What are the Bidding Revenue Models?

Table of Contents

1. Introduction
2. Cost Per Lead
3. Performance Marketer
4. E-Commerce
5. Specify by Conversion Type

Introduction

Our application supports multiple revenue models for search marketer including cost per lead, performance marketers, e-commerce, and specify by conversion type.

Cost Per Lead

Designed for lead generation search marketers who want to generate leads/conversion at a specific Cost Per Lead (CPL). A target Cost Per Lead (CPL) is set. This is the amount the search marketer is willing to pay for a conversion/sale/signup. A Max. CPL is the maximum amount a search marketer is willing to pay for a conversion when the calculated bid is below publisher reported first page bid. Setting Max. CPL to a value greater than target CPL will help maintain traffic on keywords with bids less than minimum first page bid.

Example: A search marketer wants to maximize conversions for $10 per conversion. Click “Cost Per Lead” and then under constraints, enter $10. The bidding algorithm will then set the bids to maximize revenue by spending $10 per conversion.

Performance Marketer

Used by performance marketers who have a value per conversion, and a profit margin goal as a percentage of the value. A fixed value per conversion can be entered, or included if the actual conversion value varies. The target margin is set as a percentage of the conversion value. A Min. Margin % is the minimum profit margin acceptable when the calculated bid is below publisher reported first page bid. Setting Min. Margin % to a value less than the target Margin % will help maintain traffic on keywords with bids less than minimum first page bid. Alternatively, target Margin % and Min. Margin % can be specified as a percentage or ratio of search spend, also known as ROI (return on investment) or ROAS (return on ad spend)

Example: A performance marketer generates leads for mortgage companies. Leads are worth $20-150 depending on location, property type, and credit. The target is to maximize lead volume while maintaining a margin of 30% of conversion value. In this case, the conversion value is then set to “auto” and the lead value used for bidding is based on the actual value per conversion provided via revenue tracking. If the average value per conversion for a keyword is $40, with a target margin of 30%, the conversion value would be:

$40 * (100-30)% = $28.

E-Commerce

Designed for retailers using a Cost of Goods Sold (COGS) model corresponding to revenue. The search marketer enters the estimated gross margin as a percentage of revenue, where gross margin = (Revenue – COGS)/Revenue. A target margin is then entered as a percentage of gross margin, or as a percentage of search spend.

Example: An E-Commerce retailer sells jackets with a gross margin of approximately 60%;a jacket that sells for $200 costs the retailer $80, leaving a gross margin of $120. If the retailer targets a net margin of 30% of gross margin, bidding would be set to spend $84 per conversion ($120 * (1-30%) = $84). This would leave $120-$84 = $36 net profit, which is 30% of $120 in gross margin.

Specify by Conversion Type

With multiple conversion types, you can improve your bids by basing them on them on the type(s) of conversion each keyword is driving. By specifying a value per conversion type, the bids will be based on the actual value each keyword is generating, rather than a blended average.

Example: For our fictional travel company, assume they receive $50 per Air reservation, 5% of the total Hotel reservation revenue and $10 per Car reservation.

The calculated bid is based on each keyword’s total value per conversion and the Target Margin % specified by the user, which is 10% in this example. Our bidding algorithm will calculate a bid designed to produce 10% profit margin for the keywords.

Best Practices for Bidding with Channel Connect

Table of Contents

1. Introduction
2. Folder Structure on Channel Connect
3. Folder Settings on Channel Connect
4. Limitations with Bidding on Channel Connect

Introduction

This article will explain our recommended approach to bidding when you are using the Channel Connect feature.

Folder Structure on Channel Connect

The approach for structuring folders in Channel Connect is the same as it would be for any other channel. Please refer to the guides provided here:

• A Guide for Structuring Folders
• Bidding Walkthrough

Although data will not be shared across channels, we still recommend not having Channel Connect and search mixed in one folder to help simplify management as some rules that apply to search do not apply to Channel Connect. We also recommend splitting Channel Connect ad groups of different channels (e.g. Search vs. Display) in separate folders too. This is because different channels tend to perform differently. See the Limitations with Bidding on Channel Connect section below for more info

Folder Settings on Channel Connect

The following folder settings are available when using Channel Connect:

