BigQuery Datetime Functions Mastery

Google BigQuery SQL
SQL Copilot for BigQuery

Understanding and manipulating time-based data is pivotal for data analysts, engineers, BI professionals, SQL developers, and IT consultants. This guide aims to educate you on the comprehensive use of datetime functions in BigQuery, enhancing your proficiency in data manipulation, analysis, and reporting capabilities.

As the digital world becomes increasingly time-sensitive, accurately handling, analyzing, and reporting on datetime data can significantly impact decision-making processes and business strategies. By mastering datetime functions in BigQuery, you streamline your data workflows and unlock advanced analytics capabilities. This enables you to analyze complex time series, forecast trends, and generate insights crucial for competitive advantage.

Understanding Datetime in BigQuery and How it Differs from Timestamp

Datetime and timestamp functions in BigQuery are pivotal for handling date and time data. Each serves distinct needs depending on the context of the data analysis or application. The critical difference lies in how they store time zone information.

Datetime values represent a specific date and time but do not contain time zone data, making them best suited for scenarios where the time zone is implicit or uniformly understood, such as within a single geographic location or system-wide settings where all data is localized.

On the other hand, the timestamp function is ideal for recording global events or data, where understanding the precise moment of an event requires acknowledging the time zone in which it occurred.

Key Datetime Functions and Their Usage (With Examples)

Understanding key datetime functions in BigQuery is essential for anyone looking to perform complex temporal data analysis and manipulation. These functions allow for a wide range of operations, from calculating differences between dates to adding or subtracting time intervals to formatting datetime values for more readable outputs.

We can use practical examples to explore how these functions are applied to solve real-world data challenges, such as tracking event durations, scheduling future events, or generating reports based on specific time frames.

CURRENT_DATETIME

The CURRENT_DATETIME function returns the current date and time as a DATETIME object without timezone information. This function is handy for timestamping data processing events or calculating durations from the current moment. It reflects the server's current date and time, typically used in real-time reporting and data entry timestamping.

Syntax:

CURRENT_DATETIME()

Here:

  • CURRENT_DATETIME(): Returns the current date and time. The exact format of the returned value depends on the SQL dialect you're using (e.g., YYYY-MM-DD HH:MM:SS).

Example: For a practical example that illustrates fetching the current server date and time in BigQuery without specifying a timezone, consider a situation where you must timestamp a user's action in your application's backend.

SELECT CURRENT_DATETIME() AS now;

In this example:

  • CURRENT_DATETIME(): Captures the server's current date and time, assigning this timestamp to the alias “now”.

The outcome, tagged as now, holds the datetime value, such as 2024-04-20 11:51:40, simplifying its application in logging, auditing, and event tracking within the backend of various applications.

DATETIME

The DATETIME function is pivotal in handling date and time values within BigQuery, allowing the creation of DATETIME objects from the specified year, month, day, hour, minute, and second components. It's instrumental when you need to construct a DATETIME from individual components or convert a TIMESTAMP to DATETIME for uniformity in data without timezone influences.

Syntax:

DATETIME([timestamp_expression[, timezone]])

Here:

  • DATETIME(): A function that converts a given timestamp into a datetime format.
  • timestamp_expression: The input expression, typically a timestamp, that you want to convert into a datetime format.
  • timezone (optional): An optional parameter that specifies the timezone to use for the conversion, affecting the resulting datetime value.

Example: Suppose, a hospital's administration team must schedule a critical system upgrade in a healthcare setting without interrupting patient care services. They decide on a specific downtime during a period of typically lower activity: July 15, 2021, at 10:30 AM.

To communicate and plan this precisely within their IT systems and staff schedules, they use BigQuery to create a DATETIME object representing this scheduled downtime.

SELECT DATETIME(2021, 7, 15, 10, 30, 0) AS system_upgrade_time;

In this example:

  • (2021, 7, 15, 10, 30, 0): Is used to set a precise upgrade time for July 15, 2021, at 10:30 AM, aiding the hospital's meticulous planning.

  • system_upgrade_time: It clarifies the upgrade moment, which was chosen based on historical data to minimize patient care disruptions.

This SQL query efficiently schedules a hospital system upgrade during low activity hours, ensuring seamless patient care by aligning IT enhancements with historical data insights.

💡If handling timestamp data manually is causing you trouble, find a tool that can simplify timestamp manipulation and eliminate the constraints of manual methods. Discover our comprehensive guide on using timestamp functions in BigQuery for efficient time data processing and analysis.

Dive deeper with this read

Navigating BigQuery Timestamp Functions: An Insightful Guide

Image for article: Navigating BigQuery Timestamp Functions: An Insightful Guide

DATETIME_ADD

The DATETIME_ADD function allows adding a specified time interval to a DATETIME value, making it indispensable for projections and scheduling future events. By selecting the unit of time to add, such as days or months, you can calculate future dates from a known DATETIME.

