> ## Documentation Index
> Fetch the complete documentation index at: https://docs.northbeam.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Differences in Ad Performance

> It's very common to see a difference in ad performance within Northbeam. In fact, this is 100% expected and happens more often than not. Read this article to learn why.

**This guide will answer the following question:**

* Why is my \[platform] \[metric] in Northbeam much lower/higher than in-platform?

**Examples:**<br />“My Facebook ROAS in Northbeam is 0.8, but Facebook Ads Manager is reporting a 2.1"<br />“My Google CAC in Northbeam is \$60, but Google Ads Manager is reporting \$25"<br />“My Pinterest ROAS in Northbeam is 0.5, but Pinterest Ads Manager is reporting a 1.7"

**Ad performance discrepancies are expected. This boils down to 3 core principles:**

1. [**Attribution Model**](/docs/attribution-models)
2. [**Attribution Window**](/docs/attribution-windows)
3. **Customer Journey**

## Key Takeaways

* Northbeam’s **attribution model** and **attribution window** are different from legacy models that platforms use. This creates a difference in credit allocation and performance across your campaigns.
* Northbeam recognizes touchpoints from all channels within the **customer journey,** whereas in-platform does not. That said, even if you're using the same attribution model and window, a difference in credit allocation and performance is expected.

If there are discrepancies between in-platform and Northbeam in regards to a specific campaign or channel, this is **expected**.

Our approach to weighting credit is different from how in-platform metrics traditionally weigh credit. In short, this boils down to differences between our Multi-Touch Attribution (MTA) model and in-platform traditional models. Our MTA divides credit between touchpoints, whereas in-platform models tend to give full credit to a single touchpoint (usually on a last-touch basis).

## Example

Let's say somebody placed an order for \$100. Below is their customer journey:

* User clicks on FB Campaign called "Prospecting"
* User clicks on FB Campaign called "Retargeting"
* User clicks on Klaviyo Email
* User clicks on Branded Search ad and purchases

See the credit allocation for each touchpoint across a few different attribution models.

| | Clicks-Only (Northbeam's model) | Last Touch (Northbeam's model) | Last Touch (Facebook's model) |
| :- | :- | :- | :- |
| FB "Prospecting" | \$50 | \$0 | \$0 |
| FB "Retargeting" | \$50 | \$0 | \$100 |
| Klaviyo Email | \$0 | \$0 | \$0 |
| Branded Search | \$0 | \$100 | \$0 |

Here are the biggest differences:

* Our MTA models (Clicks-Only and Clicks + Modeled Views) divide credit among all touchpoints
* The Northbeam Pixel recognizes touchpoints across the full journey (from all channels) - hence the difference in Last Touch

Other differences that aren't depicted above are:

* Attribution Windows
* Accounting Modes

All of these create differences in performance when comparing Northbeam to legacy platform reporting. We believe this is a more accurate way to make media buying decisions.

**See the article below for more information and scenarios. The article speaks specifically to Facebook, but the concept can be applied to all platforms.**

[“What are the differences between Northbeam and in-platform reporting”](/docs/what-is-northbeam#why-do-my-in-platform-metrics-look-different-from-northbeam)

## Need additional help?

If you need further assistance, reach out to [Support](https://www.northbeam.io/submit-a-support-ticket)!