Use a feed to provide business data
About feed-driven business data
This feature is only available if you use upgraded inventory management.
Feed-driven business data combines information about products and services in a Google Merchant Center inventory account with business data that you've added to DoubleClick Search. You can apply the combined data to your search campaigns for reporting insights and to automate campaign management. When you update your inventory feed, DoubleClick Search (DS) automatically updates the data you've applied to your campaigns.
For example, you can use feed-driven business data to do any of the following:
- Adjust bids or pause advertising when availability of a product or service drops below a specific threshold.
- Update text parameters in ads so your ad copy always reflects the latest prices specified in your inventory feed.
- A hotel advertiser uses an inventory feed to provide data about hotel rooms, including prices and availability.
- The advertiser also uses business data to associate ad groups with specific hotels, types of rooms, and specific locations.
For example, one ad group contains ads and keywords for Deluxe rooms in Boston hotels, so it's associated with the "Deluxe - Boston" row from a business data table. A different ad group does the same for Standard rooms in Boston hotels, and it's associated with the "Standard - Boston" row of business data.
- The advertiser wants to raise bids if room availability is more than 50%, and pause advertising when availability is down to 10%.
- To get started, the advertiser updates the business data table with a new column called Availability %.
Then, instead of manually adding availability information in the new column, the advertiser applies a formula to the availability data from the inventory feed to provide values in the column. Each time the feed is updated, the Availability % column in the business data table is also automatically updated.
- Finally, the advertiser sets up rules that adjust bids or pause advertising in the associated ad group based on the value of the Availability % column in the business data table.
The advertiser could also create a filtered report to show locations with Availability % that's greater than 50%.
A retail advertiser uses to text parameters to dynamically change product prices in ad copy.
The prices change frequently, and they're already described in an inventory feed.
The advertiser creates a business data table with a Product and Price column. Then, instead of manually adding price information in row of the table, the advertiser uses an inventory feed to provide a value in the Price column. Each time the feed is updated, the Price column in the business data table is also automatically updated.
The advertiser applies the business data to keywords.
The advertiser sets up text parameters and uses the Price business data to provide values for the parameters.
How it works: combining inventory data with business data
Here's an overview of what you need to set up in DS to enable feed-driven business data:
- A Google Merchant Center inventory feed: provides inventory data
- An inventory plan with a business data template: selects inventory data from the feed, optionally applies formulas to transform the data, and then copies the output to a business data table
- A column in a business data table: contains inventory data copied from the feed
- A matching key: matches data from specific inventory items with a specific row in the business data table
If you haven't already done so, you'll need to set up an inventory feed in Google Merchant Center, and then link the Google Merchant Center account to your DS advertiser.
The feed sends inventory data from Google Merchant Center to DS on a regular basis. The data in the feed describes items in your inventory, using standard attributes such as product ID, name, brand, category, and so on.
A business data template selects data from an inventory feed and copies it into one or more columns in a business data table. A simple template copies one inventory attribute directly from your inventory feed into one column in a table, while more complex templates use formulas to combine data from multiple attributes and may modify the data before copying it into the business data table.
Business data templates reside in inventory plans.
An inventory plan specifies an inventory feed and optionally uses filters to focus on specific types of items within the feed. A plan can contain either templates for generating campaigns or templates for copying inventory data into a business data table.
Each inventory plan that contains business data templates specifies a single business data table that the templates copy data into. If you create multiple business data templates that need the same filtered data and copy into the same business data table, you can create all of the templates in the same plan.
The inventory plan evaluates templates periodically and automatically updates your business data with the template output.
Business data in DS resides in tables that you set up. Each column in the table represents an aspect of your business (such as a product name, location, or department name), and each row contains data that you associate with campaigns or other items in a campaign. For example, each row describes a specific retail department, which the advertiser associates with specific campaigns.
Since the data from your inventory feed may change periodically, you'll need to create a separate column in your business data table to contain data from the inventory feed. Make sure the data type you specify for the column matches the type of data from the feed. For example, if you're using an attribute from the inventory feed that contains decimal numbers, either specify the Decimal data type for your business data column or use a function in the business data template to convert the feed data into some other data type.
