John loves your app. As a 27 year old gamer from Ohio with plenty of free time, John may not commit to an app quickly, but when he does find one that he likes he uses it daily.
Jane is unimpressed. She's an 18 year old college student from New York. She seems to like its potential and used it for a few weeks, but after a month there was a slow and steady decline and once she was done she was done for good.
Jimmy is a 49 year old trucker from California. He could not be less interested. He tried it for a day and left. He prefers to spend his time browsing YouTube for conspiracy theories.
John, Jane, and Jimmy all had very different experiences with your app or website. But they’re also not the only ones. There are potentially thousands of Johns and Janes, and while you hope there are very few Jimmys, chances are there are thousands of those as well.
Begin the Process of Understanding Your Customers with User Segmentation
Most businesses use some form of analytics to learn more about their company’s performance. They’ll log into a tool like Google Analytics or the iOS analytics tool, review their conversion tracking or bounce rate, and make changes to try to improve on those numbers.
But there’s a problem with this approach to analytics:
Aggregated data is, in most ways, useless.
Well “useless” may be a strong word, but its usefulness is limited. Looking at data from an aggregate level tells you nothing about Jimmy, Jane, or John as individuals, nor does it tell you anything about the users like them. It hides critical information while giving you minimal data to guide improvements.
The solution may be user segmentation.
What is User Segmentation?
User segmentation is the process of categorizing users into segments that match some characteristic about them, whether it’s:
- Tech Use and More
Instead of looking at all of the data of your users combined, you can split it into groups, like “males and females” or “married vs unmarried” or “millennials vs. boomers.” It can also be behavior related, such as “used the recommended feature vs. didn’t” or “abandoned shopping cart vs. made purchase.”
There can be more than two segments to a group, but it helps that they all have some type of defining characteristic that is different between the groups.
Why Segment Users?
To understand why user segmentation is so important, we have to take a trip back to middle school math class. There, we were taught that “average,” while easy to calculate, wasn’t necessarily that useful for understanding the specifics about large groups.
- 10 people that bought 2 items each have an average of 2 items purchased per person.
- 9 people that bought 0 items and one person that bought 20 have an average of 2 items purchased per person.
In the former example, everyone is buying your products, and so your next step may be how to find new customers, or how to increase the number of items purchased, etc. In the latter, you want to figure out who made that 20 item purchase and find more of them.
But if you’re looking at aggregate data (averages), you don’t get to know which is which. Sure, you can drilldown and figure it out in small doses, but how do you figure out the characteristics of your ideal customer? How do you determine their buying behaviors? How do you determine what you need to change to reduce dropout from the Janes and increase purchase power of the Johns?
Segmentation is what makes this possible. By splitting them into segments – for example, Janes and Johns, or recurring vs churning, etc. – you can more easily analyze their behavior, look for trends, see what campaigns work, and so much more, all in a way that give you real, tangible data that you can use to draw actionable conclusions.
How to Determine What Users to Segment
Sometimes, user segmentation will be intuitive. For example, it makes sense logically to want to look at differences between new users and recurring users. Those are two completely different groups, and within those groups you can potentially find a lot of interesting and useful information that may tell you something about your company’s future.
But there may also be segments you have yet to discover, or segments that you can imagine but may not yet realize how useful those segments may be. That is why in order to determine who and what to segment, the best way to do so is with the following:
Collect as Much Data as Possible
Choose a SaaS analytics tool and begin collecting data. It’s important to begin by collecting as much data as possible on your users, from engagement data to demographic data to behavioral data.
Good analytics conversion tracking software should be able to integrate easily with your app or website and start collecting mountains of data that you can use not only to figure out what groups to segment, but also what information those segments tell you.
Some website analytics platforms come with their own automated user segmentation system to make collecting data on different segments easier. The goal is to make sure that you have the data needed to begin determining which segments are available within the users on your website.
Focus First on Segments That Are Profitable
The purpose of data analytics (and thus ultimately, of segments) is to improve conversions and reach some type of revenue-related goal. The value of segments is going to be derived directly from whether or not you can begin moving towards those improvements based on your analyses.
So start by looking at the segments that appear related to revenue, and then focus on those groups first. This is why it is a good idea to have a “North Star Metric” – an ideal goal that is at least tangentially related to revenue that you can focus on in order to gain better insight into your customers.
For example, a website like Facebook, which makes its money off investors, stocks, and ad revenue, cares more about engagement than it cares about anything else. They want to know more active users, because the more active users they have the more they can leverage that into more money on their end.
So they wouldn’t necessarily use revenue itself to segment users. Rather, they would use engagement metrics or login data to see if users are opening the app and interacting with it. Uber would use things like total rides, AirBnB may do total bookings, an eCommerce shop may do repeat visitors, etc.
You may have your own metrics:
- # of people that signed up for a newsletter.
- # of visitors that came back at least twice in a year.
- # of users that made a second purchase, etc.
There are plenty of metrics available, both “North Star” and otherwise. You know your business, and you can try to turn that knowledge into better use of customer data.
