Let me ask you some behavioral business questions:
Can you identify users who added an item to cart and did not complete the purchase?
Do you know what are the most popular events users do before they churn?
Can you point a trend that shows that users who did not return to a game for a specific number of days will not return at all?
If these seem complex to analyze, you’re probably not applying behavioral analytics.
Traditional analytics wasn’t built to understand behavior of users, because it isn’t the way data is typically treated. Now event-based behavioral analytics, built on multi sourced, non sampled, non aggregated big data, has become essential for your online success.
Measuring KPIs like; active users, where they came from, number of new sign ups, DAU or MAU is still important for monitoring general performance. But if you really want to know what your users are doing on your application, what really made them convert, why they bought what they did, or what is the optimal path for conversion, you need to apply big data, event-based behavioral analytics.
The Why, the How, and the What
Using behavioral functions is necessary to get answers to more complex, multi-layered questions around the Why, the How, and the What. With the behavioral analytics approach, using the right tools and having the right data, you could ask essential questions such as Which items did users search for compared to the products they ended up adding to cart? or in depth retention questions like How long does it take a registered user to ask for help? and how does this vary between users from different traffic sources?
Behavioral analytics is about adding the context and the time dimension. In other words, it means looking at a time-based sequence of actions (or events), analyzing the main goals within the context of events that happened before and after, or explore user paths to really analyze causes.
Putting user events in context
Behavioral analytics relies on smart, event- based tracking that collects every action a unique user performed during his life time on the website or app. Then the multiple event properties are used to contextualize the event in the series. For example an item_ID property will be attached to the Add_Item_to_cart event, to facilitate an analysis of the user’s behavior towards buying that specific item, or even compare it with behavior of users who bought other items.
Behavioral analytics means you can analyze the user’s, or group of users’ journey, in a single visit or across many visits during the user lifetime.
It can help to find interesting, application specific metrics that could not otherwise be easily calculated with traditional analytics. Things like, clicks to conversion time – a metric for how many different clicks or page views occurred before the conversion. Or a metric to compare the amount of money a user spent with the time spent on the website before conversion.
Behavioral insights of user retention
User acquisition is important, but user retention is challenging and often requires deep behavioral analytics. An eCommerce website could gain a lot by identifying the behavioral pattern of a first-time shopper that is most likely going to return. Like in the case of one eCommerce company which gained an important insight that first-time shoppers who spent over 20 mins searching for items, are more likely to purchase.
They then turned this insight into action by targeting a specific promotion to first time shoppers after 10 mins on the site, that encouraged them to engage and spend additional time on the site. that increased the completed purchase rate dramatically.
Behavioral analytics is the next generation of BI, analyzing user behavior as a time-based series of actions. Together with advanced visualization of multi-dimensional funnel, path analysis, cohort and reverse cohort, you gain agility and quick answers to the most complex behavioral questions.
Identify the profile of players that are likely to make a deposit
A common KPI for online gaming companies is the FTD-First Time Deposit, a key “milestone” in the user journey that makes a player “hooked” on the game. With behavioral analysis, one gaming company discovered that players who win the first game are most likely to become FTD. For these first game winners, they added a “test game” before the FTD, in which the user always winds the first round. You can imagine the uplift.
Applying behavioral analytics to online gaming also helps to analyze the profile of new players that are most likely to make a deposit during the second visit and become paying users. It’s easy to think that demographic characteristics like age and gender are a predominant factor in this equation, but using behavioral analytics one company saw that a player who spends 6-times longer than the average, during the first visit, has a higher probability to make a deposit. This insight was used to offer these identified players an upgrade to their status upon entering the game in the second visit, providing a higher incentive for the player to make a deposit.
User behavior with online content
Publishers of online content are tasked with both driving a lot of traffic to their content, and then monetizing this traffic through paid advertising. Some critical behavioral questions publishers would gain from behavioral analytics would to understand:
Which content items get more views?
Which content items gain more ad clicks?
Which content items have a higher likes or share rates?
Which traffic sources refer more users who view/click/share/like these content items?
By gaining such valuable insights through behavioral analytics publishers can then promote the right content through the right channel in the most optimized way, making the greatest profit.
So don’t throw away your traditional BI, keep it in place, and use it to constantly monitor your KPIs. But surely it’s time to advance to behavioral analytics using huge data sets to understand users’ behavior and get answers to more complex questions.
To effortlessly experience big data behavioral analytics in action, sign up for free trial.