• Bid Calculations: Since there is no API with Channel Connect, no bids will actually be pushed to the publisher. Therefore Preview and Traffic modes are equivalent; we suggest setting to Traffic if you’re actively leveraging the platform’s bid calculations at the publisher.
• Revenue Model: Set exactly as you would for any other channel.
• Optimization Criteria: Set exactly as you would for any other channel.
• Boost: There is no universal boost recommendation on Channel Connect to account for headroom as every channel/publisher is different. For example, some auction models are CPC based and therefore there is no concept of headroom, and therefore the boost should be set to 0%.
• Specify by publisher and match type: This setting is redundant and should not be used. All entities are implicitly labeled as broad match on Channel Connect.
• Bid Cap: Set exactly as you would for any other channel.
• Maximum Bid Change: If Search Bid is available and accurate, set exactly as you would for any other channel. This requires the cost file to contain the Max CPC column. If that column is not provided, this setting cannot be used.
• Minimum Bid Change: Set exactly as you would for any other channel, it defaults to blank.
• Position Cap: If position is available, set exactly as you would for any other channel. This requires the cost file to contain the Avg Position column. If that column is not provided, this setting cannot be used.
• Always keep above bid for first page / Minimum Bid: this setting is redundant and should not be used. The Minimum Bid field on Channel Connect is left blank intentionally.
• Bid Floor: Set exactly as you would for any other channel, it defaults to blank.

Limitations with Bidding on Channel Connect

Whilst the majority of functionality across the platform is universal and works identically on Channel Connect, there are some known limitations and considerations:

• Folder Bid Preview and Spend Breakdown pie charts are not available for Channel Connect. These may be added as enhancements in a future release.
• Portfolio optimization (including click and profit maximization, and spend targets) is not available for Channel Connect as these rely on position models; position isn’t a required data field on Channel Connect as it often has no meaningful equivalent concept.
• Position Bidding (including position overrides) are available on Channel Connect assuming Avg Pos is in the cost files.
• If position is unavailable, the folder Position Cap setting should not be used.
• If Search Bid is not set/not available, the Maximum Bid Change setting (a.k.a. ‘bid damping’) should not be used.
• If a process isn’t setup to update Search Bid when bids are updated at the Channel Connect publisher, Search Bid in the platform will become out of sync with the actual publisher bid, and therefore bid damping is not recommended.
• Entity settings such as position and Search Bid and other dimensions will only be updated if the entity has at least one click in the cost upload. Note: Avg Position requires one impression, Avg CPC requires on click but Search Bid as a setting can be uploaded without any clicks.
• Keyword and group entities cannot be created in a standard structure upload and instead have to be done via a cost upload; the specifics of this depend on the specific integration and can be different for every client.

Best Practices for Structuring Marin Bidding Folders

Table of Contents

1. Introduction
2. Getting Started
   1. Important Note
3. Best Practices for Search folder
   1. Brand vs. Non-Brand
   2. Segmenting Non-brand Search Folders
4. Basic Non-Brand Search Folder Best Practices

Introduction

A Folder is a collection of Google Adwords Campaign Management or Groups that represent a bidding strategy. Folders should be comprised of Campaigns or Groups that have a common goal like a CPL Target, or share a budget. Keywords without much historical data will borrow data from similar keywords within the same Folder.

Getting Started

Automatic bidding optimization takes place within folders, not between folders. We therefore recommend you to consolidate your performance goals and start your planning from one folder per channel to retrieve maximum benefit. A typical initial folder structure will contain:

• Search folder: All Search Publisher Campaigns.
• Shopping folder: All Bing PLA and Google Shopping Campaigns.
• Display folder: All Display creative groups and placements.

You can read about how to make new folders in Creating a New Folder.

Important Note

Marin bidding is not compatible with Google Conversion Optimizer or some Google Flexible Bidding Strategies. In these cases Marin’s bids will be overwritten by the Google-determined bid and Google will not report the actual bid used in the auction. Contact your Platform representative to successfully launch Marin Bidding for campaigns currently opted into Google Conversion Optimizer or Flexible Bidding Strategies.

Best Practices for Search folder

Brand vs. Non-Brand

Many marketers utilize Marin’s Position-Based Bidding Algorithm to maintain an average position for Brand Campaigns and Groups.  Create separate Brand and Non-Brand search folders to use position-bidding on Brand terms and Financial bidding for Non-Brand terms.

Segmenting Non-brand Search Folders

For advertisers with one consistent goal across all Non-Brand Search Campaigns and Groups Marin PPC Management Service recommends keeping all Non-Brand Search Campaigns and Groups in one Folder.

If clients have varying business goals or products, then it is possible to create separate folders if each segment averages more than 40 conversions per month. More conversions per folder improve performance since the statistical confidence of the bid calculations is increased with the amount of sample data.

Basic Non-Brand Search Folder Best Practices

If each segment will maintain 40 conversions, consider the following approaches:

• Segment RLSA and Non-RLSA into separate folders.
• Segment by bidding goal, including position bidding, conversion (lead)-based bidding or revenue-based bidding.
• For advertisers with a large portion of spend on low-volume keywords, consider segmenting by product lines.
• Marin generally recommends against segmenting by publisher.
• Low-Volume Groups: To reduce risk from volatile bids on extremely low-volume Campaigns and Groups, consider mapping Groups with fewer than 10 clicks all-time to an Exploratory Spend folderand set to Off. As Groups accrue clicks and budgets permit, move them to the appropriate bidding folder.