Syntax:

DATETIME_ADD(datetime_expression, INTERVAL expression date_part)

Here:

  • DATETIME_ADD(): A function that adds a specified time interval to a datetime value and returns the resulting datetime.
  • datetime_expression: The datetime value to which the interval will be added. This expression should be evaluated to a datetime.
  • INTERVAL expression: Specifies the quantity of the date part to add. This expression defines how much time to add to the datetime_expression.
  • date_part: Defines the type of interval to add (e.g., DAY, MONTH, YEAR), indicating the unit of the time to be added to the datetime_expression. DATETIME_ADD supports a range of values for the part parameter to adjust a datetime value with precision. These include smaller time units such as MICROSECOND, MILLISECOND, SECOND, MINUTE, and HOUR. For broader time increments, the function allows adjustments by DAY, with WEEK being equivalent to adding seven days. It also supports larger temporal segments like MONTH, QUARTER, and YEAR, enabling users to perform adjustments that span across different lengths of time periods efficiently.

Example: In the context of a marketing department planning its annual strategy, they decide to launch a major promotional campaign precisely 3 months after the beginning of the fiscal year to align with the end of a quarter. The fiscal year starts on January 1, 2023. To accurately determine the launch date for this campaign, they utilize the DATETIME_ADD function in BigQuery.

SELECT DATETIME_ADD(DATETIME "2023-01-01", INTERVAL 3 MONTH) AS campaign_launch_date;

In this example:

  • DATETIME "2023-01-01": Specifies the initial datetime value (January 1, 2023) to which the interval will be added.
  • INTERVAL 3 MONTH: Specifies that the interval being added is 3 months.
  • AS campaign_launch_date: Renames the result of the DATETIME_ADD function to campaign_launch_date for clarity in the output of the query.

By leveraging DATETIME_ADD, the team ensures that their planning is data-driven and aligned with broader organizational timelines, facilitating seamless execution and evaluation of the promotional strategy.

DATETIME_SUB

DATETIME_SUB subtracts a specified time interval from a DATETIME value, facilitating backward calculations and historical data analysis. This function helps understand past events relative to a given date by subtracting days, months, or years from a specified DATETIME.

Syntax:

DATETIME_SUB(datetime_expression, INTERVAL expression date_part)

Here:

  • DATETIME_SUB(): A function that subtracts a specified time interval from a datetime value, returning the resulting datetime.
  • datetime_expression: The datetime value from which the interval will be subtracted. This expression should be evaluated to a datetime.
  • INTERVAL expression: Specifies the quantity of the date part to subtract. This expression defines how much time to subtract from the datetime_expression.
  • date_part: Defines the type of interval to subtract (e.g., DAY, MONTH, YEAR), indicating the unit of time to be subtracted from the datetime_expression.

Example: Suppose a university's academic department needs to determine the deadline for submitting final grades after the semester ends, they have a policy requiring grades to be submitted within 5 days after the last day of finals. The finals' week for the spring semester ends on January 10, 2023.

To ensure compliance and facilitate the smooth operation of grade processing, the department calculates the submission deadline using the DATETIME_SUB function in BigQuery.

SELECT DATETIME_SUB(DATETIME "2023-01-10", INTERVAL -5 DAY) AS grade_submission_deadline;

In this example:

  • DATETIME "2023-01-10": Specifies the initial datetime value (January 10, 2023) from which the interval will be subtracted.
  • INTERVAL 5 DAY: Indicates that the interval being subtracted is 5 days.
  • AS grade_submission_deadline: Renames the result of the DATETIME_SUB function to grade_submission_deadline for easier identification in the query's output.

This example allows the academic department to effectively communicate this crucial date to the faculty by identifying the specific deadline well in advance, ensuring that all grades are processed and recorded on time.

DATETIME_DIFF

The DATETIME_DIFF function calculates the difference between two DATETIME values, returning the difference in specified units. It's essential for measuring intervals and durations between dates, supporting a broad range of temporal analyses.

Syntax:

DATETIME_DIFF(datetime_expression, datetime_expression, date_part)

Here:

  • DATETIME_DIFF(): A function that calculates the difference between two datetime values, returning the result as an integer.
  • datetime_expression (first occurrence): The first datetime value, from which the second datetime value will be subtracted.
  • datetime_expression (second occurrence): The second datetime value to subtract from the first datetime value.
  • date_part: Specifies the unit for measuring the interval difference (e.g., DAY, MONTH, YEAR), determining how the difference between the two datetime expressions is calculated and presented.

Example: Suppose, in a construction project scenario, a project manager wants to assess the actual duration between the project's start and completion dates to evaluate project efficiency and adherence to timelines. The project officially began on January 1, 2023, and was completed on December 31, 2023.

To calculate the total number of days the project was in operation, which is vital for performance analysis, future planning, and reporting to stakeholders, the project manager uses the DATETIME_DIFF function in BigQuery.