Some user segmentation platforms come equipped with AI analysis or one click analysis, which provides feedback on segments and potentially valuable segments with minimal work by the business. Other tools may require more data knowledge and training. But no matter what analytics platform you are using, the next step is to begin analyzing the data and generating reports.
NOTE: Data analysis can sound overwhelming at this stage, but remember that if you’ve chosen the right analytics tool, the process to report on this data should be very easy. This is one of the reasons that in the battle between free web data tools like Google Analytics vs. highly specialized user segmentation and analytics software, the software always wins, because it makes analysis and reporting much easier.
Start generating reports based on the segments you discovered or chose to focus on, and then use those reports to compare. Let’s look at a scenario with how this plays out:
Bettys Boutique is an eCommerce store that sells purses, hats, and shoes.
Betty sells an equal amount of all three, and Betty decides to segment based on initial interest – ie, the first item they were looking for when they visited the site before they made a purchase. She looks for the buyers that came to the website in search of shoes, those that came into the website in search of hats, and those that came in search of purses.
When she generates a report on the acquisition of these users, the first thing she notices is that even though all three had an equal number of sales 33%, 33%, 33%), over half of the users first visited the website in search of shoes, while only 15% of visitors first came to look at purses.
What she learns is that sales and marketing strategies that focus on shoes may actually help the company sell more purses as well, because they have evidence that the shoe page is leading to more purchases overall. They also can look for ways to improve the purse page, or follow the path that buyers take that leads them towards buying purses and shoes and see if that might work for hats as well.
Create an Action Plan and Test
Once you’ve seen the data you can decide what to do next. Within the data may be even more information that helps you decide what action plans to take.
Using the above example, perhaps you split test your website and show more purse ads on the shoe pages to see if that improves revenue. Or model your purse page after the shoe page and see how that affects sales.
Then, since you are still using the analytics software, you’ll be able to see if it created a noticeable change towards your goal. Then you’ll be able to re-evaluate with either the same segment or a different segment and continue to improve your conversion rate and ultimately your revenue.
Benefits of Segmentation and What You Can Do With Segments
The primary benefit of segmentation is that it allows you to analyze across segments and use the data you find you draw conclusions about what your business needs to do in order to continue to attract more customers and retain those customers in the long term.
But that’s not the only benefit of segmentation. Other benefits include:
Learning Your Customer
Most businesses have no idea who their customers are or where they came from.
They just assume that their customers are as specific person or a specific way. We see this in the offline world all the time. People assume that the customers of a hardware store are scruffy builders. People assume that the customers of an antique store are older couples.
But are they?
A great example of this is with dentistry. If you work in dentistry, for example, you may assume that your patients are men, women, and children of all ages because that’s who you see in your office every day. But while they’re your patients, studies have shown that women make 90% of the household healthcare decisions.
Thus your patient may be anyone, but your customer is the matriarch of the family, because she is the one deciding where her family goes. You may not know this from intuition, but it is often – or can be – right there in the data for dentists that know where to look.
In the online analytics world, the same truths exist. Except the internet actually has significantly more segments and data to choose from, making it even easier to really know who your customer is, whether it’s a John or a converted Jane.
And once you have that information, you can start making changes that target those specific customers. You can try to see what else John may want to purchase as a way to quickly earn revenue, or analyze what Jane needs to turn into a long term and recurring customer. You can delve deep into their data to see if there are signs that indicate the next steps towards greater revenue.
Understand Their Habits
In addition to a customer profile, segmenting the data may also help you understand how they move down the pipeline and funnel into a sale. You can see if there is a way to improve efficiency during that pipeline in a way that users will eventually appreciate, or you can simply model their habits and find ways to target that for profit in the future.
By creating a customer profile and learning as much as you can about their likes, dislikes, and habits, you can also start creating laser focused marketing that is geared towards those individuals.
On a broad level, it means trying to attract your primary demographic towards your website or app more often. Segments let you learn a lot about each customer, which in turn means that you can figure out how to target them with in order to convert more users in the future
You can market items of interest to them, such as a new products (ie, send emails only to those that you think would be interested). You can offer better customer service by having a specific idea of who they are and what they need. You can price products differently depending on which ones you think will get the most revenue and why. You can come out with new products, new services, and more, and focus your marketing efforts on the demographics or other data that seems like it will be most effective.
Continue to Analyze
The best part of using an analytics conversion tracking tool is that your data doesn’t stop just because you’re analyzing it. It continues onward, trying to pick up more useful data as you continue to try to see whether or not your changes had any effect. This will also tell you if there are trends within the data that can or cannot be identified as resulting from the changes you made to your business model.
Segmenting users takes some of the stereotyping and assumptions out of the website or app workplace, and instead focus on what the data tells you instead. For example, if you run a shopping website that seems like it is geared towards millennials, you can measure that assumption and see if it continues to hold water.
Segment Analytics for Actionable Results
Segmentation is not just an added valuable analytics tool function. Segmentation gives you insight that you cannot see if you only looked at the aggregate data. It may seem like an extra step, but the value of the information you get from segmentation is second to none, as they are on the front lines of your home looking through each nook and cranny and then providing you with the best next options.
If you are interested in a powerful web and app analytics software that uses AI to perform some of the more complex tasks, contact Air360 today for a free trial.