Ads Manager and the metrics within it are important in helping you track the performance of your ads, ad sets and campaigns. To help you understand how metrics are calculated and which metrics are new or being tested, you may see metrics labelled as estimated or in development. This article provides an overview of what these metric labels mean and guidance on how to use these types of metrics for planning or decision-making for your business.

In this article:

• Identifying which metrics are estimated or in development
• Estimated metrics
• In-development metrics

Identifying which metrics are estimated or in development

When you hover over the name of a column in Ads Manager, you'll see more information with the definition of the metric represented in the column (Overview), details on how the metric is calculated, as well as recommendations on how to use the metric (Details), and any related resources for the metric (Related). Below the definition, you'll be able to see if the metric is labelled as estimated or in development. Click on the label to learn more about each type of metric.

Let's take a look at what estimated and in-development metrics are and how to use them.

Estimated metrics

Estimated metrics are calculated using sampling or modelling. These types of metrics are helpful because they can provide insights for outcomes that may be hard to precisely measure, such as the number of unique people that your campaign reached or the estimated ad recall lift of your ad. These metrics are meant to provide directional insights into the value of your marketing results and can factor into strategic planning for your business.

To help you measure the value of your advertising in near real time, we often use sampling methods to look at a portion of the data that is a good representation of the larger population included in the entire set of data. Sampling allows us to calculate metrics quickly with a high level of accuracy that may otherwise take a long time or be difficult to calculate. People-based metrics, which can help you understand your campaign's performance based on the people who use the Facebook family of apps and services, are sampled because of the large amount of data it requires to calculate these metrics quickly.

We also use sophisticated models that leverage a range of data to provide you with valuable information that you may not get elsewhere. For example, estimated ad recall lift uses a machine learning model to estimate the number of additional people who would remember seeing your ad when asked within two days. This machine learning model is trained on real polling responses from similar campaigns; it leverages high-intent signals, such as people's interactions with your ads. These models and algorithms may evolve based on factors such as changes in user behaviour or advertiser need.

Similar to metrics, breakdowns in Ads Manager are also estimated to provide dimensions, including age, gender and placement, quickly. With this information, you can gain insights into your audience and actions taken on your ads that may otherwise be hard to precisely quantify.

All estimated metrics and breakdowns in Ads Manager are reported based on your campaign's actual performance. You may also notice estimated metrics while you're first setting up a campaign, such as estimated daily reach. These estimations are different, as they're based on the potential reach or results of your ads depending on the targeting and delivery criteria that you've selected, and they don't represent actual results.

How to use estimated metrics:

• Use for directional insights
• Use to gauge advertising performance
• Factor into strategic planning and business decisions
• Compare historical campaign performance, unless we've specified that the model or sampling algorithm has changed (in which case the data may not be comparable to previous campaigns)

Bear in mind that estimated metrics may evolve as we gather more data, or as we improve our sampling or modelling. You may see a notification at the top of Ads Manager that provides additional details if there are any updates to estimated metrics.

In-development metrics

In-development metrics are still being tested and may change as we improve our methodologies. We encourage you to use these metrics for directional guidance, but recommend that you use caution when using them for historical comparisons or strategic planning.

We frequently launch new features for ads that may require new, innovative metrics that are still being tested to determine how to best offer insights on the feature. New metrics are typically in development for the first 60 days. However, certain metrics can be in development for longer while we collect information and make adjustments.

Once testing is complete and a metric's calculation is established, the metric will no longer be labelled as in development. You shouldn't expect to see a notification if there are any updates to in-development metrics.

How to use in-development metrics:

• Use for directional insights
• Use to gauge advertising performance
• Do not use for historical comparison
• Do not factor into strategic planning