SELECT DATETIME_DIFF(DATETIME "2023-12-31", DATETIME "2023-01-01", DAY) AS project_duration_days;

In this example:

  • DATETIME "2023-12-31": Specifies the end datetime value for the calculation.
  • DATETIME "2023-01-01": Specifies the start datetime value for the calculation.
  • DAY: Indicates that the difference is calculated in days.
  • AS project_duration_days: Renames the result of the DATETIME_DIFF function to project_duration_days for easier identification in the query's output.

This precise measurement of the project timeline enables the project manager to compare planned versus actual durations, assess efficiency, and identify areas for improvement.

DATETIME_TRUNC

DATETIME_TRUNC truncates a DATETIME value to a specified component, such as year or month, facilitating data aggregation and simplification. It helps group data by larger time units without losing the context of the original dates.

Syntax:

DATETIME_TRUNC(datetime_expression, date_part)

Here:

  • DATETIME_TRUNC(): A function that truncates a datetime value to the specified component, effectively rounding it down to that component.
  • datetime_expression: The datetime value to be truncated. This is the value that the function will modify based on the specified date_part.
  • date_part: Specifies the unit of time to which the datetime should be truncated (e.g., YEAR, MONTH, DAY). This determines how much of the datetime expression is preserved and what is rounded down.

Example: Imagine, a retail company wants to analyze monthly sales data to identify trends and prepare for inventory adjustments, it needs to aggregate sales data monthly. The analysis will include determining the total sales starting from the first day of each month.

To streamline this process, especially when dealing with sales transactions recorded with specific dates and times, the company utilizes the DATETIME_TRUNC function in BigQuery.

SELECT DATETIME_TRUNC(DATETIME "2023-07-15", MONTH) AS month_start;

In this example:

  • DATETIME "2023-07-15": Specifies the datetime value to be truncated.
  • MONTH: Indicates that the datetime should be truncated to the start of the month, effectively rounding it down to July 1, 2023.
  • AS month_start: Renames the result of the DATETIME_TRUNC function to month_start for easier identification in the query's output.

By applying this function across their sales data, the company can uniformly align all transaction dates to the start of their respective months, simplifying the aggregation process.

EXTRACT

EXTRACT pulls a specific component from a DATETIME value, such as year, month, or day, offering precision in temporal data analysis. This function enables the isolation of particular date parts for detailed examination and comparison.

Syntax:

EXTRACT(part FROM datetime_expression)

Here:

  • EXTRACT(): A function used to extract a specific part (e.g., year, month, day) from a datetime or interval expression.
  • part: Specifies the component of the datetime expression to be extracted, defining specific segments of the datetime value. Options range from precise units like MICROSECOND and MILLISECOND to broader divisions such as SECOND, MINUTE, HOUR, and DAY. Additional components include DAYOFWEEK (returns Sunday as day 1), DAYOFYEAR, and weekly calculations with WEEK (numbered 0 to 53, starting Sunday) or WEEK() for a specified start day. ISOWEEK tracks the ISO week number starting Monday. For longer periods, MONTH, QUARTER, YEAR, and ISOYEAR are available, along with basic DATE and TIME components for full date or time extraction.
  • FROM datetime_expression: Indicates the source from which the specified part should be extracted. The datetime_expression is the datetime value or field from which the extraction is to occur.

Example: Imagine, in a marketing context, a company plans to analyze the effectiveness of its email marketing campaigns throughout the year to identify which months generate the highest engagement rates. Each campaign's launch date is recorded in their database as a DATETIME value.

To focus their analysis on the timing of these campaigns, the marketing team decides to extract the monthly component from each campaign's launch date.

SELECT EXTRACT(MONTH FROM DATETIME "2023-04-15") AS campaign_launch_month;

In this example:

  • MONTH: Specifies that the month part of the datetime expression is to be extracted.
  • FROM DATETIME "2023-04-15": Indicates the datetime value from which the month will be extracted.
  • AS campaign_launch_month: Renames the result of the EXTRACT function to campaign_launch_month for easier identification in the query's output.

By performing this operation across all campaign records, the marketing team can aggregate engagement data by month to pinpoint when their efforts have been most and least effective.

FORMAT_DATETIME

FORMAT_DATETIME converts a DATETIME value into a formatted string based on a specified pattern, enhancing readability for reports and presentations. This function allows for customized date and time representations.

Syntax:

FORMAT_DATETIME(format_string, datetime_expression)

Here:

  • FORMAT_DATETIME(): A function used to format a datetime value according to a specified format string.
  • format_string: The template string that defines the desired output format of the datetime value. This string includes placeholders that specify how various components of the datetime (like year, month, day) should be displayed.
  • datetime_expression: The datetime value to be formatted. This expression evaluates to a datetime that will be converted into a string according to the format_string.

