Controlled experiments are the gold standard for measuring the effectiveness of social marketing campaigns and platforms.
To prove what works and what doesn’t in social marketing campaigns and platforms, consider how British epidemiologist Austin Bradford Hill fought tuberculosis in 1948.
Hill was determined to verify and demonstrate which medications and treatments worked and which were a waste of time to develop because they didn’t achieve the desired effect. He randomly assigned patients to both test and control groups — creating one of the first rigorous examples of randomized controlled trials (RCTs), which proved causality of various treatments.
Controlled experiments are often the most affordable and conclusive measurement option
Today, the RCT methodology is the gold standard for measuring the effectiveness of social media marketing campaigns as well as medical treatments.
Constant change and lack of sound alternatives for measurement and metrics has meant that social marketers have relied on engagement metrics such as likes and shares to justify their efforts. These metrics, while useful in understanding content performance, often lead marketers to chase audiences who aren’t likely to convert or encourage marketers to “game” content for social media interactions that fail to increase consideration, advocacy or sales.
As social media matures as a marketing playbook staple, marketers must shift focus from community growth to revenue growth and prove that their investments can deliver results.
Design campaigns with experiments in mind
For those who seek proof, there is no methodology like experimental design. Many of the major social networks offer measurement solutions that employ causal experiments. The experiments either use surveys to assess recall or intent or they match exposure to marketing messages to actual purchase data to measure sales lift.
Benefits of controlled experiments
- Hypothesis-driven. Experimental design brings goal-setting to the front of the process. This is a major plus in a social marketing world where hypothesis development and goal setting often take a back seat to the excitement of a novel execution.
- Gold standard of measurement. Just as Hill raised the bar for scientific rigor, a valid, controlled social marketing experiment will tell you whether marketing on a specific platform creates incremental value, all other factors being equal. Other measurement methods are useful, but still depend on assumptions and correlations.
- Simple to understand. Controlled experiments are transparent and easy to understand. Marketers and executive stakeholders don’t need a Ph.D. to understand the results (although an understanding of statistics is required to set up an experiment).
Experiments are, however, limited to paid social campaigns because they must access and track statistically significant and similar control groups. Experiment results are also applicable only to the social platform used for testing (e.g., a Twitter test applies to Twitter and not all social platforms). Despite these limitations, controlled experiments are still a compelling measurement option for social marketers in an environment of increasing data volumes and measurement complexity.
Just as medical professionals need to prove a particular treatment improves a patient’s health, today’s marketing leaders must embrace new metrics and build new competencies in various analytics methodologies to demonstrate the value of social marketing.
Gartner for Marketers clients can read more in Embrace a New Approach for Social Marketing Measurement by Jay Wilson and Christi Eubanks.