A campaign's learning phase is generally a few days, as it takes a few days for the Amobee DSP to gain statistically significant insights for real-time optimizations. During this learning phase, new sites are delivered on and new users are reached.

For additional tips, refer to the Campaign Performance, Optimization & Delivery Best Practices.

A line item's learning period depends on the following factors:

• — The speed of accumulated impressions (this depends on budget, bid price, and targeting criteria). The larger the number of impressions accumulated for a line item, the more a line item has learned.
• — Other similar line items. If there are other line items with similar settings that have been live in the past 30 days on the Amobee DSP, then the historical data of those similar line items can increase the learning period of Amobee's machine learning model.

Format refers to the actual creative format of the displayed ad, whereas environment is where an ad will be displayed. Available formats include display, video, and audio. Available environments are desktop web, mobile web, mobile app, CTV, and DOOH. Each format has a default set of environments based on the most common use cases. The format/environment mappings are as follows:

• — Display includes desktop web, mobile web, mobile app, and DOOH.
• — Video includes desktop web, mobile web, mobile app, CTV app, and DOOH.
• — Audio includes desktop web, mobile web, and mobile app.

Cross-channel implies reaching a user via multiple ad formats, e.g., reaching the same user with both banner and video formats. Cross-device means reaching the same user on multiple devices, such as a laptop and a mobile phone. Campaigns can be set up for both cross-channel and cross-device delivery when targeting multiple formats and devices.

Yes, the in-platform notes feature is available via the Flyout panel at IO/package/LI levels and can be used to collaborate on campaigns. Team members on a campaign can also see the notes history on any of the campaign objects (IO/package/LI). See Using the Flyout Panel.

Yes, the Amobee DSP includes bulk edit functionality.

The bulk editing feature enables group configurations of budgets, bid prices, frequency rules, pacing, inventory sources, and more. You can edit hundreds of packages and/or line items across all channels in a single in-platform workflow. This is very advantageous for large multinational brands that have a diverse set of products, territories, and a multi-level business structure. For example, if you select bulk edit on the bid column, you can add, subtract, or replace a certain monetary value across all of your line items.

Power Edit is a bulk editing tool for managing insertion orders (IOs), packages, line items, ads, beacons, and creatives in a single Microsoft Excel file. With Power Edit, you can make multiple changes at once in a spreadsheet. Since the Excel file is saved outside of the Amobee DSP, there are no concerns about system timeouts and loss of unsaved work.

Bulk editing allows you to be more efficient with campaign optimizations by making quick edits to multiple packages and/or line items simultaneously. You can bulk edit up to 100 line items.

If you want to make edits across a single object (e.g., several packages in an IO), you should use bulk edit. If you want to make updates across multiple IOs, then use Power Edit. Power Edit can also be great for bulk adding Geo or Brand Intelligence targets to several line items in an IO.

Yes, creatives can be bulk uploaded for ads served through the Amobee DSP. For ad servers such as Doubleclick, Sizmek, and DoubleVerify wrapped assets, Amobee's drag-and-drop functionality offers a seamless way to upload and test multiple creatives together, without having to reformat your tag sheet. Creatives can also be bulk uploaded using Power Edit, a functionality that allows you to create, delete, and update any campaign object on the Amobee DSP. Power Edit can be used to bulk upload creatives regardless of where they are saved.

When copying a package or a line item, the last 30 days of learnings are carried over.

Since the strongest signals on learnings are kept between the line item to the creative, it is recommended that you copy a package or a line item to retain as much learnings as possible, which helps to start them with the maximum amount of historical learnings.

As you progress higher in the hierarchy of a campaign from line item to IO, the retention of learnings is not as strong.

Exclusion lists override inclusion lists. When a site is on both list types, it will be blocked.

On the Amobee DSP, inclusion and exclusion lists are computed together as a logical “AND”. Both criteria need to pass to qualify for ad serving. If a site moves from an inclusion list to an exclusion list, updates to an inclusion list are not required. However, if a site moves from an exclusion list to an inclusion list, then both lists need to be updated. When moving from one list to another, if the campaign is live, then results take time to propagate.

Amobee supports location-based targeting at multiple levels, including: continents, countries, DMAs, regions, cities, global postal codes, US zip codes, and live location groups.