Here's a list of format string elements described along with applicable function contexts like date, datetime, and timestamp:

  • `%A`: Full weekday name in English (e.g., "Wednesday"). Applicable to date, datetime, and timestamp functions.
  • `%a`: Abbreviated weekday name in English (e.g., "Wed"). Applicable to date, datetime, and timestamp functions.
  • `%B`: Full month name in English (e.g., "January"). Applicable to date, datetime, and timestamp functions.
  • `%b`: Abbreviated month name in English (e.g., "Jan"). Applicable to date, datetime, and timestamp functions.
  • `%c`: Complete date and time representation in English (e.g., "Wed Jan 20 21:47:00 2021"). Applicable to datetime and timestamp functions.
  • `%D`: Date in the format mm/dd/yy (e.g., "01/20/21"). Applicable to date, datetime, and timestamp functions.
  • `%d`: Day of the month as a zero-padded decimal number (e.g., "20"). Applicable to date, datetime, and timestamp functions.
  • `%e`: Day of the month as a space-padded decimal number (e.g., " 9" for the 9th of the month). Applicable to date, datetime, and timestamp functions.
  • `%F`: Full date in the format yyyy-mm-dd (e.g., "2021-01-20"). Applicable to date, datetime, and timestamp functions.
  • `%H`: Hour (24-hour clock) as a zero-padded decimal number (e.g., "23"). Applicable to time, datetime, and timestamp functions.
  • `%h`: Same as `%b`, abbreviated month name in English (e.g., "Jan"). Applicable to date, datetime, and timestamp functions.
  • `%I`: Hour (12-hour clock) as a zero-padded decimal number (e.g., "09"). Applicable to time, datetime, and timestamp functions.
  • `%j`: Day of the year as a zero-padded decimal number (e.g., "020"). Applicable to date, datetime, and timestamp functions.
  • `%k`: Hour (24-hour clock) as a space-padded decimal number (e.g., " 9"). Applicable to time, datetime, and timestamp functions.
  • `%l`: Hour (12-hour clock) as a space-padded decimal number (e.g., " 9"). Applicable to time, datetime, and timestamp functions.
  • `%M`: Minute as a zero-padded decimal number (e.g., "47"). Applicable to time, datetime, and timestamp functions.
  • `%m`: Month as a zero-padded decimal number (e.g., "01"). Applicable to date, datetime, and timestamp functions.
  • `%n`: Newline character. Generally used in string formatting.
  • `%P`: Lowercase 'am' or 'pm' (e.g., "pm"). Applicable to time, datetime, and timestamp functions (formatting only).
  • `%p`: Uppercase 'AM' or 'PM' (e.g., "PM"). Applicable to time, datetime, and timestamp functions.
  • `%Q`: Quarter of the year as a decimal number (e.g., "1"). Applicable to date, datetime, and timestamp functions.
  • `%R`: Time in the format hh:mm (e.g., "21:47"). Applicable to time, datetime, and timestamp functions.
  • `%S`: Second as a zero-padded decimal number (e.g., "00"). Applicable to time, datetime, and timestamp functions.
  • `%s`: Seconds since the Unix Epoch (e.g., "1611179220"). Applicable to time, datetime, and timestamp functions.
  • `%T`: Time in the format hh:mm:ss (e.g., "21:47:00"). Applicable to time, datetime, and timestamp functions.
  • `%t`: Tab character. Generally used in string formatting.
  • `%U`: Week number of the year, with Sunday as the first day of the week, as a zero-padded decimal number (e.g., "03"). Applicable to date, datetime, and timestamp functions.
  • `%u`: Day of the week as a decimal, Monday as 1 (e.g., "3"). Applicable to date, datetime, and timestamp functions.
  • `%W`: Week number of the year, with Monday as the first day of the week, as a zero-padded decimal number (e.g., "03"). Applicable to date, datetime, and timestamp functions.
  • `%w`: Day of the week as a decimal, Sunday as 0 (e.g., "3"). Applicable to date, datetime, and timestamp functions.
  • `%X`: Time representation in HH:MM:SS format (e.g., "21:47:00"). Applicable to time, datetime, and timestamp functions.
  • `%x`: Date representation in MM/DD/YY format (e.g., "01/20/21"). Applicable to date, datetime, and timestamp functions.
  • `%Y`: Year with century as a decimal number (e.g., "2021"). Applicable to date, datetime, and timestamp functions.
  • `%y`: Year without century as a zero-padded decimal number (e.g., "21"). Applicable to date, datetime, and timestamp functions.
  • `%%`: Literal '%' character. Used to escape the '%' in format strings.

Example: Suppose, a marketing team needs to prepare a clear and understandable report for the board meeting, they aim to present the start date of their summer promotional campaign in an instantly understandable format.

To achieve this, they use the following SQL query in BigQuery.

