Experimentation in Marketing

At Google, experimentation has consistently been a guiding principle behind our marketing strategy. In the unprecedented and unique times we are living in, the significance of experimentation has only become more important.

Unique times = Sum (Tech Environment + Data Privacy Laws + Consumer Behavior)

The present moment is a direct result of the interaction between the current technological landscape, evolving data privacy laws, and shifting consumer behavior.

• Technological Environment: The tech landscape is constantly evolving. For instance, Chrome initiated a move towards enhanced data privacy by restricting third-party cookies by default for a small percentage of its browsers. This aligns with various data protection regulations, including GDPR, CCPA, LGPD (Brazil's General Data Protection Law), and India's DPDP (Digital Personal Data Protection).
• Data Privacy and Compliance Laws: These regulations worldwide are becoming increasingly strict, highlighting the protection of consumer data and privacy. Marketers must adapt their strategies to comply with these laws while maintaining effective results.
• Consumer Behavior: Consumer behavior is rapidly changing. Individuals are becoming more tech-savvy and expect tailored and relevant experiences. Marketers must comprehend these evolving behaviors and modify their strategies accordingly.

Marketers are faced with a critical problem: how to preserve solid brand-consumer relationships in this quickly changing environment. How can businesses ensure that their tried-and-true methods remain relevant and effective?

One powerful solution is to conduct controlled experimentation. By testing different channels, creative approaches, and levels of personalization, marketers can identify what resonates with their audience and what does not. This data-driven strategy allows for real-time adjustments, ensuring that campaigns remain engaging and impactful.

What is Marketing Experimentation ?

Marketing experimentation is a data-driven approach to identifying which marketing tactics resonate most effectively with a target audience. This involves testing different marketing strategies, messages, and campaigns to determine which ones are best suited for achieving specific marketing goals. For example, marketers may test different website layouts, A/B test email subject lines, or run social media ads with different targeting criteria to see which approaches generate the most conversions.

By running marketing experiments, you can gain valuable insights into customer behavior and preferences. This information can then be used to improve the effectiveness of your marketing efforts and ultimately drive better results.

As I had mentioned before, the Test and Learn Culture needs to be developed in every organization - which contains the same very steps.

There are various types of Marketing Experimentation techniques

• A/B Testing: This is a common method where you compare two or more versions of a marketing asset (e.g., website landing page, email subject line) against a control group to see which performs better.
• Multivariate Testing: This tests multiple variations of several elements simultaneously, like headlines, images, and layouts, to find the optimal combination. It's more complex than A/B testing but allows for a broader understanding of what resonates with your audience.
• MAB testing (Multi-Armed Bandit testing) is like having a smart testing assistant. Instead of manually dividing traffic between various variations (like A/B testing), MAB uses an algorithm to dynamically adjust traffic allocation based on real-time results. Imagine this: you have various website layouts you want to test. MAB starts by showing each layout to a small group of visitors. As it gathers data, it automatically sends more visitors to the layouts that appear to be performing better, while gradually reducing traffic to less successful ones. This helps you quickly identify the best-performing layout without having to wait for pre-defined testing periods to end.

Quick explanation of how these different tests differ

A Case Study of Marketing Experimentation

I would like to share with you a real-life example where a company called Sigma Sport in the United Kingdom used Google analytics and Google Optimize to solve this problem.

While reviewing Google Analytics data along with CRM data, the team discovered that 40% of Sigma Sport’s returning customers came through the homepage, but less than 2% of them were using the prominent homepage carousel to browse highlighted brands.

The next step was to understand the “why” behind that low 2% number.

Multiple customer journey analyses were conducted, which revealed that nearly half of all user journeys began on the Sigma Sport homepage. However, the most important aspect was that, rather than using the carousel, the majority of users preferred to use either the site search or the main navigation to browse through the brands they were interested in.

In other words, users had distinct preferences but were all receiving the same experience. Sigma Sport has 9 brands featured on the homepage, so the majority of users were having difficulty locating their favorite brand on the page.”

That gave them a new idea: If the homepage showed users the brands they had interacted with on a previous visit, Sigma Sport should see more engagement and sales.

Using Optimize 360, they created an experiment where they replaced Sigma Sport’s homepage carousel with brand-specific images of the site’s three top-performing brands: Castelli, Specialized, and Assos. Then they targeted the experiment to the audiences they had already built in Google Analytics.

They used the Analytics audience targeting feature in Optimize 360 to serve bespoke experiences to subsets of users. three distinct Analytics audiences who had earlier bought or interacted with the top three brands, then used these as targeting rules in Optimize 360. Anyone who had looked at or bought a Specialized bike in the past, for instance, now saw Specialized products in their carousel.

The experiment drove a 28% rise in revenue and a 32% increase in e-commerce conversion rate during the experiment. In fact, Sigma Sport saw uplift across the entire customer journey with a 90%+ probability to beat the baseline

Conclusion

Marketing experimentation is not just a set of techniques but a mindset that should permeate the entire organization. By embracing a Test and Learn culture, businesses can remain competitive, agile, and customer-focused in an ever-changing market. The case studies presented demonstrate the powerful impact of a well-executed experimentation strategy, leading to improved performance, customer satisfaction, and business growth.

As the digital landscape continues to evolve, so too must our approaches to understanding and engaging with our audiences. Experimentation is the key to unlocking that potential.