eCommerce App Analytics Best Practices - Whiteboard Wednesday

Localytics Senior Solutions Consultant, Richard Cipolla, dives into a best practices series about kickstarting app analytics to get the deepest insights. First up: eCommerce Apps.

 

Transcript

Hello and welcome to another Whiteboard Wednesday. I’m Richard Cipolla, Senior Solutions Consultant here at Localytics, and today we’re going to be talking about how to tag your e-commerce or retail app with the right metrics. This will also apply to any apps that have a major shopping or e-commerce component to them, even if it’s not the sole focus of the app.

 

Tagging Events

So, let’s starting talking about some of the events to tag in your app. The local SDK automatically collects things like the device type, either Android or and iPhone, or if it the users first time in the app. Events are a good place to start tagging your app, because they describe the actual actions users take within the app. Events also have attributes associated with them, which are good details for those actions and will help you get some good, granular insights about those actions.

Let's move to some of the events you should tag. The first one is an item-viewed event. This will describe the actual content that users are viewing, and should have attributes for the brand, price, name, category of the item, and so on, to see what content users are looking most often, even if they’re not ultimately buying. Also, you’ll definitely want some sort of events for any actions they take with that item. Are they sharing that item, favoriting that item, rating that item, commenting on it, etc? This shows what kind of interest users have in the items, even if they’re not ultimately buying them.

Next, you’ll also want events for every time the user gets to a results page, whether it’s a browsing page by category, if they sorted it from prices high to low, one that describes what the search was, any sort of filters applied to that search, and again, how far they went down on the page. Did they get through the tenth item on the results or the fiftieth item?

You’ll definitely want to tag all the events that are part of the checkout process as well, after the user views an item, adds it to a cart, and then starts the checkout process and completes each step, including the final confirmation screen, and then actually confirms of the purchase. For all these events, make sure you tag them as late as possible to describe some of the key details of the action. Again, tagging the category or viewed event that we talked about after it’s happened lets you see how far down the viewing they get.

 

Choosing Custom Dimensions

Now, let’s talk about some of the custom dimensions you’ll want to tag in your app. First, you’ll want to tag if the user is logged in or not logged in currently. A lot of apps will let you view items even if you’re not logged in, but to actually checkout and buy the item you’ll have to log in. So you should see how many users are logging in early on in the process, before they’ve even begun the checkout process, and thus avoiding that step to see how it’s affecting the checkout process.

Next, if you have some sort of measurement for this, and you have a campaign running for that promoter score or something similar, you can have a custom dimension that describes a positive user, a mutual user, or a negative user based on the results of a messaging campaign, to see how the behavior works for the positive users and try to mimic that for future new users in your app.

 

Defining Funnels

With the technical details out of the way, let’s turn our attention to some of the dashboard features you should have in the app, like funnels and segments. Let’s start with funnels. These will be really critical to your analysis of your e-commerce app. The major funnel is going to be purchasing related. A user is actually viewing categories and getting to search results, and then looking at items, adding them to a cart, entering the address and credit card information, confirmation, and/or actually checking out. That’s the main purchasing funnel, but you’ll definitely want to include the different variations of this funnel too. Consider if they started the funnel by searching instead of viewing the category, or if they added the items in the cart before checking out.

 

Establishing Segments

Next, I'll move to the segments you’ll want to tag in the app. This will be a good way to differentiate how features, particularly discovery features of the app, are working, and if they’re making users convert. You can create a segment for a user who searches and a segment for users who actually browse the app to see if one leads to more success than the other. If you’ve got some sort of differentiator, like a special item comparison view, or some sort of carousel view, you can see if users who engage with that are more likely to buy, or if it’s not impacting the results as much as you want. Then, you can either make it better or turn your attention elsewhere.

 

Refining The Messaging

With the segments, events, and custom dimensions all together, they flow into some good messaging campaigns you can run in your e-commerce app to push users to buy something. The first messaging campaign that you might want to run for users who have things left in their cart, could be an offer for free shipping in the message or something to entice them to actually finish the checkout funnel.

Also, once they’ve checked out, you can send them a message that asks them if they enjoyed their experience or not. If they rate it highly, you send another message suggesting they rate your app in the app store. That way, you can boost your app store rating with users that had a really good experience. Then, you can also have messaging campaigns to encourage users to interact with those features of your app that help differentiate it from the competition, like a special item comparison or carousel view to see if more user interaction would make it more successful.

 

Conclusion

These are some basics to start tagging your e-commerce app, events, funnels and messaging. Hopefully these will get you some of the granular insights you need with Localytics. Once again, I’m Richard Cipolla; I hope to see you again next week. 

 

New Call-to-action

X