This article will help you:
- Understand how derived properties can benefit your analysis
- Review the functions and operators that can be used when creating your derived properties
In some cases, you may want to run analyses based on properties that were not sent to Amplitude but can be derived from existing properties. Amplitude Data’s derived properties allow you to create new event and user properties retroactively, based on functions and operators that you can apply across multiple existing properties. These do not affect your raw data and will be computed on the fly.
For example, you may want to create a chart that groups by whether an item added to a shopping cart is eligible for a discount. In that case, you could create a derived property whose value is a boolean based on whether the price exceeds a certain amount.
Create a derived property
NOTE: You must be in your project's main
branch to create a derived property.
To create a derived property, follow these steps:
- In Amplitude Data, navigate to Properties and click the Derived tab.
- Click + Add Derived Property.
- In the Derive New Property modal, enter a new property name (required) and description (optional).
- Enter your formula. See below for the list of valid functions and operators.
- Click Save.
Preview your results
As long as the formula you entered is valid, you can preview the results in the space below the formula editor. Do this by selecting existing values for properties used in the formula, or test out any free-form values. You can do this from either the Create/Edit modal or from the side panel for a saved derived property.
Derived property use cases
Taking our previous referrer URL example, you can write a formula using string operators that looks like this:
SPLIT(referrer_url, "/", 2)
This formula will convert a value like "https://www.google.com/search?q=amplitude" into the value "www.google.com." But what if you want to strip this down even further, to just "google"? You can achieve this by wrapping the result of a SPLIT function inside another SPLIT function. The resulting formula would look like this:
SPLIT(SPLIT(referrer_url, "/", 2), ".", 1)
Amplitude also supports math operators. Let’s say you have events that contain subtotal and tip properties, and you want to run some analyses based on the total amount. You can use this formula:
SUM(subtotal, tip)
Maybe you're also interested in knowing how many orders you would have given discounts to if the total order size was over $50. This formula will tell you whether a particular order would receive a discount:
IF(SUM(subtotal, tip) >= 50, true)
NOTE: Queries using derived properties may experience longer query times depending on the complexity of the formulas. There is also a limit of up to 10 property references per derived property.
Functions and operators
String functions
Function | Description | Example | Result |
---|---|---|---|
REGEXEXTRACT (text_property, regular_expression) | Extracts substrings matching the regular_expression | REGEXEXTRACT("shirt-150", "[0-9]+") | "150" |
REGEXREPLACE (text_property, regular_expression, replacement_text) | Replaces the property's values with text matching the regular_expression with replacement_text | REGEXREPLACE("en-US", "-.*", "") | "en" |
CONCAT(property1, property2) |
Concatenates a property with another property or text value. | CONCAT("firstName", "lastName") | "firstName lastName" |
LOWERCASE (text_property) | Lowercases all characters in property's values | LOWERCASE("John") | "john" |
UPPERCASE (text_property) | Uppercases all characters in property's values | UPPERCASE("John") | "JOHN" |
SPLIT (property, separator, [index]) | Split a property based on a delimiter and return an array of split elements. Takes an optional index that returns the element at that index. |
SPLIT("a_b_c", "_")
SPLIT("john@example.com", "@", 0) |
["a", "b", "c"]
"john" |
REMOVE (property, text) | Remove all occurrence of text in property | REMOVE("en-US", "en-") | |
EXTRACT_FROM_DICT (property, text) | Extract a value from a dictionary string based on a specific key | EXTRACT_FROM_DICT("{'id': 1, 'name': 'John', 'country': 'US'}", "name") | "John" |
Math functions
Function | Description | Example | Result |
---|---|---|---|
SUM(num_property1, num_property2) or ADDITION( |
Adds a property with other properties or with numbers. Equivalent to the `+` operator |
SUM(subtotal, tip) >>> SUM(10, 2) |
12 |
MINUS(num_property1, num_property2) or SUBTRACT( |
Subtracts a property with other properties or with numbers. Equivalent to the `-` operator. |
MINUS(total, tip) >>> MINUS(12, 2) |
10 |
MULTIPLY (num_property1, num_property2) |
Multiplies a property with other properties and/or with numbers. Equivalent to the `*` operator. |
MULTIPLY(price, quantity) >>> MULTIPLY(2.50, 4) |
10 |
DIVIDE(numerator, denominator) |
Divides a property by another property or number. Equivalent to the `/` operator. |
DIVIDE(calorie_intake, calorie_goal) >>> DIVID(1000, 2000) |
0.5 |
POWER(num_property, exponent) |
Takes the property's values to the exponent power |
POWER(property, 3) >>> POWER(2, 3) |
8 |
MIN(num_property1, num_property_2) |
Returns the minimum value between two numbers. |
MIN(5, 10) |
5 |
MAX(num_property1, num_property_2) |
Returns the maximum value between two numbers. |
MAX(5, 10) |
10 |
CEIL(num_property) |
Rounds up to the nearest integer. |
CEIL(3.8) |
4.0 |
FLOOR(num_property) |
Rounds down to the nearest integer. |
FLOOR(3.8) |
3.0 |
Object functions
Function | Description | Example | Result |
---|---|---|---|
EXTRACT_FROM_DICT (property, text) | Extract a value from a dictionary string based on a specific key | EXTRACT_FROM_DICT("{'id': 1, 'name': 'John', 'country': 'US'}", "name") | "John" |
Date/ time functions
NOTE: Amplitude requires all Unix timestamps to be expressed in milliseconds.