SELECT FORMAT_DATETIME("%B %d, %Y", DATETIME "2023-07-15") AS formatted_date;

In this example:

  • "%B %d, %Y": The format string indicating how the datetime should be formatted. "%B" is for the full month name, "%d" is for the day of the month, and "%Y" is for the four-digit year.
  • DATETIME "2023-07-15": Specifies the datetime value to be formatted.
  • AS formatted_date: Renames the result of the FORMAT_DATETIME function to formatted_date for easier identification in the query's output.

This SQL operation transforms a DATETIME into a reader-friendly format, significantly enhancing report clarity for board meetings by presenting key dates quickly and optimizing communication effectiveness.

LAST_DAY

LAST_DAY returns the last day of the month for a given DATETIME value, aiding in monthly closing processes and period-end reporting. This function simplifies the calculation of end-of-period dates.

Syntax:

LAST_DAY(datetime_expression[, date_part])

Here:

  • LAST_DAY() Function: A function that returns the last day of the period specified by the date_part in the given datetime_expression.
  • datetime_expression: The datetime value for which the last day of a specified period is sought. This expression should evaluate a datetime from which the function will calculate the previous day.
  • date_part (optional): An optional parameter that specifies the period (e.g., MONTH, YEAR) for which the last day should be calculated. If omitted, the function typically defaults to finding the last day of the month.

Example: Let's consider a scenario where a company wants to calculate the last day of the month for the end date of a promotional campaign. The campaign starts on July 15, 2023, and they want to ensure it runs until the last day of that month to maximize exposure and participation.

Given this requirement, the company can use the LAST_DAY function in BigQuery to determine the exact date the campaign should end. The SQL query for this would be the following.

SELECT LAST_DAY(DATETIME "2023-07-15") AS campaign_end_date;

In this example:

  • DATETIME "2023-07-15": Specifies the datetime value for calculating the last day of the month.
  • AS campaign_end_date: Renames the result of the LAST_DAY function to campaign_end_date for easier identification in the query's output.

This result helps the company plan and executes the promotional campaign by providing a precise end date. It also ensures that the campaign is aligned with the company's monthly financial and marketing activities, allowing for better resource allocation and performance tracking.

PARSE_DATETIME

PARSE_DATETIME converts a string into a DATETIME value based on a specified format, bridging the gap between textual date representations and DATETIME objects for processing and analysis.

Syntax:

PARSE_DATETIME(format_string, string)

Here:

  • PARSE_DATETIME(): A function that converts a string into a datetime value based on the specified format string.
  • format_string: The template string that specifies how the parts of the datetime are represented in the input string. This format string defines the pattern that the input string must follow.
  • string: The actual string to be converted into a datetime value. This string should match the pattern defined by the format_string.

There are a few additional things we need to remember for conversion while using PARSE_DATETIME:

  • When using the PARSE_DATETIME function in SQL, the format elements within the format string must correspond precisely to the elements within the datetime string being parsed.

The following works because elements match on both the sides:

SELECT PARSE_DATETIME("%m/%d/%Y", "12/25/2023");

  • Each element in the datetime string must have a corresponding element in the format string, and their positions must align.
  • Mismatched elements or missing elements between the format string and the datetime string will result in errors during parsing.

The following code will produce an error.

SELECT PARSE_DATETIME("%Y-%m", "2023-12-01");

  • However, certain format elements like %c in the format string can match multiple elements in the datetime string, allowing for flexibility in parsing datetime values with varying formats.

The following is one such example:

SELECT PARSE_DATETIME("%c", "Thu Dec 25 07:30:00 2008");

Example: Let's consider a scenario where we have a string representing a datetime in a format that includes the day of the week, month name, day of the month, time in hours, minutes, and seconds, and finally, the year. We want to convert this string into a DATETIME object.

Given the datetime string "Mon Jan 1 15:00:00 2018", we will parse this string using a format string that matches its structure with the following syntax.

SELECT PARSE_DATETIME("%a %b %e %H:%M:%S %Y", "Mon Jan 1 15:00:00 2018") AS datetime;

  • "%a %b %e %H:%M:%S %Y": The format string indicating the pattern of the datetime components in the input string. Here, %a is the abbreviated weekday name, %b is the abbreviated month name, %e is the day of the month, %H is the hour (24-hour clock), %M is the minute, %S is the second, and %Y is the four-digit year.
  • "Mon Jan 1 15:00:00 2018": The string to be converted into a datetime value, matching the pattern defined by the format string.
  • AS datetime: Renames the result of the PARSE_DATETIME function to datetime for easier identification in the query's output

This query will correctly parse the string and return a DATETIME object representing January 1, 2018, at 3:00:00 PM, as shown below:

This approach effectively converts a textual datetime representation into a structured DATETIME object, allowing for further manipulation and analysis within BigQuery.

