- start_date - Start date in Excel date serial number format.
- end_date - End date in Excel date serial number format.
- unit - The time unit to use (years, months, or days).
The DATEDIF (Date + Dif) function is a "compatibility" function that comes from Lotus 1-2-3. For a long time, official documentation on DATEDIF was sparse. Even now (May 2021), Excel will not help you fill in arguments for DATEDIF like other functions. As the late Chip Pearson once wrote in his immortal words: DATEDIF is treated as the drunk cousin of the Formula family. Excel knows it lives a happy and useful life, but will not speak of it in polite conversation. Yet DATEDIF remains a very useful function for certain problems.
The DATEDIF function can calculate the time between a start_date and an end_date in years, months, or days. The time unit is specified with the unit argument, which is supplied as text. The table below summarizes available unit values and the result for each. Time units can be given in upper or lower case (i.e. "ym" is equivalent to "YM").
|"y"||Difference in complete years|
|"m"||Difference in complete months|
|"d"||Difference in days|
|"md"||Difference in days, ignoring months and years|
|"ym"||Difference in months, ignoring years|
|"yd"||Difference in days, ignoring years|
Example 1 - Basic usage
In the example shown above, column B contains the date January 1, 2016 and column C contains the date March 1, 2018. In column E:
E5=DATEDIF(B5,C5,"y") // returns 2 E6=DATEDIF(B6,C6,"m") // returns 26 E7=DATEDIF(B7,C7,"d")// returns 790
Example 2 - Difference in days
The DATEDIF function can calculate the difference between dates in days in three different ways: (1) total days, (2) days ignoring years, and (3) days ignoring months and years. The screenshot below shows all three methods, with a start date of June 15, 2015, and an end date of September 15, 2021:
The formulas used for these calculations are as follows:
=DATEDIF(B5,C5,"d") // total days =DATEDIF(B6,C6,"yd") // days ignoring years =DATEDIF(B7,C7,"md") // days ignoring months and years
Note that because Excel dates are just large serial numbers, the first formula does not need DATEDIF and could be written as simply the end date minus the start date:
=C5-B5 // end-start = total days
Example 3 - Difference in months
The DATEDIF function can calculate the difference between dates in months in two different ways: (1) total complete months, (2) complete months ignoring years. The screenshot below shows both methods, with a start date of June 15, 2015, and an end date of September 15, 2021:
=DATEDIF(B5,C5,"m") // complete months =DATEDIF(B6,C6,"ym") // complete months ignoring years
DATEDIF always rounds months down to the last complete number of months. This means DATEDIF rounds the result down even when it is very close to the next whole month. In addition, DATEDIF may not work as expected when start and end dates are "end of month" dates. This example provides more information and alternatives.
Example 4 - Difference in years
The DATEDIF function can calculate the difference between dates in complete years with just one method, shown below:
=DATEDIF(B5,C5,"y") // complete years =DATEDIF(B6,C6,"y") // complete years =YEARFRAC(B7,C7) // fractional years with YEARFRAC
Notice in row 6 the difference is almost 6 years, but not quite. Because DATEDIF only calculates complete years, the result is still 5. In row 7 we use the YEARFRAC function to calculate a more accurate result.
Example 5 - Age from birthday
The DATEDIF function can be used together with the TODAY function to calculate current age from a birth date. With a full birth date in A1, the formula is:
Read a complete explanation here.
- Excel will not help you fill in the DATEDIF function like other functions.
- DATEDIF will throw a #NUM error if start_date is greater than the end_date. If you are working with a more complex formula where start dates and end dates may be unknown, or out of bounds, you can trap the error with the IFERROR function, or use MIN and MAX to sort out dates.
- Microsoft recommends not using the "MD" value for unit because it "may result in a negative number, a zero, or an inaccurate result".