Campaign Mobile Bid Adjustment Recommendation Exclusions

Table of Contents

1. Introduction
2. How to set up mobile bid adjustment exclusions
3. Important note
4. Additional information
5. Useful links

Introduction

Mobile Bid Adjustment allows you to dynamically alter the amount you’d like to bid for mobile devices.

With the Campaign Mobile Bid Adjustment Recommendation Exclusions feature for Google and Yahoo! Japan, you can choose to exclude specific campaigns from the platform’s recommended adjustments. For more information about the way the platform calculates mobile bid adjustment recommendations, please read our Help article.

The exclusion feature gives you even more flexibility and control over your ad spend and mobile bidding strategy, and comes in handy in a number of different scenarios. For example, you could use it to completely switch off mobile traffic to a specific campaign by excluding it from the platform’s bid adjustment recommendations and setting a -100% mobile bid adjustment.

How to set up mobile bid adjustment exclusions

The exclusion can be set to on or off via either single edit, multi-edit, or bulk upload for any Google/Yahoo! Japan campaign. When a campaign’s override is set to on and the campaign-level bid adjustments are set to Traffic, the platform will not update mobile bid adjustment for the campaign.
To set up the mobile bid adjustment exclusion in single edit:

1. Go to the main Campaigns tab.
2. Click the name of any of your campaigns.
3. Navigate to the Settings sub-tab.
4. Use the Mobile Bid Audit Adjustment Recommendation Exclusions radio buttons to enable the feature.

 

To set up the mobile bid adjustment exclusion in multi-edit:

1. Go to the main Campaigns tab.
2. Use the checkboxes in the grid, select any one of your campaigns.
3. Click Edit above the grid (yellow ‘pencil’ button).
4. In the bubble, choose Mobile Bid Adj. Exclusion from the drop-down menu, then select On.

Important note

Group mobile bid adjustments will always trump the campaign mobile bid adjustment setting. If you are using campaign mobile bid adjustment recommendation exclusions, please ensure that no groups within that campaign have a mobile bid adjustment set and are not configured to receive mobile bid adjustment recommendations.

Additional information

• Group level recommendations, if enabled, will still be calculated for groups within a campaign, even if it has Mobile Bid Adjustment Recommendation Exclusions enabled.
• The recommended mobile bid adjustment will still be calculated and can be viewed in the Campaigns and Campaign Devices grids.
• You can view the status (on or off) of this setting within the main Campaigns grid.
• Historical data for Mobile Bid Adjustments is not currently available.

How to Reduce the Time Taken by Long-Running Bidding Jobs

Table of Contents

1. Introduction
2. Recommended solutions

Introduction

There are limitations on how many bid changes can be pushed to the various publishers in any one bidding cycle.

This largely depends on the number of keywords and groups in the client account, but can also depend on other background tasks occurring in the Platform, and even on publisher API’s. Any long-running bidding job taking more than 12 hours should be investigated. We are actively monitoring this to increase capacity longer term, but if you suspect this is impacting your account then please file a ticket through the help section.

It should be noted that there are two distinct phases to a bidding cycle: (i) calculating the bids (typically very quick) and (ii) pushing these changes to the publisher (dependent on other queued/scheduled actions and the publisher APIs); usually it is the second phase which takes the longest.

We generally recommend limiting Marin client accounts to less than 2 million keywords and less than 200,000 groups for optimal results.

Recommended solutions

If long-running bidding jobs are impacting your account, we recommend the following solutions, in order:

1. Exploratory Spend Folder: Put majority of Google Adwords Expert AdGroups (e.g. 0 clicks all time, or <10 clicks all time) into an “Exploratory Spend” Folder which is not to be trafficked (set to “Off”). As budget becomes available, add some of these Groups to trafficked Folders. Leave better performing Groups in bidding Folders and move poorer performing Groups back to Exploratory Spend Folder. Repeat as budget becomes available.
2. Stagger Boost & Target Changes: When you change the Target CPL/Margin/ROI/ROAS and Boost settings, the calculated bids change for all groups and keywords in a Folder. By staggering these changes per Folder throughout the week, you can spread the number of bid changes across multiple days reducing the daily impact.
3. Activate/Increase Folder Minimum Bid Change: It is possible to only push new bids if they are sufficiently different from the current bid; which can drastically reduce the number of bid changes pushed to the publisher on a daily basis. Consider setting to $0.02 or $0.05 (or the equivalent in the local currency), and continue to increase as needed.
4. Bid Override: Put majority of Keywords (e.g., <5 clicks all time) on Bid Override. If these keywords are in a trafficked Folder, we will still calculate a bid, but we will not post a change with the publisher.