💡 If managing date data manually is causing you trouble, find a tool that can simplify date manipulation and eliminate the constraints of manual methods. Discover our comprehensive guide on using date functions in BigQuery for efficient date data processing and analysis.

table

Get BigQuery Reports in Seconds

Seamlessly generate and update reports in Google Sheets—no complex setup needed

Start Reporting Now

Advanced Datetime Conversion Techniques

Familiarizing with datetime conversion techniques in data analysis within BigQuery is crucial for effectively managing and interpreting temporal data. Two advanced methods particularly stand out:

Converting Datetime to Date Using the CAST Function

To convert a datetime value to a date in BigQuery, you can use the CAST function. This method is proper when the time component of a datetime is not needed for analysis or reporting. By casting a datetime to a date, you effectively remove the time portion, focusing solely on the date aspect.

Syntax:

FORMAT_DATETIME(format_string, CAST(string AS DATE))

Here:

  • FORMAT_DATETIME: This is the function used in BigQuery to format a datetime value according to a specified format string.
  • format_string: This is a parameter of the FORMAT_DATETIME function. It is a string literal that specifies the format in which the datetime value should be displayed. This string can include format elements such as %Y for the year, %m for the month, %d for the day, %H for the hour in 24-hour format, %M for the minute, %S for the second, and so on. For example, "%Y-%m-%d %H:%M:%S" would format the datetime value as "YYYY-MM-DD HH:MM:SS".
  • CAST(string AS DATE): This is another parameter of the FORMAT_DATETIME function. It is used to convert a string value to a date data type. The CAST function is used for this purpose, where the string is the value to be converted, and AS DATE specifies the target data type. This part of the function ensures that the input value is a valid date before formatting it.

Example: Let's consider a scenario where you have a log of customer registrations in a database, and each registration includes a timestamp indicating when the registration occurred. You want to extract and format just the date part of each registration timestamp in the "YYYY-MM-DD" format for analysis.

Here's how the SQL code can be applied in this scenario:

SELECT
  FORMAT_DATETIME("%Y-%m-%d", CAST('2021-07-15 08:30:00' AS DATETIME)) AS event_date;

This query will output 2021-07-15, removing the time component and leaving just the date.

Converting Timestamp Into Datetime

The operation removes the timezone information from the timestamp to convert a timestamp into a datetime in BigQuery. This gives a datetime value representing the same time point in a specific timezone, typically UTC. This conversion is proper when working with local time representations without considering timezone differences.

Syntax:

DATETIME(timestamp_expression[, timezone])

Here:

  • timestamp_expression: This is the value or field to be converted into a DATETIME object.
  • timezone (optional): Indicates the timezone for the conversion. If specified, the DATETIME function will use this timezone to interpret the timestamp_expression.

Example: Here's an example of converting a timestamp into a datetime in BigQuery, where you must account for or ignore timezone differences.

Suppose you have a timestamp value of 2023-03-29 15:00:00 UTC and want to convert it into a datetime in the UTC timezone.

SELECT DATETIME(TIMESTAMP "2023-03-29 15:00:00 UTC", "UTC") AS datetime_value;

This query converts the timestamp 2023-03-29 15:00:00 UTC into a datetime value, keeping the same point in time but now represented as a datetime without explicit timezone information. This makes handling local time operations or comparisons in the UTC timezone easier.

💡 If converting data types manually is causing you trouble, find a tool that can simplify data conversion and eliminate the constraints of manual methods. Discover our comprehensive guide on using conversion functions in BigQuery for efficient data transformation and processing.

table

Simplify BigQuery Reporting in Sheets

Easily analyze corporate data directly into Google Sheets. Query, run, and automatically update reports aligned with your business needs

Simplify Reporting Now

Best Practices for Application of Datetime Function

Managing dates and times is a fundamental aspect that plays a critical role across many applications, from scheduling and logging to expiration checks and historical data analysis. The application of datetime functions, which enable developers and analysts to represent, manipulate, and perform calculations with dates and times, requires a better understanding of best practices.

Filling in Missing Parts of Dates

Ensure completeness using functions like COALESCE or IFNULL to provide default values for missing date parts. This approach helps maintain your datasets' integrity, especially when dealing with incomplete date information. For instance, if your data only includes a year and month, you can use these functions to assume the first day of the month for any missing day parts, ensuring that your dataset remains usable and consistent.

Managing Overlapping Datetime Format Elements

Be cautious with formats that could be ambiguous, and use the most specific format strings possible. This is crucial for preventing confusion between similar-looking date parts, like month (MM) and minute (mm), especially in international contexts where date formats vary. Choosing the correct format string helps clarify your data's meaning and ensures others interpret it correctly.

Flexibility with AM/PM Indicators in Datetime Formats

Use FORMAT_DATETIME to handle AM/PM indicators flexibly, catering to different user preferences. This function lets you format date and time most intuitively for your audience, whether they prefer a 24-hour clock or the 12-hour format with AM/PM. It's a simple yet effective way to make your reports and analyses more accessible to a broader audience.

Handling Date Transitions at Month-Ends with Special Attention

Pay extra attention to month-end transitions, using LAST_DAY to avoid date overflow errors. This is particularly important when calculating expiration dates or scheduling events towards the month's end, as it ensures you don't accidentally roll over into the next month. Utilizing LAST_DAY helps accurately capture the intended last date of the month, avoiding common pitfalls in date-related calculations.

