Pages

Monday 3 August 2015

Discover the Hidden Value in your Customer Data

As the field of analytics turns more and more predictive, emphasis on using data moves to discovering unknown landscapes. But moving beyond the obvious to completely understand your customers’ journey is a very complex step. How can you derive patterns from countless data sources to optimize customer experience and create a competitive advantage?
 
CREATING A 360-DEGREE CUSTOMER VIEW
 
Companies today are focusing on creating a 360-degree customer view. To do so, the first step is to have your data collection up and running, making sure that you can deliver data to a centralized environment, from which it can be used for further processing. In many cases, this environment is a data warehouse. In an ideal situation, all data from online channels, CRM technology, campaigns, call centers and order management is structured an gathered there to draw the best possible conclusions for optimization. However, this comes with many challenges to collect “good data” related to the volume, velocity, variety and veracity of data. We previously highlighted the struggle to stream web data to a data warehouse, but in the entire ETL process, there are many more. Luckily, one factor that has disappeared from the list of challenges is costs of data storage. That has become very affordable.

DISCOVERING HIDDEN VALUE
 
Nevertheless, if you manage to deliver all your data to a centralized environment, where do you start if you want to find patterns and insights to outperform the competition? In other words: how can you discover the hidden value in your customer data? The traditional approach is to put analysts and data scientists to work. They define a hypothesis based on the events they think are important. For instance: which events lead to churn or conversion. And then they start collecting data to see if factors like income and the event are correlated. To really validate their hypotheses, often an iterative process, may last long time. In practice analysts and data scientists spend up to 70 or 80% of their time on data preparation.
 
DIMENSIONS OF DATA
 
With the internet of things and web data, vast amounts of data are being added to those of traditional sources, making data processing even more complex and running the risk of not seeing the wood for the trees. Especially when this data is poorly structured or customers are high up in the conversion funnel and many variables have to be taken into account. The more dimensions the data has, the more complex it gets to obtain insights, making it very hard to create substantial business value.
 
FUNDAMENT FOR DATA ANALYTICS AND DATA DISCOVERY
 
As in practically every data challenge before, smart technology can make your life easier. There are discovery platforms available to recognize patterns in vast amounts of data. These can identify the most important variables that need further research to determine their business relevance. Analysts and data scientists can use this information to perform business analysis and determine the actual value of these variables in the customer journey as a whole. However, for these platforms to reach their maximum potential, you need to have the right infrastructure in place. This has to cope with the volume and velocity of the data you are collecting. Also, you have to be able to connect all relevant databases to create your 360-degree customer view.
 
APPROACH

So to discover the hidden value of your customer data, here’s the way to go:
  1. Make sure you can collect data from all relevant sources
  2. Store this data in an environment powerful enough to process it and with no limitation when it comes to storage capacity
  3. Run a discovery platform that can determine which variables have most effect on the important events of your customer journey.
  4. Make a predictive model to foresee which visitors will match this variables
    Take actions to optimize this customer journey and reduce churn or improve conversion

    SOURCE: processexcellencenetwork
 

0 comments:

Post a Comment