It was supposed to be a nice relaxing day of sorting through his email and getting caught up from a long yet enjoyable holiday break. Unfortunately for Artemus Brown, Analytics Manager for Geometrixx, his expectations were not to be met on that particular day.
Artemus Brown approached the first marketing team meeting of the year like all the others, he was ready to speak to the most recent sales cycles and prepared to answer detailed requests with “Let me look into that and get back to you” When asked by his director “What were the gross sales totals in revenue as well as orders for the week leading up to Dec 25th?” he was armed and ready with his numbers to which he said, “We witnessed 92,000 sales that equal $2,281,000 in gross revenue. This shows an increase in average order value of 26 percent for the period” Given that his answers were above the projections he and all the others in the room smiled and relaxed a bit. His director responded with “Fantastic it looks like our product recommendations software along with our holiday marketing blitz was effective. Can you please break down those numbers to expose how much of this increase can be assigned to our efforts? Also, given that we were netting larger sales than previously, are those sales from returning customers or new customers? Can you also tell me what percent of our traffic is from customers and what percent is from prospects? What are the differences in user behavior in our customers when they were prospects and prospects that never convert? If we can identify customer behavior that we can apply to prospects via targeting, then we can raise KPI’s.” Artemus fired off his standard “Let me look into those requests and put something together for us to review” then sat quietly for the rest of the meeting trying to stay focused on what was being discussed in the meeting and not on building his analysis framework in his head.
The ability to divide datasets into customers and prospects has been a challenge for many Web Analytics practitioners since the early days of our rapidly evolving field. Analyzing the customer’s behavior when they were prospects has been even more of a challenge. Marketers want to understand the differences in acquisition patterns as well as behavior that exist between these two key groups. Understanding and exposing those differences allow them to manipulate prospects via content targeting and personalization techniques to achieve their specific business objectives. Manipulating the outcome by raising average order value, increasing average order size or some other unstated objective is made possible through segmented analysis.
Leveraging Adobe Insight these challenges have been overcome. The transformational capabilities of Adobe’s Multi-Channel solution, allow the analyst to identify the point in time when a visitor, or prospect, becomes a customer. This opens the doors for different kinds of insightful analysis. The analyst is easily able to divide the dataset into the two groups, customers and prospects and witness the changes in size of those groups over time. From there, additional segments can be created and analyzed within the parent segment classifications of customer and prospect.
Below we see how Artemus has divided his multi-channel dataset into two major visitor types: customers and prospects. This enables him to quickly trend the breakout of these two groups over time as well as to apply any additional metrics against each group for further analysis. What is revealed is interesting. In the weeks leading up to the holiday break, Artemus notices that the percentage of known customers visiting the site remains relatively constant. This is very encouraging as he sees this as an unanticipated benefit from their retention efforts. The week of 12/10/2012 highlights this in a unique way, although overall Web Visitors decline week over week, the percent of customer’s remains constant. This leads him to make a note in future analyses to focus on multi-purchase customers and compare their buying cycle with that of single purchase customers to expose the most common durations and apply those to known single purchase customers in subsequent visits.
This functionality also allows the analyst to breakdown the separate prospect and customer behavior. Understanding the behavior of your customers in their visits prior to their first conversion visit is important. This analysis yields insight into how much time elapses between the initial visit and the purchase visit as well as what content aids in the subsequent conversion visit. Understanding the behavior of customers when they were prospects also allows for marketers to better understand the impact of marketing efforts, that may not be tied to last click conversion and thus unknown in most analysis sets. Isolating “assisting” content as well as isolating behavioral nuances can be exploited during targeting and user design testing. Adobe Insight allows for these segments as well as additional an more refined segments to be leveraged in Adobe Test&Target thus completing a full multi-channel optimization system.
Drilling into the behavior of customers during non-conversion visits that occurred prior to the first conversion gleans great insight into
- The amount of research customers need prior to conversion
- The quantity of visits the average customer makes prior to conversion
- The types of content a customer consumes prior to conversion
Understanding these aspects and the customer buying cycle opens the door for content targeting and also user design enhancements to decrease the amount of time between the initial non-conversion visit and the first conversion visit. These techniques can also be applied to known customers in efforts to increase additional purchases.
In order to understand some quick insights into the differences between customer and prospect behavior, Artemus needs to understand some basic information on entry and exit content. Below he sees that most prospects come in through the home page, which is in line with the landing page strategy tied to most of the field marketing efforts. However, an unexpected discovery is that Facebook is accounting for the top two entry pages for customers by percentage during the period being analyzed. This is exactly the small nugget he can report and track to support the marketing efforts of the Social Media team. The hidden gem he finds is that the Women’s section title page is the third most popular entry page by % for customers. This interesting insight reveals a demographic trend that is counter to non-seasonal trends.
The most glaring take-away for Artemus is not something that he wants to report up but rather a concern that he needs to address with the user design team. The top exit page for prospects is a retrieve password page, which is a page that a prospect should never be presented with. Artemus is curious how prospects are reaching an authentication point in their visits and makes a note to run path analysis including this page on the prospect segment. Additionally, he wants to understand what prospects are searching for that is causing the no results pages to be the second highest exit page. The final takeaway he sees is that customers exit at a very high rate on the order satisfaction survey pages, this is typical but he makes a note to discuss any possible incentives to decrease this as an exit point post conversion with the added bonus of having more completed surveys.
The next step for any analyst witnessing these trends in the data would be path analysis to determine the most common paths for prospects to the authentication point as well as looking at specific internal keywords to identify what keywords are returning No Results. This small example highlights how differentiating between customers and prospects can be leveraged in multiple ways. With the unlimited correlation and segmentation capabilities of Adobe Insight, analysts are empowered to quickly drive the analysis in any direction the data is suggesting.
Understanding your customer’s behavior is the first step towards enriching all future communications. Creating high-level segments to differentiate customer groups allows the analyst to expose the hidden nuances of those segments behaviors. Once customer segment behavior has been defined the doors for user design enhancements, content targeting, personalization and retention efforts are thrown wide open. Adobe Insight allows for the rapid query of vast arrays of data over multiple online as well as offline data sources. This enables the analysts to begin their analysis broad and shallow and then quickly go deep and narrow to unearth actionable data points.
In the coming months will be writing more about the powerful analysis capabilities your business can take advantage of today. We will explore, in depth, Adobe’s analytics technologies and solutions. We will lay out the process for leveraging these solutions in Adobe Insight and offer some actionable strategies for immediate results in your business. Stay tuned for the ongoing cultivation of our hero Artemus Brown as he evolves from Analyst to Web Analytics Action Hero.