Uncover in-depth insights

Modern Data Management Guide

Download now

Bonus for readers

Modern Data Management Guide

Broaden Your Skills with More BigQuery Functions

If you are looking to advance your skills in Google BigQuery, it's essential to understand its more complex functions and capabilities.

  • Array Functions: Explore array functions to handle and process arrays within your queries for more dynamic data structuring.
  • String Functions: Utilize string functions to manipulate and analyze textual data efficiently in your BigQuery queries.
  • DML: Gain proficiency in DML to perform updates, deletions, and insertions that modify data stored in BigQuery efficiently.
  • Aggregate Functions: Understand aggregate functions to perform calculations over a range of values, which is crucial for summarizing data insights in your queries.
  • Numbering Functions: Assign unique or ranked numbers to rows in a result set, enabling effective ordering and partitioning of data.
  • Conditional Expressions: Perform logic-based operations, returning different results based on specified conditions.
  • Navigation Functions: Access values in other rows without the need for self-joins, facilitating lead or lag data operations within partitions.

Troubleshooting Errors in Datetime Functions in BigQuery

When dealing with datetime functions in BigQuery and encountering errors, it often boils down to a few common issues. Let’s take a look at those down below.

Avoiding Overflow Errors in DATETIME_DIFF Calculations:

⚠️ Error:

Overflow errors occur when the difference between two dates, calculated with DATETIME_DIFF, exceeds the maximum value of the chosen unit (e.g., seconds, minutes). This often happens when the date range is extensive, and the unit of difference is too granular, leading to numbers that BigQuery cannot handle, which can halt data processing.

✅ Solution:

To prevent overflow errors, select units for DATETIME_DIFF that match the scale of your analysis – opt for days, months, or years instead of seconds or minutes when dealing with extended date ranges. Plan your queries to anticipate the range of differences you expect to encounter. Conduct basic checks to ensure datetime ranges are within a reasonable and expected bound.

Syntax:

SELECT
  DATETIME_DIFF(datetime_end, datetime_start, 'Unit') AS time_difference
FROM
  your_dataset

Here:

  • DATETIME_DIFF(datetime_end, datetime_start, 'Unit'): This function calculates the difference between two datetime values (datetime_end and datetime_start) in the specified unit ('Unit').
  • datetime_end and datetime_start: These represent the ending and starting points of the datetime range you're analyzing. Ensure that these dates are within a reasonable range to avoid overflow when calculating differences in granular units.
  • 'Unit': Specifies the unit of time to calculate the difference in (e.g., 'DAY', 'MONTH', 'YEAR').
  • time_difference: The alias for the result of the DATETIME_DIFF calculation, representing the difference between the two datetimes in the chosen unit.

Example: If you're analyzing the documents' age and choose to calculate the difference in seconds, you might exceed the function's capacity for old documents. A safer approach would be to calculate the difference in days or months. For instance:

SELECT DATETIME_DIFF(DATETIME '2024-03-29', DATETIME '2000-01-01', MONTH) AS diff_months

This calculates the difference in months, reducing the risk of overflow compared to calculating in seconds or minutes.

Format Mismatch Errors in PARSE_DATETIME

⚠️ Error:

Format mismatch errors arise when there's a discrepancy between the format string specified in PARSE_DATETIME and the actual format of the input datetime string. Even slight differences, such as an unexpected separator or an incorrect order of date and time components, can cause these errors. Such mismatches make the datetime data unusable, as BigQuery fails to interpret the input correctly based on the provided format string.

✅ Solution:

To prevent format mismatch errors, ensure that the format string used in PARSE_DATETIME precisely matches the format of your input datetime data. Before parsing, validate the format of your input data to confirm it aligns with your format string.

Use conditional logic or regular expressions to pre-process and clean your data if it does not conform to the expected format. Regularly reviewing your input data for consistency and format variations can also help identify and correct mismatches before they lead to errors in your queries.

Syntax:

SELECT
  PARSE_DATETIME("%Y-%m-%d %H:%M:%S",
    REGEXP_REPLACE(your_datetime_string, "Pattern_To_Replace", "Replacement_Pattern")) AS parsed_datetime
FROM
  your_dataset

Here:

  • PARSE_DATETIME("%Y-%m-%d %H:%M:%S", your_datetime_string): Converts a string to a datetime object.
  • REGEXP_REPLACE(your_datetime_string, "Pattern_To_Replace", "Replacement_Pattern"): Before parsing, this function is used to adjust the format of the datetime string, ensuring it conforms to the expected pattern.
  • your_datetime_string: This represents the datetime string being converted.
  • parsed_datetime: This is the alias for the result of PARSE_DATETIME, representing the correctly parsed datetime object.

