As you may have learned last month, Boost is up to some exciting things! Previously, we discussed using lexical analysis as a tool to identify words or phrases that have a positive or negative impact on your results. This month we will address hypothesis testing and putting the lexical analysis to work.
A hypothesis test in simplest terms is an A/B test to optimize the performance of one element against another. We define an element as one word or exact phrase. Through hypothesis testing we are able to work toward our end goal of identifying the best messaging in each ad group or category.
The example above shows a lexical analysis of an account. Our Analytics team identified two price-related elements, “$19.99” and “50% Off,” that we will test against each other across multiple ad groups. Our platform inserts “$19.99” and “50% Off” into the exact same ad groups. The headline, description and display URL will remain the same; we will only change the two price point variables.
During testing, we evaluate multiple metrics to determine a winner. Overall, we are looking for a price point that is statistically significant at driving more clicks and conversions at a more cost effective cost-per-acquisition. Based on our analysis, we concluded with a 99.8% confidence rate that using “$19.99” in copy instead of “50% Off” will result in a better conversion rate.
About Boost Media
Boost Media increases advertiser profitability by using a combination of humans and a proprietary software platform to drive increased ad relevance at scale.
The Boost marketplace comprises over 1,000 expert copywriters and image optimizers who compete to provide a diverse array of perspectives. Boost’s proprietary software identifies opportunities for creative optimization and drives performance using a combination of workflow tools and algorithms. Headquartered in San Francisco, the Boost Media optimization platform provides fresh, performance-driven creative in 12 localized languages worldwide.