6 Mistakes to Avoid in Power Automate

Many users schedule triggers in workflows. While scheduling triggers is useful for generating summary reports, running a bot frequently to check for new emails is a waste of resources.

Solution: Instead, adopt an efficient approach with event-based triggers. These triggers allow a flow to run when a specific action occurs.

There are numerous triggers available, divided into two categories:

• Automated
• Instant

These triggers let you connect with a wide range of services, such as running a flow when you send or receive a Teams message or an email.

Power Automate workflows have an action limit of 500, but it’s advisable to use no more than 400 actions. Exceeding 400 actions can degrade workflow performance, slowing down the flow.

Solution: Create child flows instead of using too many actions.

There are two ways to do that

1. Using APIs Calls
2. Use the inbuilt action “Run a Child Flow“

This helps in maintaining optimal performance and preventing slowdowns.

An array is a collection of similar items, such as a list. For instance, all the records of a patient grouped together form an array. Extracting specific data from an array by having a bot iterate through the entire dataset can be time-consuming. This slows down the bot and the overall process.

• Solution: Apply relevant Filters to remove unnecessary data, allowing the bot to read important data faster.

For example, in healthcare, data ca be filtered based on active records, name, number of fields available, etc.

Speed up data extraction and improve processing efficiency.

When calling APIs, the information is received in JSON format, which can have varying structures. To make Power Automate understand which entity and value to take, we use the Parse JSON action. This action requires a JSON schema with key types and values, which could be strings, integers, etc.

• Issue: Sometimes, certain values may be blank, and many developers struggle to handle this situation.
• Solution: Use the “any of” function with null to inform the bot that the data can be in the given format or may be blank.
• Benefit: This prevents errors and allows efficient handling of thousands of lines of data.

• Example:

Imagine you are processing customer data from a JSON payload. One of the fields is “middleName”, which might be null for some customers. If you define the schema as:

jsonCopy code{"properties": {"middleName": { "type": "string" }}}

The flow will fail when “middleName” is null. Instead, define it as:

jsonCopy code{"properties": {"middleName": { "type": ["string", "null"] }}}

This way, the Parse JSON action will handle null values gracefully, and the flow will continue without errors.

By using Parse JSON correctly and allowing for null values, you ensure robust and error-free automation that can handle a variety of data scenarios.

jsonCopy code{"type": ["string", "null"]}

Running independent actions one after the other or sequentially takes a lot of time of the bot and in turn slows down the whole process. When you wait for one action to complete to start another action, the whole flow disrupts. It leads to:

1. If the first action fails, the second action can never start.
2. It slows down the whole process.

To overcome this issue, simply run independent actions parallelly. Since both actions have no repercussion on each other, it won’t affect any part of your process and will lead to a reduction in execution time and improve efficiency.

In Power Automate Desktop, a retry policy ensures that if the bot cannot interact with a UI element, it will retry a number of times by refreshing the page. This resolves issues caused by pop-ups or conflicting elements.

• Issue: The default policy can lead to indefinite retries without alerts, making errors difficult to detect.
• Solution: By implementing a proper retry policy to ensure retries are limited to 3 or 4, errors are promptly identified.

1. Efficient Data Processing for Reports

You have a large dataset but need to create reports for specific users.

Solution: Instead of letting the bot read the entire dataset, filter out the data to improve the bot’s efficiency.

Example: Stakeholders need a report on late arrivals in the manufacturing industry. Rather than going through all data, apply a filter for arrivals after a certain time. This allows the bot to complete the task quickly by avoiding unnecessary data.

2. Parallel Execution for Independent Reports

You need to generate 10 different reports for a user, each independent of the others.

Solution: Run these reports in parallel since they have no interdependencies.

Example: If you need to upload a report to both OneDrive and Azure Blob Storage, these independent actions can be executed simultaneously. This parallel processing speeds up the overall task completion.

3. Event-Based Report Generation

Instead of generating reports on a fixed weekly or monthly schedule, use event-based triggers.

Solution: Trigger report generation based on specific events, such as the submission of an adaptive card.

Example: When a person fills out an adaptive card, a process is automatically triggered to generate the required report. This approach ensures timely and relevant report generation based on actual events rather than predetermined schedules.

These 6 strategies not only save time and resources but also improve the accuracy and reliability of your automated processes.

Utilizing these best practices in Power Automate will enable your business to:

• Save time by 20 hours per month per employee on repetitive tasks
• Reduce Operational costs by 27%
• 30% increase in overall process Efficiency

Start optimizing your Power Automate workflows today with Sunflower Lab’s expert Power Automate team. Contact Us today and our team will be happy to assist you further.