Function | Description | Example | Result |
---|---|---|---|
DATE_TO_LONG (date_property) |
Convert date into unix timestamp |
DATE_TO_LONG("2020-12-01") |
1606780800000 |
TIME_TO_LONG (time_property) |
Convert date time (YYYY-MM-dd[T]HH:mm:ss) into unix timestamp |
TIME_TO_LONG("2020-12-01 12:00:00") |
1606780800000 |
LONG_TO_TIME (number_property) |
Convert unix timestamp into date-time |
LONG_TO_TIME (1606780800000) |
"2020-12-01 12:00:00" |
LONG_TO_DATE (number_property) |
Convert unix timestamp into date |
LONG_TO_DATE (1606780800000) |
"2020-12-01" |
DATE_TIME_FORMATTER (datetime_property, old_format, new_format) |
Convert format of a datetime property to a new format. See Java SimpleDateFormat for more details. |
DATE_TIME_FORMATTER ("05.01.2021 12:00:00:000", "MM.dd.yyyy hh:mm:ss:SSS", "yyyy/MM/dd") |
"2021/05/01" |
TODAY() |
Current day represented as a long in epoch time in UTC. |
TODAY() - start_date_in_ms >>> 1609459200000 - 1577836800000 |
31622400000 |
EVENT_HOUR_OF_DAY() |
Get hour of day from the event's timestamp. (0-23) |
EVENT_HOUR_OF_DAY() |
10 |
EVENT_DAY_OF_WEEK() |
Get day of week from the event's timestamp as string. i.e. Monday |
EVENT_DAY_OF_WEEK() |
Monday |
Array functions
When performing computations on derived properties created from array properties, Amplitude assumes that only the first child is an array property, and only considers the first value of the other children, even if they are also array properties.
