How DoubleClick Search tracks website conversions
DoubleClick Search reports conversions tracked by Floodlight, and can be set up to report on conversions tracked by other systems such as AdWords Conversion Tracking and Google Analytics. To report conversions tracked by AdWords and other external systems, DoubleClick Search periodically imports conversion data collected by the external system. Each external system uses its own methods for collecting conversion data. DoubleClick Search simply reports the data that these systems have collected.
To report conversions tracked by Floodlight, DoubleClick Search relies on Floodlight tags and DoubleClick cookies.
DoubleClick sends a cookie to the browser after any impression, click, or other activity that results in a call to the DoubleClick server. If the browser accepts the cookie, the cookie is stored on the browser.
Most commonly, DoubleClick sends a cookie to the browser when a user clicks a search ad managed by DoubleClick Search or visits a page that shows DoubleClick ads. Pages with DoubleClick ads include ad tags that instruct browsers to request ad content from the DoubleClick ad server. When the server delivers the ad content, it also sends a cookie. But a page doesn’t have to show DoubleClick ads for this to happen; it just needs to include DoubleClick ad tags, which might load a click tracker or impression pixel instead.
What’s changing in DoubleClick Search conversion tracking
In some cases, conversions are hard to measure, such as when the website cookies that store information about your ad clicks aren't available due to factors including browser settings. In these cases, Floodlight uses scaling to account for the conversion data that can't be directly measured. That is, Floodlight uses machine learning and historical data to model the number of conversions and amount of conversion revenue that cannot be measured. Floodlight can then add the resulting estimate to its conversion metrics to provide a more complete picture of how your advertising drives conversions.
By default, all Floodlight columns include scaled conversions, but we will provide the ability to exclude scaled conversions.
We may make additional upgrades in the coming months in alignment with other Google products to further improve measurement.
You may find some discrepancies between conversion counts in DoubleClick Search and other DoubleClick products, but these discrepancies are expected to diminish over time as upgrades are released.
When we are able to store information in cookies, conversion measurements will remain unchanged.
When we apply statistical modeling to account for conversions that cannot be measured due to factors such as browser settings, the result is often a fractional number.
To make DoubleClick Search reporting easier to understand, we round these fractions in some locations within the UI, in executive reports, in downloaded reports, and in the reporting API. This rounding can cause discrepancies between reporting columns, data cards, and other places in reports, depending on how each part of the report rounds and sums the fractional data.
Here are some examples of discrepancies due to rounding:
- Conversion totals in the UI compared with totals in downloaded reports
- Conversion totals compared with calculated metrics such as "Cost per action" or formula columns
- Conversion totals in unsegmented reports compared with segmented reports
- Conversion totals on different tabs, such as the Campaigns tab compared with the Ad groups tab
These discrepancies are expected to diminish over time as upgrades to further improve measurement are released in alignment with other Google products. You may need to update the conversion-tracking tags on your site to benefit from some of these upgrades.
Impact to bidding
We recommend that you monitor your conversion reports over the next few months and consider making any necessary changes to your bids. If you’re using Smart Bidding strategies and you notice a decrease in traffic, you may want to adjust your targets during this time to compensate for the underreported conversions. If you’re setting bids manually, keep in mind that some conversions might not be reported when you’re evaluating performance and adjusting bids.
Bid strategies are not impacted by rounding discrepancies because they use the raw fractional numbers that result from applying statistical modeling to conversion data.