When available, Amobee uses latitude/longitude coordinates from a bid request for location targeting. When latitude/longitude information is not available, Amobee uses exchange data and standard IP address-based targeting.

Amobee leverages one of the industry leading technologies, Digital Envoy, to ensure that our targeting is as accurate as possible. Digital Envoy is also one of the providers that IAS leverages for geo-verification, implying that Amobee's geo-filtering aligns well with IAS's in-geo verification.

When deriving location, Amobee uses the following approach, depending on the data available in each bid request:

• — Latitude/longitude coordinates
• — Non mobile carrier full IP Address
• — Exchange provided location IDs
• — Exchange declared country, region, DMA, city, zip
• — Full IP Address
• — Partial IP Address

Yes. The Amobee DSP supports bid modifiers at the campaign level, and these modifiers can adjust all bids for a campaign based on a number of parameters, including: site, app, daypart, device, inventory source, first-, second-, and third-party data segments, and location.

Amobee’s Bid Multipliers tool allows you to override bidding algorithms by manually increasing or decreasing specific targeting parameters. It offers the flexibility of letting the user choose what device types, sites, exchanges, times of day, and audiences should be bid on more aggressively to drive performance. It is also transparent, showing what dimensions are driving campaign performance based on historical data.

Yes, the Amobee DSP supports goal stacking of up to three KPIs for campaigns with multiple media objectives. The platform also supports goal prioritization, i.e, supporting dozens of potential campaign goal types. As an example, you can set up a video campaign to optimize towards both viewability rate and Cost Per Completed Views.

The Amobee DSP automatically analyzes over 2,500 variables and incorporates them into a custom optimization model for each campaign.

Goal stacking avoids the need to split budgets or build multiple line items by each KPI. This reduces the need for multiple line items and manual optimizations on your end.

For additional tips, refer to the Campaign Performance, Optimization & Delivery Best Practices.

Supported KPIs include::

• — CPA
• — CPL (Lead)
• — Engagement
• — ROI%
• — CPM (fixed cost)
• — CPC
• — CTR
• — Cost per Install (mobile)
• — Click to call (mobile)
• — In-view %
• — Viewable CPM
• — In-demo reach via Nielsen
• — CPCV (video)
• — 100% completion goal (video)

Yes, the Amobee DSP supports conversion and custom event tracking. Since the Amobee DSP includes a full, natively-integrated DMP, our pixel capabilities are extensive and highly customizable.

Look-back window is fully customizable during pixel creation. These pixels can be targeted by packages and line items in the platform for retargeting using a boolean workflow to customize frequency, recency, & velocity (targeting a user based on how their engagement changes over time).

There are two primary categories of pixels on the Amobee DSP::

• — Campaign-focused pixels that are easy to create, capture a limited amount of data, and are built for retargeting and action tracking. Such pixels can be image-only or Javascript tags, and can be added to a campaign object (e.g., package or line item) for action tracking. These can also be configured to track and/or optimize based on cost to action or ROI (if a shopping cart value is passed). These pixels support a maximum of two macros for passing custom values: cid: customer and ID bprice: shopping cart value (must be a number)
• — DMP-focused pixels built to store a variety of data types for categorization, analysis, segmentation, and activation. These pixels allow Amobee to support more complex data asks that may involve a greater number of variables.

Amobee can optimize to offline sales and foot traffic. With the Amobee DSP, you have access to in-flight sales optimization/measurement verified by partners such as ODC, NCS, IRI, iBotta and Cubeiq. Amobee accepts all third-party attribution pixels and can work with you to integrate automated data feeds with your preferred verification partners.

Creative sizes can go higher in some cases, but we recommend communicating with your publisher for approval of higher creative sizes prior to launch.

A frequency cap limits impression(s) for an ad served to a user to x number of times. Campaign managers can set up the maximum number of times that a user can view an impression within a certain interval for a campaign. This can be configured at the IO/package/LI levels. When the Amobee DSP delivers more impressions to a user than what you have specified for a campaign, it is considered a frequency cap violation. Sometimes this limit is not maintained. If a campaign delivers more than 5% of specified impressions, we will investigate it further.