Adobe Analytics Offers New Features for Improving the Customer Journey

This week, Adobe Analytics introduced new features that provide more understanding and visibility into the customer’s journey and behavior. We are continuing to mature the opportunities for improving the experience you are providing to your customers. This new release will help you to drive better-informed business decisions and properly explore customer behavior around key events. Much like the Lego movie, everything is awesome. Here are a few of the feature highlights for this release.

Best Fit Attribution (Algorithmic)

http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2014/10/Best-Fit-Attribution-e1413460266229.jpg

How well are current marketing channels contributing to incremental success? Or in other words, when customers engage with marketing, how likely is it that they will end up converting? Best Fit Attribution objectively determines the fractional impact of each marketing touch along the customer’s journey toward success. This algorithmic attribution offering uses advanced statistics and machine learning techniques to help lead to a better understanding of marketing campaign effectiveness. It can be used broadly at a high level with marketing channels and the entire customer population—or on specific areas of interest such as individual campaign initiatives and specific customer segments. Analysts can use this feature to measure the effectiveness of programs that you are currently running, inform budget decisions, and even model around nonrevenue (or micro-success) events such as application starts and email signups. Stay tuned for some more follow-up blogs on this topic.

Cohort Analysis

http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2014/10/Cohort-Analysis-e1413460310242.jpg

What are customers doing before they purchase a product or register an account? What marketing efforts, pages, features, and workflows are driving that desired behavior? To truly analyze the customer journey, it is important to understand that it is more than a matter of knowing what customers are interacting with and how frequently. A key component is understanding cause and effect. What preceded a particular activity and what occurs afterward? The Latency Visualization feature provides a means to freely perform cohort analysis on your customer base. On the day that users convert what percent have additional visits? What other activities and features are users engaging with the day before they convert? What remarketing effort is bringing customer back the next day? In this release we are enhancing that analysis with the ability to explore from the customer engagement perspective, outside of specific time associations. Consider this scenario in the travel industry.

Tim, your customer, has been researching for the next big family vacation and wants to make sure he’s got it all planned out before booking. He’s close to being done but has this upcoming business trip that’s going to take his attention away from this activity for a week. But no matter, as soon as Tim returns he completes the booking.

Now what’s interesting about that scenario? One quick observation is the time boxing of customer engagement as well as the assumption that customers think in the same manner. Here, Tim was researching a great trip for his family but got pulled away for a short time. This didn’t impact his intentions or expectations to book, but it did break his activity apart when evaluated over time. It’s pretty easy to see how customer commitments and responsibilities, like Tim’s, can impact your exposure to the true customer journey. To help uncover richer customer trends, we’ve enhanced our Latency feature to support the analysis of a cohort purely around customer engagement and actions. It’s no longer just a question of what did a customer do yesterday but now what did a customer do in “the visit” before they purchased? Whether it is the same day, yesterday, four days before, or even a month before, you will be able to expose a more complete customer journey. Check out these blog posts to learn more about the Latency feature and cohort analysis.

What-If Analysis

http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2014/10/What-if-e1413460379701.jpg

When planning a new marketing campaign, the goal is to impact the business by increasing a key performance indicator (KPI), such as revenue or customer registrations. But it is typically easier to understand the potential lift on traffic metrics like Visits and Page Views versus those critical KPIs. Now with a Regression Analysis workflow we’ve released directly into the graph visualization you’ll be able to easily project the impact of driving more traffic during a target period on revenue, or other such KPIs. For example, generating a 10 percent increase in visits for a target customer segment during a week could increase revenue by 20 percent. Having that information and perspective will help drive more data-driven prioritizations and in establishing target goals, those based on actual trends and opportunities from within customer data.

Easily Export Data to R

http://blogs.adobe.com/digitalmarketing/wp-content/uploads/2014/10/Export-data-to-R.png

We enabled integrated statistical and predictive workflows within Adobe Analytics over the past two years to put advanced capabilities in the hands of the analyst. And we know there are more opportunities out there for the data scientist to continue to build new models and statistical approaches. To help facilitate those highly opportunistic efforts we are releasing an update to our very popular Segment Export feature. This feature allows you to take the fully transformed data, with all your business rules and segments applied, out quickly to utilize within other applications, like R. And with the increase in offline data integrations from CRM systems, interactive voice recognition (IVR) systems and transactional data warehouses we realized the need to provided more standardized output formats. These formats help accommodate different data standardizations and collaborative information sharing. Now by offering a more native configuration for including a header row and an automated CSV reformatting (applying the delimiter and escaping) these data exports can easily be processed into R to enable custom statistical models for reintegration back into Analytics and the Marketing Cloud.