Can Endangered Florida Manatees Increase Your Online Conversion?
Manatees, the gentle giant of the sea, were routinely hunted pre-historically by the Taino and other Caribbean people, for everything from their hides and meat, to their bones themselves, which were ground up to treat various ailments such as asthma. Hunting of manatees was banned in 1893, and so the question of whether manatees could also be used to increase your website conversion were generally thought to be unanswerable. Until today.
A History of Manatee Interest and Affinity within Adwords
“But Sayf!” you might exclaim. “Manatee bones couldn’t possibly be used to increase your website conversion.” There, we can only assume, you would be correct. Though one could hypothesize that if someone were to offer manatee bone powder for sale online, its general rarity might, in fact, increase your website conversion. This however, is not my meaning, as I could never wish harm on such gentle creatures simply to increase a conversion metric with a short term boost in black market goal funnel completions.
I am, instead, referring to the new Demographics and Interests reports within Google Analytics.
In June of 2011, Google announced “Interest Categories”. These were a list of over 1,000 interest categories from sports, to interest in nature and wildlife such as sea cows (ie our friend the manatee, through wildlife interest categories).
This allowed advertisers to do contextual targeting on other sites. So if a user showed interest in a site talking about the differences between Amazonian Manatees (Trichechus inunguis) and the West Indian Manatee (Trichechus manatus) and the West African Manatee (Trichechus senegalensis), they might be tagged with an interest category by Google. They could then be targeted for ads even if they were on say a site about cameras, because they had previously shown interest in that wildlife category.
In September of 2012, Google added audience demographics to their Adwords tricks.
Here we could target either men or women, based on Google’s guess as to the user’s gender, based on predictive algorithms. Now we could, should we desire, place adds for our hypothetical Manatee Boat Tour Company, but only have them appear to women. Or combining with the interest categories, to women who had shown interest in other wildlife webpages, possibly focusing on the manatee itself.
Then in June of 2013, Google announced “Affinity Segments” which were an even broader attempt at creating interest cohorts from user interests.
Now by way of a new algorithm, Google could not only guess as to your gender, but as to your additional broad areas of interest. Are you a man in your 30s interested in Florida and marine mammal wildlife? You might like Manatees.
As the gentle Amazonian manatee lives solely in the fresh water of the mighty Amazon river, never entering the salty sea, these interest and affinity reports lived within Adwords, and were never seen in Google Analytics. Now however these reports have spread, much more like the familiar West Indian Manatee of North America, living both in the fresh and salt water environments of Adwords AND Google Analytics.
Much like how a group of manatees may huddle near the warm water outflows of power plants, rather than migrating south to warmer waters, analysts have hovered over the new reports, looking for the warmth of insightful data, unaware of the dangers posed in their actions. For the new demographic reports can be lifesaving, just as warm power station outflow can be to the manatee which cannot survive in water temperatures below 60°F (15°C), but they can also be misleading, or even dangerous.
The New Reports in Google Analytics
Now age, gender, affinity categories (ie affinity segments) and other categories (interest categories) are available within the Audience Reports, and for use in Advanced Segments. You can use these dimensions for analysis just like any other. How do men convert versus women? How do different age groups compare? How do people interested in wildlife rate in goal conversion?
Example Age Report
Example Gender Report
Example Affinity Categories Report
Example Other Categories Report
Eliminating Danger by De-labeling the Buckets
The important thing to understand, is that for the most part these are all guesses by Google. Other Categories, the interest categories, are based on the sites you visit. These are then crunched by an algorithm to guess your gender, age, and your overall affinity categories.
Visit a gaming site, and then a coding news site? Google might label you an 18-24 year old male, and drop you in the “Technophiles” affinity category. It doesn’t mean you ARE that old, or that Gender, or even really a technophile. This is the primary “rub”, much like a manatee might rub against your boat on a tour of certain Florida waterways, of this data. It’s all best guess by Google. It’s important to understand this conceptually, because it’s far too easy to think “this site is popular with men” or “this site is popular with women aged 35-44 who like wildlife”. It doesn’t mean either of those. It mean you might have a higher conversion rate with users who Google’s algorithm GUESSES are aged 35-44 and who they GUESS like wildlife.
If you have offline information that your primary audience is 18-24 year old men, you might not want to place ads in Adwords based on that information. We have very little way of knowing the actual accuracy of that data “in the wild”, or with your own users. It could be very accurate with your users, or depending on your market your users could be so scrambled as to make it difficult to really sketch a line between the Google defined bucket, and the offline reality.
But here’s where the power comes into play by adding this data to Google Analytics… Does it matter if it’s accurate? Does it matter if your 25-34 female wildlife enthusiasts are ACTUALLY 25-34 years old, female, or wildlife enthusiasts? What matters is THAT BUCKET, however you define it, converts higher.
What if I said “Google has used an algorithm to create cohorts of similar users. It’s done one dimension where it breaks them into two cohorts. A second dimension where it breaks them into 6 cohorts, and a third where it breaks them into over 80 additional ones.” With enough traffic you can look at these and say “Well this cohort 1.4.65 actually converts at twice the rate of any other cohort.” Would you be interested in targeting that cohort with additional adspend? I would. The labels aren’t as meaningful as they are tied to site goal conversions, where you can determine which, if any groups, actually make more sense for targeting in the Display Network.
So if you’re trying to sell more tours on your manatee watching boat, don’t focus on the fact that you mostly attract older couples over 55, who show an interest in wildlife. If you break down your numbers and are seeing that your visitors in cohort 2.2.13 are converting really well, then focus on that, even if it that cohort is made up of late twenties female hockey fans.
Using Affinity and Other Categories, as well as the Gender and Age demographics can be a powerful way to focus on the cohorts of users that engage with your content, and convert better on your website. Just be sure to treat the actual cohort labels with a grain of salt, and focus on what these similar groups of individuals do in common when it comes to on site conversion, and then have that behavior affect your ad spend.
If you still aren’t seeing these reports in your data under Audience, give it time, as Google has not completed the roll out of this functionality. Others might see instructions on how to implement these reports, which involves upgrading to the dc.js code, and enabling the reports in Google Analytics. If you have not yet done so, and are interested in seeing these reports, you should upgrade your code. These reports are not yet available through Universal Analytics.
Lastly, if you are interested in helping out the manatees, give the good people at Save The Manatee Club a visit.
About Sayf Sharif
Sayf Sharif is a Web Analyst, and expert in Usability and UX, who has worked with businesses large and small to maximize their online presence since the beginning of the Web, winning numerous awards along the way. Sayf has studied human tool use from the stone age (he went to graduate school for Archaeology) to the information age (he started programing on his father’s TRS-80), and is always interested in what goals people wish to accomplish using their tools, and how successful that experience was.