The Heritage Foundation

How a target impression share bid strategy decreased ROI

Experiment ID: #16990

The Heritage Foundation

Founded in 1973, The Heritage Foundation is a research and educational institution—a think tank—whose mission is to formulate and promote conservative public policies based on the principles of free enterprise, limited government, individual freedom, traditional American values, and a strong national defense.

Experiment Summary

Ended On: 07/11/2019

Within the Google Ads platform, we decided to test a target impression share bid strategy to see if we could increase our level of competitiveness and generate more clicks and donors more effectively than with a typical impression-based bid strategy.

Research Question

Would a target impression share bid strategy increase donor conversion rate or revenue metrics?

Results

 Treatment NameClick RateRelative DifferenceConfidence
C: Control 10.8%
T1: Treatment #1 9.8%-9.0% 98.8%

This experiment has a required sample size of 7,591 in order to be valid. Since the experiment had a total sample size of 25,890, and the level of confidence is above 95% the experiment results are valid.

Flux Metrics Affected

The Flux Metrics analyze the three primary metrics that affect revenue (traffic, conversion rate, and average gift). This experiment produced the following results:

    9.0% decrease in traffic
× 0% increase in conversion rate
× 0% increase in average gift

Key Learnings

With 98.7% level of confidence, we saw a -9.0% decrease in ad clicks.

Combined with this decrease in clickthrough rate from our ads, we also saw a spike in cost per click value (+107% increase in CPC, with 100% level of confidence). 

Further, we noticed that the target impression share bid strategy (which is designed to focus on “maximizing opportunity” from the Search Network) did not increase revenue (+2.3% in revenue with a 53.4% LoC). Finally, it showed a directional decrease in donor conversion rate, as well (-17% in donor conversion rate with a 51.4% LoC).

In summary, do not use a target impression share bidding on a brand keyword strategy, as it only serves to boost your costs dramatically, with no evidence that it will increase your results.


Experiment Documented by Greg Colunga
Greg Colunga is Executive Vice President at NextAfter.

Question about experiment #16990

If you have any questions about this experiment or would like additional details not discussed above, please feel free to contact them directly.