Analytics

Learn more about how to perform audience segmentation based on people's behavior, and which metrics are currently available for this. Or, you might decide to analyze funnels based on user activities and their profile information. 

Events Segmentation Analytics

This basic type of event analytics serves to allow users a better insight into their audience. Events Segmentation has three core blocks:

  • Selecting segments
  • Selecting events to show
  • Events data visualization

Select Segments

Behavior data helps you better understand how your audience interacts with an app or a website. The audience segmentation lets you dig a little deeper and see the difference in that behavior. The segments are based on the person attributes - static data related to each end user whose information is stored in the People module. This is usually some very current information that is likely to change in the future.

The segments that have been created earlier can be used again. Click Add Segment and the right-side panel will appear.
 

adding existing segments to analytics
 

If you don’t find what you need in the existing segments list, the other option is to create a new segment for your report.
 

creating segments for cpaas analytics


Up to 4 segments can be created in the Event Segmentation analytics. The image below displays a basic 2-segment scenario.

selecting audience segments on cpaas

Select Events to Show

Events reflect user actions inside an app or on a website. Events Segmentation analytics can display up to 4 different events on the user activity graph.

On the Events Segmentation page, scroll down to Select events to be shown on graph and decide which events you want to include.
 

event for analytics graph

Events Data Visualization

Click the +PROPERTY link to set the event properties. When you include properties, the graph will also contain certain conditions that have been met and are related to that event. Take a music streaming app as an example: if you select the Song Played event only, the graph will show how many people listened to the songs., and if you include a song genre or name, the graph will show how many times people listened to the specific genre or song.  

The data related to the event will be shown as a line graph chart. When you’re done specifying segments and events that should appear on the graph, click RENDER GRAPH.

Remember to select a time period that you would like visualized (modify this later if you need to). Choose the last 7 days, last 30 days, this week, previous week, or a custom range that you define yourself. You can also get information for different periods within your set timeframe. For example, if you need a weekly graph, but also want to see daily data, select Daily too.

graph for audience analytics

Metrics

Four metric types are currently available on the Events Segmentation page:

  • Events
  • People 
  • Adoption 
  • Engagement

Events metrics show the number of selected events that happened on a specific date. You will get a better idea of whether the users` amount of activity aligns with the goals you have set for your product.

To understand how many people performed specific actions within different time frames, turn to the People metric.

The Adoption metric reveals whether certain parts of the app’s functionality are being used by your audience or a specific audience segment. Keep in mind that the best practice would be to set up a segment of people based on the registration date. You can select a segment of people who registered in May and then check out the report for Adoption related to the months following the registration and to the most important features that you have. If you are going to use the Adoption metric without a registration-based segment, you may get incorrect information because the audience will include new users who were not able to perform any actions before they registered!

The Engagement metric helps you understand how many actions are generated by one person on average. If you are trying to inspire your audience to perform many actions during the day, this report will uncover whether there is a growth or decline in engagement. As with the Adoption metric, you will need to select a segment of people based on the registration date. Otherwise, calculations will include 'new' users who simply could not have performed any actions before the registration!
 

user events shown on graph

Funnel Analytics

A sequence of user actions on your website or mobile app tracked with Funnel Analytics shows how different audience segments behave so you can better adapt your engagement strategy.

Let`s take user activation as an example. Activation often means that a customer goes through a number of steps before realizing that a specific product is something they need. The ultimate goal of marketing teams is to encourage more customers to reach that realization and complete the funnel. Before you are ready to analyze the funnel data and adapt campaigns accordingly, you need to have the customer information stored inside the Infobip People module and be able to track events.

Next, in the Moments module on the Infobip web interface, select Funnel Analytics and do the following:

  • set up an audience segment or use the whole audience
  • select the sequence of events that represent your funnel
  • specify conversion window (the adequate timeframe for the funnel)
  • specify the timeframe for funnel calculation

We will quickly go through each step, and then you can go ahead and start analyzing and adapting your strategy!

Segments

Calculate a funnel either for all the people you have or for certain audience segments. Up to  2 segments can be used in funnel analytics. To specify the segment, select the existing standard attribute (such as country, city, etc.) or create a custom one.

Segments are helpful if you believe that certain groups of your audience might be completing the funnels differently. If that`s the case, you will need to figure out the reasons and how to make your ‘lower performing’ users convert at the same rate.

You can use the existing segments or create new ones for this report.
 

segments in funnel analytics

Events in Funnel

The next step is to define the funnel. Think about what sequence of user events has to happen for the funnel to be completed. This order of events is important, and it can`t involve any ‘optional’ in-between steps. Keep in mind that you can include up to 7 events in one sequence.

Imagine an example case with the user registration as the starting event and consider a possible timeframe for the entire sequence: Registration - 12:05pm; Watch video - 12:10pm; Payment for full video - 12:30pm.

The screenshot below shows how to set up events that you will include in the funnel.
 

customer payment event in a funnel

Conversion Window

Define an approximate amount of time that will pass between the first and last step in the funnel. For example, you can set the conversion window to 15 minutes, as we just did for the Registration - Watch Video - Payment for Full Video sequence.
 

funnel analytics


 If all the steps are not completed within 15 minutes, the funnel will show a drop-off.

Calculate the Funnel

To record one unique person within a given timeframe and a given conversion window, the whole funnel should happen. If it happens successfully at least once without drop-offs, then for that person the funnel is marked as complete. If that person enters the funnel multiple times but has at least 1 complete funnel, this still counts as a success and all other entries will be disregarded.

conversion window

Funnel Graph

Pick a specific timeframe for the funnel calculation. But consider the fact that your app/website evolves constantly and the same is true for your acquisition efforts and communications strategy. That also means that the behavioral pattern of your users from six months ago may not match the present one.

When you are ready, click Render Graph. You will see the data breakdown for both People and Events. 

Breakdown by the number of people

The resulting graph will reflect the funnel that you have specified. The number on top of the bar represents the percentage of people that have arrived at each specific step in the sequence.

customer funnel graph
 

Hover over the bars to see both the percentage of people who reached the specific step and of those who dropped off from the previous step, as well as the exact numbers for both.

breakdown by the number of people in funnel
 

Breakdown by events

Once the first step happens, that’s when each new funnel starts. The data displayed on the top of each step in the sequence takes into consideration the events that have reached that specific step. But, same as with the Breakdown by the number of people, when you hover over the bars, a pop-up will appear showing the percentage of people who reached the specific step and who dropped off from the previous step, as well as the exact numbers for both.

This graph is helpful when you want to understand, for example, how many sequences were started but not finished, which events in the funnel might need improvement, etc. 
 

events graph on spaac
 

Table View and Tags

The information from the graph can be shown in the table view as well. There, you will find both exact numbers and percentages. On the image below, the values in the segment columns show the number of people that arrived at the specific step from the previous one.

There is one additional feature in the "Performed by the number of people" table view: you can add a tag to some group of people from the funnel so that later you can use it for communications or further analytics. Hover over the exact number and click the tag icon that will appear next to the exact number. On the pop-up menu, select the existing tag or create a new one.

Tagging will help with future communication or analytics in situations when you want to select this specific group of people.