/
Help Center
/ Meta
/ Meta Managed Partner Ads Faq

meta-managed-partner-ads-faq - Articles Page

Meta

Meta Managed Partner Ads FAQ

4 min read • Last updated 14 October 2024

Answers to Common Questions about Managed Partner Ads

What distinguishes Managed Partner Ads (MPA) from standard Dynamic Product Ads (DPA)?

Managed Partner Ads (MPA) represent a specialised Meta offering that allows for direct management of dynamic product advertisements on behalf of retailers, featuring SKU-level reporting. This contrasts with the traditional Dynamic Product Ads (DPA), which offer basket-level reporting and are part of Meta's standard advertisement offerings.

How are audiences formulated in MPA campaigns?

MPA leverages retailers' transaction data to construct both prospecting and retargeting audiences, enhancing the relevance and efficiency of ad targeting.

Could you explain the distinction between retargeting and prospecting audiences?

  • Retargeting Audiences
    Retargeting audiences are composed of individuals who have previously interacted with a product, aiming to re-engage past visitors.
  • Prospecting Audiences
    Prospecting audiences are created using a combination of Meta’s machine learning algorithms and the Retailer’s website pixel data to identify potential new customers who share characteristics or behaviours with the Retailer’s existing customers, but who have not yet interacted with the Retailer’s business

Is it possible to utilise custom audiences within MPA campaigns?

Absolutely! You can leverage custom audiences provided by your Retailer to enhance your MPA campaigns. This allows you to target specific customer segments more effectively, ensuring your campaign reaches the right audience and maximises impact.

How do Custom Audiences work with MPA’s Ad Sets?

  • Retargeting: Ads will be delivered to users who match both the retargeting criteria and the custom audience, sharpening your focus on specific users.
  • Prospecting: Ads will be shown to users who match either your existing targeting criteria or the custom audience, expanding your reach to new potential customers.

*Note: When selecting custom audiences, they are automatically applied to both ad sets, ensuring consistency across your campaign strategy.

Is there a risk of overspending in MPA campaigns?

Thanks to the system's safeguards, overspending in MPA campaigns has not been observed to date.

Once my campaign is live, am I able to make adjustments?

Post-launch, campaign budgets can be modified, either increased or decreased, and the campaign's end date can be adjusted. However, changes to the products advertised are discouraged, as this would necessitate a re-entry into the learning phase.

How do MPA campaigns ensure they target the appropriate audience, and what is the duration of the learning phase?

MPA campaigns typically undergo a 7-day learning phase before optimising the reach to the intended audience pools, ensuring that advertisements are shown to the most relevant users.

What implications arise from altering the products in an ongoing campaign?

Modifying the products in a live campaign is possible, yet not advised due to the necessity for the campaign to re-enter the 7-day learning phase. It's recommended to initiate a new campaign for different products.

What happens to the remaining budget if I terminate a poorly performing campaign prematurely?

The decision on how to handle the remaining budget upon early campaign cessation varies by retailer. While Zitcha charges the wallet in full upon order approval, retailers might allocate the unspent budget to alternative ad types or charge based on actual expenditure. This is determined on a case-by-case basis, according to each retailer's policy.

How is the budget allocated between the prospecting and retargeting audience sets?

The budget is evenly distributed between the two audience sets, ensuring balanced exposure across both targeting strategies.

What criteria must a supplier meet to qualify for MPA?

Suppliers are integrated into the MPA framework by mapping each brand to a brand group and assigning a vendor ID. Eligibility is achieved once Meta records 100 conversions linked to the vendor ID, facilitating targeted advertising.

Why are there products appearing in the SKU report that I did not select for advertising?

MPA campaigns generate three types of reporting pools: Promoted, Halo, and All, accounting for the appearance of non-selected products in the SKU report.

Is it feasible to set frequency capping for MPA campaigns to avoid overexposure?

The control over frequency capping in MPA campaigns resides with Meta, and currently, custom adjustments are not available to advertisers.

How many ad sets and SKUs are advisable regarding the setup of MPA campaigns?

The effectiveness of MPA campaigns increases with the number of SKUs promoted, as this enlarges the audience pools for targeting. The optimal number of ad sets will depend on the campaign's scope and objectives.

How do MPA advertisements differ from standard Meta ads, and under whose account do they run?

The execution account for MPA ads, whether from the retailer's or the supplier's handle, is contingent upon the retailer's configuration. This flexibility allows for tailored brand representation within the campaign.

What performance metrics does Zitcha provide for MPA campaigns, and how frequently are these metrics updated?

Performance metrics for MPA campaigns are updated daily within Zitcha. The last update timestamp is visible at the bottom of the Metrics tab, offering timely insights into campaign performance.

Why can't I initiate MPA directly on Meta's platform?

MPA is designed for operation through Zitcha's interface, integrating directly with Meta's systems to streamline campaign management and reporting for retailers and suppliers.

What is the recommended duration for an MPA campaign, and how should the budget be set?

Campaign duration and budgeting should be set according to the supplier's objectives and the campaign's scope, with a minimum of 7 days recommended for the learning phase.

Give us feedback
Was this article useful?