This is why cohort analysis should be a key tool when you are measuring any marketing campaign. First, let’s define the term:
Cohort analysis is a subset of behavioral analytics that takes the data from a given eCommerce platform, web application, or online game and rather than looking at all users as one unit, it breaks them into related groups for analysis. These related groups, or cohorts, usually share common characteristics or experiences within a defined timespan. Cohort analysis allows a company to “see patterns clearly across the lifecycle of a customer (or user), rather than slicing across all customers blindly without accounting for the natural cycle that a customer undergoes.” Source: Wikipedia
At Nanigans, we’ve developed a cohort analysis tool specifically for Facebook advertising campaigns that empowers advertisers to understand and compare the value generated by ad placements and customers in the days, weeks, and months after initial ads are delivered.
For example, in the cohort analysis below, the advertiser is comparing revenue generated by weekly groupings of Facebook ad placements during the 15 days after the ad campaign went live. In this screenshot we can see the total number of purchases generated by each cohort, and easily surface details about groups of users within this ad campaign. A darker shaded cohort cell indicates better performance, and the point at which text turns from black to white indicates when these cohorts generated a positive return on ad spend.
We can see clearly that the November 25th group of users represented the highest volume of purchases, as well as the largest revenue from the included time period.
Some other key callouts from this cohort analysis, which may help you evaluate your own:
- Which was the worst performing cohort? In this example the October 7th cohort took the longest time to generate positive revenue, and generated the least amount of total revenue. We’ll talk more about how to find out why in just a minute.
- Which was the best performing cohort? As mentioned, the November 25th cohort was only one of two groups to start generating positive revenue on the third day, and has generated the most revenue of any campaigns for the dates included.
- Does anything else stand out? There are two other cohorts that stand out: the November 18th group because it took the longest to generate revenue, and the December 9th cohort because – although it’s the newest group – it began generating revenue on the third day.
Within Nanigans Ad Engine, advertisers can immediately dive into Performance Analysis to further investigate why some groups of users performed well and why others did not.
After identifying your top-performing ads (based on revenue, or any other key metric you are using to judge campaign performance), take these steps to find out why they performed differently:
- Look at overall campaign attributes. Where was this ad delivered, and was it different than previous ads? For example, if the November 18th ad was only delivered to Facebook’s right hand side (RHS), but expanded to RHS and News Feed for the November 25th group, that could be one reason why this cohort outperformed another.
- Drill down to more finite campaign attributes, such as creative (image, video, and copy), targeting, ad type, geo-location, gender, interests, and any other attributes used within your campaign.
- Once you know how ad placements performed along with these key attributes, compare them against each other by date range to see why they performed better.
In Nanigans Ad Engine, this is as easy as sorting your table and viewing the attributes you would like to see. Running this analysis not only gives you perspective on what creative, targeting, and other attributes you should use in your next campaign, but also tells you which creative did not perform well so you can be sure not to use it next time.
Want to see Cohort Analysis in action? Contact us today for a demo.Contact Us Today