The ID column cannot contain data from an inventory feed. All business data tables are required to specify an ID column. The ID column can only contain static data, and its value in each row must be unique, so it isn't possible to use feed-driven business data to manage values in the ID column.
Manual columns and feed-driven columns
In addition to the two columns described in the previous section, a business data table can contain additional columns that are either maintained manually or are maintained by a business data template. For example, in the Hotels business data table below, the ID and Location columns are managed manually (that is, you add, change, or remove data in the column), while the Availability % column is managed by a business data template.
If you manually change or set a value in a feed-driven column, the inventory plan will replace the value with its own value the next time DS evaluates your inventory feed.
A business data template copies data from one or more items in an inventory feed into a specific row and column in a business data table. To determine which row to copy into, the items in the feed and the ID column of the business data row need to contain a matching value. This matching value is called a key. On the business data side, only the ID column can be used as a key. In the inventory feed, any inventory attribute or combination of attributes can be used.
In Example 1, the ID column of the business data table matches values from the Product ID attribute in the inventory feed.
In the example below, the key is constructed from a formula that combines data from the inventory feed's Product type and City attributes. As long as the output of the formula exactly matches the ID column of the business data table, the template can copy data into a row within the business data table.
No matching key
If there is no inventory item with a matching key for a row in a business data table, DS removes any value that may be in the row's feed-driven column. DS does not make changes to any manual columns within the row.
Advanced example: Calculate highest availability for each location
Instead of changing bids based on the availability of specific types of rooms, a hotel advertiser wants to take action if the availability for any type of room in a city is 50% or higher.
To enable this scenario, the advertiser does the following:
- Split the business data table into two tables: one for specific hotel products and the other for locations. Then include the location business data table in the original Hotels table.
- Now that there's a business data table whose ID column contains a city name, the advertiser uses the city name as a matching key.
When selecting data to copy into the business data table, the business data template selects from all inventory items with the same matching key. Because of this grouping feature, the advertiser can use the max function to take the highest availability value for each city.
- Because the Location table is included in the Hotels table, the highest availability will also be included in each row of the Hotels table. When availability data changes in the feed, the business data template will update the availability percent in the Location table, and that in turn will update the availability percent in the Hotels business data table.
Managing business events
You can use business data to describe frequent, recurring business events such as sales and promotions. Then you can create automated rules that change bids, pause or unpause ads and campaigns, and make other changes before and after each instance of the event.
In a business data table, you need at least three columns to define a business event:
- A column for the date on which the event starts
- A column for the date on which the event ends
- One or more columns that describe what changes during the event. Each of these description columns is called an event variable.
For example, in a "back to school" sale, you may discount some brands of shoes by 20%. So you may want one event variable column to describe the name of the sale, and another event variable column to describe the discount amount.
Feed-driven business data can provide values for the event variable columns. Any time a business data template updates an event variable column, the template always makes the following changes to start and end date columns:
- The Event start time column is always set to the current day.
- The Event end time column is set to a day many years into the future, effectively creating an event that never ends.
The date columns are automatically updated only when a business data template changes the existing value in an event variable column.
If you use feed-driven business data to manage business events, it's important that you ensure the business data templates do not change the event variable columns until you're ready for the event to start and end. For example, you may want to ensure that your inventory feed does not provide data for the event variable until the event starts.
You can also keep the business data template paused until the day that the event starts. However, when you activate the template, the next time DS evaluates templates the inventory plan updates the event value and dates only if event value is different from the value currently in the business data table.
Ready to get started?
- Create an inventory feed in Google Merchant Center and link the Google Merchant Center account to your DS advertiser.
- Create a business data table, and add a column to contain data from the inventory feed. Be sure that the data type for the new column matches the type of data you plan to copy from the inventory feed.
- Create an inventory plan and add a business data template. The template selects data from the feed and copies it into your business data table.
- Apply the business data to campaigns and other items.
- Use the business data for reporting or to automate campaign management.