Example: If your datetime string is "2024-03-29 15:00:00", but you use a format string expecting a different order or separators, an error will occur.

For example, a correct usage would be:

SELECT PARSE_DATETIME('%Y-%m-%d %H:%M:%S', '2024-03-29 15:00:00') AS parsed_datetime

This aligns the format string '%Y-%m-%d %H:%M:%S' with the input, ensuring accurate parsing.

Unspecified Fields Takes Default Value in PARSE_DATETIME

⚠️ Error:

When fields are unspecified in PARSE_DATETIME, BigQuery assigns default values to these fields, which can lead to unexpected results. This automatic filling of default values, such as setting the time to 00:00:00 when no time is specified, may not align with the actual context or intended interpretation of your data, especially when working with partial dates or times.

✅ Solution:

To mitigate issues with default values, specify as complete a datetime as possible when using PARSE_DATETIME. When dealing with partial datetime information, consider utilizing additional functions within BigQuery to manually set or adjust the unspecified parts of your datetime values to accurately reflect your data's actual context.

Being aware of and proactively managing how default values affect your data's interpretation is crucial for maintaining the accuracy and integrity of your analysis and reports.

Syntax:

SELECT
  PARSE_DATETIME("%Y-%m-%d %H:%M:%S",
    IFNULL(your_datetime_string, {{Default_DateTime}})) AS accurate_datetime
FROM
  your_dataset

Here:

  • PARSE_DATETIME("%Y-%m-%d %H:%M:%S," your_datetime_string): This function converts a string into a datetime object based on the specified format.
  • IFNULL(your_datetime_string, {{Default_DateTime}}): In cases where parts of your datetime string are missing, IFNULL can replace null values with a default datetime.
  • your_datetime_string: This variable holds the datetime string you're converting. It may come from a column in your dataset.
  • {{Default_DateTime}}: represents a placeholder for the default datetime value you provided.

Example: When using PARSE_DATETIME and omitting parts of the datetime, BigQuery fills in missing fields with default values, often leading to unintended results. For instance, if you parse a date without specifying the time:

SELECT PARSE_DATETIME('%Y-%m-%d', '2024-03-29') AS parsed_datetime

BigQuery assumes a time of 00:00:00 for the missing time component. This behavior is crucial to understand when dealing with partial datetime values to avoid misinterpreting your data.

Enhance Your Reporting Capabilities With OWOX BI BigQuery Reports Extension

Utilize the OWOX BI BigQuery Reports Extension to seamlessly integrate and visualize your time-based data analysis, enhancing your reporting capabilities with rich, dynamic, and actionable insights. This powerful tool allows you to effortlessly merge your BigQuery data with intuitive visualization features, making it easier than ever to uncover trends and patterns in your time-based datasets.

table

Get BigQuery Reports in Seconds

Seamlessly generate and update reports in Google Sheets—no complex setup needed

Start Reporting Now

This comprehensive guide on BigQuery datetime functions equips you with the knowledge to effectively manipulate, format, and analyze time-based data, ensuring accurate and insightful data-driven decisions. By mastering these functions, you'll unlock the full potential of your time-based datasets, extracting valuable insights that drive business growth and innovation.

Whether you're a data analyst, BI professional, SQL developer, or IT consultant, this guide provides the tools and techniques to succeed in today's data-driven world.

FAQ

Expand all Close all
  • What is the difference between datetime and timestamp in BigQuery?

    Datetime represents a date and time without timezone information, ideal for calendar dates. Timestamp, however, includes date, time, and timezone, used for precise moments in time across different time zones.
  • How can I add a specific time interval to a Datetime value in BigQuery?

    Use the DATETIME_ADD function, specifying the Datetime value, the interval to add (e.g., hour, day), and the amount. 

    For example, 

    DATETIME_ADD(DATETIME "2023-04-01", INTERVAL 1 DAY) adds one day.

  • How do I extract the month from a Datetime value in BigQuery?

    Use the EXTRACT function with the Datetime value to get the month. 

    Example: 

    EXTRACT(MONTH FROM DATETIME "2023-04-01") returns 4.

  • How can I format a Datetime value into a more readable format in BigQuery?

    Use the FORMAT_DATETIME function, specifying the desired format string and the DateTime value. 

    For instance, 

    FORMAT_DATETIME("%B %d, %Y", DATETIME "2023-04-01") formats the date as "April 01, 2023".

  • What is the correct way to handle timezone conversions in BigQuery?

    Convert Timestamp to a specific timezone with TIMESTAMP functions like TIMESTAMP_SECONDS, adding the timezone parameter. For Datetime, use the DATETIME functions, considering it has no timezone.
  • What are the best practices for troubleshooting errors in datetime functions in BigQuery?

    Ensure correct format matching, anticipate overflow errors by choosing appropriate units for DATETIME_DIFF, and understand the default values behavior in functions like PARSE_DATETIME. Regularly validate and preprocess your datetime data to prevent common mistakes.