	true	Whole numbers are valid floats too
isFloat(‘1,250.75’)	true	Comma is the thousands separator in the invariant culture
isFloat(‘10,5’)	true	Read as 105, not 10.5, because the comma is a thousands separator
isFloat(‘10,5’, ‘de-DE’)	true	German uses the comma as the decimal separator
isFloat(‘1.250,75’, ‘de-DE’)	true	German thousands and decimal separators
isFloat(‘-3.75’)	true	Negative numbers are fine
isFloat(‘$25.00’)	false	Currency symbols aren’t allowed
isFloat(‘abc’)	false	Not a number
isFloat(”)	false	An empty string returns false, without an error
isInt(‘10.5’)	false	isInt rejects decimals

Two results surprise most people. First, “10” returns true, so isFloat can’t tell you whether a value has a decimal point. Second, “10,5” returns true without a locale, but it means 105. If your users type commas as decimals, pass their locale.

Check a value and convert it safely

The common pattern is: validate with isFloat, then convert with float(). Here’s how I set it up in the expense flow, where the amount comes in as text from a form or a SharePoint text column:

1. Add a Condition action.
2. In the left box, select fx and enter isFloat(triggerBody()?['text']). Set the operator to is equal to and the right box to the expression true.
3. In the True branch, convert the value with float(triggerBody()?['text']) and use it, for example in an Update item action.
4. In the False branch, send the submitter an email that the amount isn’t a valid number.

If you prefer one expression, use if() to fall back to zero. This is handy for initializing a Float variable from user input:

if(isFloat(triggerBody()?['text']), float(triggerBody()?['text']), 0)

To round the result or drop the decimals afterward, see how to convert a float to an integer.

Use the locale parameter for European formats

The same text means different numbers in different countries. Pass the locale to both isFloat and float, so the check and the conversion agree:

Text	Locale	isFloat	float() result
1,250.75	none (invariant)	true	1250.75
1.250,75	de-DE	true	1250.75
10,5	de-DE	true	10.5
10,5	none (invariant)	true	105

if(isFloat(variables('Amount'), 'de-DE'), float(variables('Amount'), 'de-DE'), 0)

If you work with users in several countries, read the locale from the user’s profile or a list column instead of hard-coding it. A wrong locale doesn’t throw an error, it just returns false or the wrong number.

isFloat vs isInt vs float

Function	Returns	Use it to
isFloat(string, locale?)	true/false	Check that text is any valid number
isInt(string)	true/false	Check that text is a whole number (no locale parameter)
float(value, locale?)	a number	Convert text to a decimal number (errors on invalid text)
int(value)	a number	Convert text to a whole number

To accept only values with decimals, combine the two checks: and(isFloat(x), not(isInt(x))). The isInt side is explained in the isInt function tutorial.

Filter an array to keep only numeric values

isFloat works inside the Filter array action too. To keep only rows where the Amount text is a number, switch the Filter array condition to advanced mode and use:

@isFloat(string(item()?['Amount']))

The string() wrapper matters: if a row has an empty Amount, item()?['Amount'] is null, and isFloat expects a string. With string(), null becomes an empty string, which returns false instead of failing the action.

Common errors

• InvalidTemplate … expected a string: you passed a Number column, an integer variable or null. Wrap it: isFloat(string(coalesce(x, ''))).
• isFloat says true but the math is off by a factor of 10 or 100: the decimal separator is a comma. Add the locale, such as ‘de-DE’, to isFloat and float.
• Currency values fail: remove the symbol first, for example replace(x, '$', '').
• Values with spaces fail: trim the text with trim() before checking it.

If you’re new to writing expressions, testing them in a Compose action first, as I did here, saves a lot of failed runs.

More Power Automate expression tutorials:

• Easiest Way to Convert a String to a Decimal Number in Power Automate
• Power Automate addDays() Function + Examples
• How to Use formatDateTime() Function in Power Automate?
• How to Send Email Using If Condition in Power Automate?

Frequently Asked Questions