One of the more powerful aspects of our Explore application is its ability to segment the data on the fly. Allowing you to drill down into your unlimited data. Let’s take a fairly common marketing question and see how we can answer it in Explore: How is my latest campaign doing with customers using tablets and mobile phones? Mobile users are an important target market.
When you get into Explore, first, create a view of your campaign.
Here is what that view might look like without a segment:
Now, let’s create a segment for just users of tablets:
Notice the scope of the segment is for sessions. It could be for events or visitors, but we are focused on sessions here.
This is the resulting view for campaigns with the tablet segment applied:
Since we’re really trying understand the impact of the campaign on mobile too, we’ll create a segment for mobile devices:
Since the segment is almost the same as the one we made for tablets, we copied the tablet segment and just changed the device name.
Here is the resulting view with both segments applied:
That is not what we were expecting. We should see more traffic than just the tablet segment. Why? Well, in Explore, the conditions for ALL segments must be met for data to get to the view.
Let’s create the segment again:
Now that we have our segments established, let’s focus on getting all the data. Move the slider bar from faster query to more accurate on the right side:
This looks good. However, it raises the question how did the tablets and mobile compare with desktops? Let’s look at this another way. We’ll use a view with all the device types:
Now we’ll add a segment for sessions with a campaign:
It’s the same as the original! The issue here can be seen in the original campaign id view:
It is because sessions without a campaign are represented by a “-“. So in fact campaigns exist in EVERY session!
Let’s try a different route:
Since we know that campaigns exist in every session, we just exclude those dashes.
Great! This is exactly what we were looking for. We’re seeing mobile phone users at about half of desktop users. Tablets are a very small set of campaign respondents.
From here we can dig in more if we want. You’ll see the bounce rate for mobile users is quite high. Perhaps we need to understand if there are issues with the mobile phone experience. Or, we could look at conversions, since that’s what we’re really aiming for.
As mentioned earlier, your ability to drill into data is limitless!