AudienceFlow Overview

Audience Flow is a SmartFeeds tool designed to automate the management of your audience lists. The primary goal is to improve the precision and performance of your media campaigns by moving from periodic, manual list uploads to a fully automated system that synchronizes your audiences daily.

This guide explains the tool's functionalities, the data required, how to configure your audiences, and how to interpret the results and troubleshoot common issues.

Prerequisites

To use this functionality, you must have access to the Audience Flow module within the SmartFeeds platform.

Audience Flow: Loading, Segmenting, and Activating Your Data

The Audience Flow workspace provides a central hub for managing your audiences, from the initial data connection to the final export to media platforms.

Step 1: Create or Select a Flow

Before you can add data sources or build audiences, you must first create a new "Flow" or open an existing one. This Flow acts as a container for your entire audience configuration. The process for creating a flow is standard across the SmartFeeds platform. For a detailed guide, please see the ProductFlow’s article

Step 2: Prepare Your Data

To use Audience Flow effectively, your data must be structured into two main categories:

Onboarding Data
This information is essential for matching your users with their profiles on media platforms. This is your own first-party data, meaning it has been collected directly from your customers. It is typically extracted from your company's internal systems, such as a Customer Relationship Management (CRM) tool, a Customer Data Platform (CDP), or a central data warehouse.
As a security standard, all personally identifiable fields listed below must be hashed using the SHA256 algorithm. SHA256 is a standard security function that irreversibly transforms data (like an email address) into a unique string of characters, ensuring that the original personal information remains protected. Specific formatting rules must be applied before hashing.

Required Fields: email, firstname, lastname
phone_number: Must be formatted to the E.164 standard (e.g., +33678787878) before being hashed.
country_code: This field has a conditional requirement:
- For Meta platforms, it must also be hashed.
- For all other platforms, it must be provided as a clear, unhashed two-letter uppercase code (ISO 3166-1 alpha-2).
postal_code

Hashing Formula Example: To hash a field, you can use a formula similar to this: UPPER(TO_HEX(SHA256(your_table.your_field)))

Segmentation Data
This behavioral and transactional data allows you to create refined audience segments. This can include online/offline purchase history, basket composition, and other custom indicators available in your system. You can verify that this data is correctly implemented and ready to use when you create your first audience in Step 4.

Step 3: Add Data Sources

A data source is required to begin.

Click the + icon in the "Data Sources" section to add your first source. You have several options to connect your data:
- SFTP (Secure File Transfer Protocol): This method allows for the automated and secure transfer of files. It is highly recommended for recurring updates but typically requires assistance from your IT or technical team to set up.
- URL Link: Ideal for files hosted in the cloud, such as a Google Sheet. For this to work, the file must be publicly accessible or shared with our service account. For a Google Sheet, you must share it with the following email: adscale@smart-feeds.iam.gserviceaccount.com.
- Cloud Data Sources: Direct connections to platforms like Google Cloud Storage.
- Local File: A manual upload of a file (e.g., Excel or CSV) from your computer.
To enrich your data, you can add multiple data sources. Simply click the + icon again and repeat the connection process for each new source. If you use more than one, you will need to configure the Join Key to link them. Important: Before you can configure a join, each data source must have been successfully fetched at least once. If the join is not configured correctly, only the data from your primary (first) source will be accessible for mapping and audience creation. This setting is found by clicking the gear icon (⚙️) in the data sources column.
Once connected, your data source will appear, showing its name, the last fetch time, and the total number of imported rows.

Step 4: Create and Manage an Audience

With a data source active, you can build and modify your audiences.

To create a new audience, click + Add an audience in the "Audiences" section. This will open the Formula Editor, where you can define your segmentation rules.
To edit an existing audience, click the pencil icon on the audience card. This will take you back to the Formula Editor to make any necessary changes at any time.

Step 5: Configure an Export and Map Your Data

An audience must be exported to a media platform to be used in campaigns.

On the audience card you created, click Add an export. Audience Flow provides automatic exports to platforms like Google Ads and Meta. It's important to note that you must select an export channel before you can proceed to the mapping stage. The system needs to know the destination platform to understand which specific fields are expected for the mapping.
Next, you must configure the Mapping. This defines how the columns in your source file correspond to the fields required by the ad platform. While ProductFlow is structured around the data transformation process (input, modification, output), Audience Flow is designed with the audience cards as the central focus to provide a clear overview of each audience and its destinations. As mapping is a supporting function in this view, it is accessed through the data source settings. To do this, click the gear icon (⚙️) in the data sources column.

Note: Is the platform you're looking for not on the list? We are constantly expanding our integrations. Contact us to discuss your project, and we can explore creating a custom connector for you.

Understanding Audience Size: Match Rate vs Activable Audience

To help you understand the performance of your audiences, Audience Flow uses two key concepts: the Match Rate and the Activable Audience size.

1. The Match Rate: A Quality Score for Your Data

The Match Rate is a percentage that measures the quality of your list. It compares the number of contacts you provided to the number of contacts that the advertising platform (like Google or Meta) was able to successfully recognize. The calculation method is specific to each platform, so a score on Google Ads may differ from a score on Meta for the same list.