Here are some illustrative examples:
Example 1
propA = [1,2,3], propB = [a,b,c]
CONCAT(propA, propB) = [1a, 2a, 3a]
Example 2
propA = [1, 2, 3], propB = [a]
CONCAT(propA, propB) = [1a, 2a, 3a]
Example 3
propA = [1], propB = [a, b, c]
CONCAT(propA, propB) = [1a]
Function | Description | Example | Result |
---|---|---|---|
ITEM_COUNT (property) |
Length of array property; defaults to 1 for non-arrayed properties |
ITEM_COUNT(products*)
*products is an array property (e.g. ['apple', 'orange', 'banana']) |
3 |
GREATEST(property) |
Get max value of the array |
GREATEST(prices*)
*prices is an array property (e.g. [3.5, 10, 2]) |
10 |
LEAST(property) |
Get min value of the array |
LEAST(prices*)
*prices is an array property (e.g. [3.5, 10, 2])
|
2 |
COALESCE(property) |
Get the first non-null value of the array |
COALESCE(locations*)
*locations is an array property (e.g. [null, 'California', 'New York'])
|
'California' |
Conditional operators
Operator | Description | Example |
---|---|---|
IF(logical_expression, value_if_true, value_if_false) |
Returns value_if_true if logical_expression is true, otherwise return value_if_false |
IF(price == 0, "true", "false") IF(property == "(none)", "Property was not set", "Property was set") IF(OR(region == "California", region == "New York"), "USA", "Other") |
AND(logical_expression_1, logical_expression_2) |
Returns True if both logical expressions are true, false otherwise |
AND(is_subscribed == "true", has_valid_promo == "true") |
OR(logical_expression_1, logical_expression_2) |
Returns True if any logical expression is true, false otherwise |
OR(has_email == "true", has_phone == "true") |
SWITCH(expression, case_1, value_1, [case_2, value_2 ...], [default]) |
Evaluates an expression and returns values based on defined cases. Returns a default value if no cases are met if defined, otherwise null. |
SWITCH(tier, "gold", 2, "silver", 2, "bronze", 1, 0) |
String/numerical operators
Operator |
Example |
---|---|
== |
action == “purchase” |
!= |
item_count != 0 |
contains |
email contains “@gmail.com” |
does not contain |
title does not contain “officer” |
<, <=, >, >= |
duration >= 60 |
glob match |
url glob match “https://www.google.*/*” |
glob does not match |
query glob does not match “*/query=*“ |
has prefix |
title has prefix “sir” |
Set operators
Set literals (e.g. ("apple", "orange")) must appear on the right hand side of the operator
Operator |
Example |
---|---|
== |
IF(product == (“apple”,“orange”), "true", "false") product = “apple”, Returns "true" |
!= |
IF(product != (“apple”,“orange”), "true", "false") product = “banana”, Returns "true" |
Derived Properties Formulas
Get the difference between two dates
DIVIDE(
SUBTRACTION(
DATE_TO_LONG(
$1 ),
DATE_TO_LONG(
$2 )
),
86400000
)
Sample Output:
In the derived property above, the properties end_date and start_date are converted into UNIX timestamps and then we find the difference between them. Finally, we divide that number by 86400000 which is the number of milliseconds in 1 day. Please note that the output will be a double type (e.g. 2.0). At the bottom, you can preview some of the results of the derived property. Replace the $<number>
here with the actual properties
Standardized Date Format
IF(DATE_TIME_FORMATTER(
$60 ,
"yyyy-MM-dd'T'HH:mm:ssX",
'yyyy-MM-dd'
) contains '-', DATE_TIME_FORMATTER(
$61 ,
"yyyy-MM-dd'T'HH:mm:ssX",
'yyyy-MM-dd'
), IF(DATE_TIME_FORMATTER(
$72 ,
"yyyy-MM-dd HH:mm:ss",
'yyyy-MM-dd'
) contains '-', DATE_TIME_FORMATTER(
$73 ,
"yyyy-MM-dd HH:mm:ss",
'yyyy-MM-dd'
), DATE_TIME_FORMATTER(
$76 ,
"yyyy-MM-dd",
'yyyy-MM-dd'
)))
Sample Output:
One possible way to format dates to Standard Date Format would be to use a series of IF statements, but you need to make sure the higher specificity conditional comes first. Replace the $<number>
here with the actual properties
Get Sign Up Month and Year
CONCAT(
REGEXEXTRACT($107,
'\d\d\d\d\-\d\d'
),
"-01"
)
In the example above, the derived property will pull the sign-up month and year from a property that contains a more detailed value and append the "-01" to set it to the beginning of the month. You can use the REGEXEXTRACT() to pull the Year and Month from the value and use CONCAT() to append the "-01" to the sign-up month and year. Replace the $<number>
here with the actual properties.
Replace Property Values
IF(
OR(
REGEXEXTRACT($47,'Casual) =='Casual',
REGEXEXTRACT($61,'1 Job Posting')=='1 Job Posting',
REGEXEXTRACT($67,'1 Basic')=='1 Basic'
), 'Casual', 'False')
Sample Output:
To replace multiple property values, you can use REGEXEXTRACT() to pull the string in the property and use OR statements inside an IF statement to see if the value pulled from the properties contains any of the values you wish to replace. In the example above, if the property value matches any of the values specified, it will replace the value with "Casual". Otherwise, it will replace the property with "False". Replace the $<number>
here with the actual properties.