"Data is everything" — that's what you'll hear from most digital marketers, especially those who work in the ecommerce realm of multichannel marketing and advanced analytics. Generally speaking, data is valuable, and it's certainly true that no brand in today's world can reach its full potential without using data to inform its marketing strategy. But is all data useful? No. That's why cohort analysis is essential.
Marketing cohorts, or cohort audiences, help brands make informed decisions about their businesses by providing specific information that enables a more accurate analysis of the motivations behind customer behavior. A cohort is a group within an audience segment united by a shared experience or trait that goes beyond simple demographics.
There is a wide range of "experiences" that can define a cohort group. Examples include major life events such as military service, marriage, divorce, having children, or moving to a new location. Others may be technical factors such as when users subscribed to emails, likes or shares on social media, or event attendance. Companies often use cohorts to present small groups of users with highly relevant and personalized content and advertising or to deepen their understanding of audience behavior and improve strategies and campaigns.
Cohort analysis is the process of reviewing data reports for helpful information about customer behavior that can be used to craft new campaigns and refine marketing strategies. It uses key traits and shared experiences to explore how different groups interact with the brand in the near term and a broad sense. Factors that are commonly used in cohort analysis include:
Acquisition dates
Specific product purchases
Event attendance
Social media engagement
Purchase time frames
Page bounces
Post-purchase behavior, including returns
Marketers analyze cohorts to isolate factors that influence business success. Cohorts can help marketers evaluate:
Customer lifetime value (CLV)
Churn rates and potential reasons for customer turnover
Factors that push customers to convert
Overall business health
Effectiveness of campaigns and reasons for success or failure
Most major marketing platforms, such as AdRoll's Growth Marketing Platform and Google Analytics, allow users to perform cohort analysis. Google's cohort analysis tool is commonly used but limits marketers to creating cohorts by acquisition dates. Other platforms may allow for more flexibility in reporting, letting users drill down to the details they need.
Before performing cohort analysis, marketers need to determine how to define their cohorts. How cohorts are created depends on the brand and marketing activity, as the information will change depending on the time periods set and the behaviors they want to analyze. For example, let's pretend that an ecommerce skincare brand hopes to learn why they are losing customers in the middle of the conversion funnel before a purchase is completed.
In this example, the company would want to split the audience into cohorts according to shared experience. In this instance, they could explore three groups: users who started on the homepage and completed a purchase, users who dropped off at the checkout page, and users who bounced off the website from a product or category page inside the digital storefront.
For the first group — customers who completed a purchase — the company could determine that the prospective customers could successfully find what they wanted and faced no issues with the checkout process. But if the marketing team wants to know why their brand is losing customers, they should focus more effort on the latter two groups.
For the second group — those consumers who bounced at the checkout page — the company could guess that they either struggled with the checkout experience or discovered something that discouraged them from completing their purchases. High shipping costs and long wait times for delivery are some possible factors that are common in ecommerce. This could prompt the company to reevaluate its shipping costs or remove back-ordered or delayed products from the storefront. These changes would help prevent consumers' frustration, which causes them to leave the brand for good.
For the third group — consumers that left while still shopping on product pages or category listings — the company might hypothesize that they could not find what they wanted. This might be a logical conclusion if this group of website visitors viewed several pages in the shop before leaving. Alternately, this cohort could have had difficulty navigating the website. If they only explored one or two pages before departing, that might be a clue.
Cohorts provide deeper insights into consumer behavior to help businesses make significant adjustments to their overall brand experiences and website functionality. Some examples might be altering the checkout process or making the website navigation more intuitive, so it's easier to find products with fewer clicks. Companies may also decide to make changes on a smaller scale that could tackle suspected problems, such as creating a discount for users who might have been discouraged by shipping charges or creating membership options that earn free shipping.
Cohort analysis can be constructive for brands hoping to refine their customer experience and create more trust with consumers. It's also a highly effective way of gleaning information that can help offer a higher degree of personalization in marketing content. Presenting users with campaigns that acknowledge their prior experience with the brand is an excellent way for companies to strengthen relationships with their audiences and help them feel seen. The more a brand understands its customers' interests and behaviors, the more personalized the experience will feel, which is the key to building brand loyalty. Cohort analysis can help ecommerce brands identify the most valuable data points, refine their marketing approaches, and make improvements that result in positive business outcomes.
Last updated on November 16th, 2021.