In Audience Flow, a color-coded status helps you interpret this score at a glance:

Blue (Processing): When you first send an audience, it will appear in blue. This means the platform is still processing your list, which can take time. The final match rate is not yet guaranteed.
Green (Good): Once the platform has finished processing, the audience will turn green if the match rate is high. This indicates that your data is of good quality.
Orange (Average): An orange status means the match rate is acceptable but could be improved.
Red (Poor): A red status signals a low match rate. This is a clear indicator that you should review and rework your audience data to improve its quality.

2. Activable Audience: Who You Can Actually Target

It is crucial to understand that even with a "Good" (green) match rate, the Activable Audience—the final number of users you can target in a campaign—will almost always be smaller than your initial list.

For example, you might import a list of 565,990 contacts and get a high match rate, but find the final activated audience is only 60,000 users. This is normal and happens for two main reasons:

User Consent: A user may have a profile on the platform, but if they have opted out of ad personalization in their privacy settings, they cannot be targeted.
Recent Activity: Platforms like Google only include users who have been recently active on their services (e.g., Search, YouTube). A matched user who has been inactive for a long period will not be included in the activated audience.

Strategies and Best Practices

To maximize the effectiveness of your audiences, it is important to combine strategic segmentation with a few key operational best practices.

Strategic Use Cases

Optimize Acquisition by Excluding Existing Customers: One of the most powerful and immediate uses is to improve the efficiency of your prospecting campaigns. Create a comprehensive audience of all your existing customers and use it as an exclusion list in your acquisition campaigns. This ensures you aren't spending money to acquire users you already have.
Boost Retention and Re-engagement: Go beyond simple retargeting. Build dynamic audiences of customers who show signs of lapsing, such as those who haven't made a purchase in the last 90 or 180 days. Target them with specific "we miss you" offers to encourage their return.
Drive Cross-Sell and Up-sell Opportunities: Use transactional data to identify purchasing patterns. For example, create an audience of customers who bought "Product A" but not the complementary "Product B". You can then target them with ads showcasing the benefits of using both products together.
Build High-Quality Lookalike Audiences: The performance of lookalike (or similar) audiences depends heavily on the quality of the source list. Use Audience Flow to create a highly-qualified "seed" audience, such as your top 5% of customers based on lifetime value. Use this hyper-qualified list to let ad platforms find new users with similar profiles, significantly improving your prospecting results.

Key Operational Practices

Respect the Initial Population Delay: When you send a new audience to a platform like Google Ads, always wait 24 to 48 hours for it to be fully processed. Analyzing its size or making changes before this period will not give you an accurate picture.
Prioritize Data Security with Hashing: Always ensure your personally identifiable information (PII) is securely hashed using SHA256, ideally before it is even uploaded to SmartFeeds. This is not just a best practice; it is essential for security and compliance.
Choose the Right Update Frequency: While the tool can support daily updates, a weekly refresh of your audience lists is a reliable and recommended standard. This provides a good balance between audience freshness and system stability.
Meet Minimum Audience Size Requirements: Be aware that platforms have minimum size thresholds to protect user anonymity. For a Google Ads campaign, for instance, an audience must contain at least 1,000 matched and activated users to be eligible for targeting.

Quick Answers to Common Issues

My Google Ads audience hasn't updated in several days, what could be the reason?

As of April 7, 2025, Google Ads requires a lifespan value for all customer lists, indicating how long a user should remain in the audience. Previously, this field was optional. If an audience created before this date does not have a lifespan defined, it will fail to process. To fix this, you must edit the channel configuration for that audience in SmartFeeds and set a lifespan (e.g., 540 days, which is the maximum). All new audiences created in SmartFeeds now have a default lifespan of 540 days to prevent this issue.

The audience size in the Google Ads interface doesn't match the size in SmartFeeds after an update. Why?

This is typically due to a processing delay on Google's end. After an audience is updated in SmartFeeds, it takes Google Ads between 24 and 48 hours to fully match and process the new list. The numbers you see in the Google Ads UI during this window likely still reflect the previous version of the audience. Check back after 48 hours for the updated figures.

Can I upload audiences to a Google Ads MCC (Manager) account?

Yes, this is possible and considered a good practice. You can add an MCC account directly in SmartFeeds by following the standard channel connection process. When you export an audience to an MCC, it will automatically be available in all the linked sub-accounts, provided the audience sharing feature is enabled in your MCC.

My audience is in an error state, but the message is unclear or mentions "non-execution." What does this mean?

This error can occur if the "non-execution alert" is triggered. This alert flags an audience if it hasn't been processed within a set time frame (e.g., 2 days). If your data source is only set to update once a week, this alert will naturally be triggered between updates. In this scenario, it is not a blocking error but an expected notification based on your source file's update frequency.

For further assistance with SmartFeeds, consider reviewing these articles:

For additional help, feel free to reach out via our Contact Us page.