I often meet with our customers to help them understand how to connect modern technology to business success. The ever-present question at these encounters is “Where do I start?” For them, they may understand that they need a data-driven strategy or the culture may aim to take a shift to being guided by data. These are often goals set by the executive team with little guidance on how to execute or implement. This can be frustrating and impede even aggressive initiatives. Cloudera often partners with these companies to define the desired path, set reasonable milestones and success metrics, and aide solution design. The broad nature of customer insights across industries and use cases means that there is no one single journey applicable for all users. But when it comes to starting your efforts it may be comforting to know that there are common patterns to adoption that can help you experience use case success, faster.
Cloudera and TDWI have recently partnered to outline key attributes to success for implementing customer insight solutions. These are the solutions that help us better capture customer data (across systems and touchpoints), open up that data to advanced analytics, and build data products that deliver insights and actions to customers. We have introduced the research in a blog released last week called “Introduction to Six Strategies for Advancing Customer Knowledge” and the full research can be accessed here. In this blog we will focus on the critical first step to advancing customer knowledge.
Did you know that over 80% of companies who choose Cloudera to start with the same prescriptive activity? It is not the heart-racing, bleeding-edge use case that you hear about in WIRED magazine but a foundational and required step in getting to almost any customer insights solution; bringing data together.
While it is true you can buy a BI or analytics tool that will analyze data in different locations, we believe that their are expressed benefits to combining data in a physical location. These benefits include but are not limited to decreasing operational overhead, joins and aggregates that are not possible when data in stored separately, and obstacles in opening up self-service data discovery. David Stodder hints at the problems this can create in the following section.
“When decision makers can only see fragments of customer data at a time, it limits their understanding of both individual customers and larger trends across populations of customers and prospects. Organizations dependent on such fragmentary views often cannot be agile with information because the limited sources support only one or a few types of queries and reports. Data science and advanced analytics deliver more value if organizations can deploy models and algorithms on detailed data brought together from many sources, not just on summaries or samples contained in a single data warehouse or data mart.”
And teases at the reward.
“Bringing together as much of this data and information as possible will help organizations gain a richer, more detailed picture of customers. As their intelligence grows from the newly combined data, organizations can add insights to customer profiles, increasing their value exponentially. These enhanced profiles can improve real-time customer engagement, sharpen marketing campaigns, and support use of analytics to predict customers’ future behavior. For these reasons, it is important for modern organizations to seek a comprehensive, 360-degree view of customers and prospects.”
Real World Examples
Let’s look at some real-world applications of some of these outcomes that David discusses. SFR is the second largest telecommunications operator in France. They believe the proliferation of telecom networks and digital technologies allows people to enrich their lives through easy interactions with the world around them. SFR implemented Cloudera to improve real-time engagement by creating a shared, detailed view into the customer journey that would be available to employees across the company for real-time search, reporting, and analysis. RocketROI provides a real-time, cross-channel, software-as-a-service (SaaS) performance marketing solution that enables small and medium-sized businesses to precisely target online customers. RocketROI partnered with Cloudera to sharpen marketing campaigns by improving click-through rates at a dramatically lower cost-per-click. As a result, RocketROI customers have seen significant performance improvements with some customers reporting a more than 3,000 percent increase in sales and conversions. Quotient Technology (formerly Coupons.com) is a leading digital promotion and media platform, distributing digital coupons, coupon codes, and card-linked offers. Quotient relies on Cloudera’s platform to predict future behavior patterns and target relevant opportunities. One of the challenges Quotient faced in building behavioral targeting and personalization solutions was a siloed legacy environment. Quotient can now target promotions based on shoppers’ purchase behavior, helping retailers and CPG companies increase consumer engagement and loyalty while better tailoring outreach to consumer needs.
These companies were able to receive notable benefits from their data collection and aggregation efforts. You can too, as David describes…
“The best way to achieve this is to bring customer data together into a single, central data repository or enterprise data hub that gathers together diverse data generated by customers in different channels and at different touchpoints. This is not trivial to do in a data warehouse because customer data is often held in disconnected data silos or applications that have different data types and models. Customer transaction records are in one or several databases linked to OLTP systems or business applications, while CRM and sales force management systems hold other bits of vital data. Then there are the even less structured customer service records, call center interaction records, customer satisfaction survey data, and information generated by online behavior and engagement. External social media data could also be valuable for analysis.”
A New Approach
Endeavouring to do this on legacy technology is often met with the same stale results and experienced pitfalls. Data Warehouses and data virtualization may offer some remedy but as it is pointed out in the research…
“Although collecting all this information in a classic data warehouse using a single structured data model might seem ideal, in implementation data warehouses are difficult and slow to create. They are usually not comprehensive enough and have limitations in terms of the types of analytics that they can support. Some organizations try to use data virtualization or federation middleware to expand the data reach by creating virtual views across sources. This option would work for accessing some additional data sources, but virtualization can require complex up-front modeling and setup and does not enable organizations to access the full mix of customer information at the file level.”
“Organizations need a strategy for creating the single customer view. Even if it is not possible to move absolutely all of the data and information into a single source such as an enterprise data hub, the more data they can locate in this source, the sooner they will be able to decrease time to insight through the application of big data analytics.”
We will be back with you shortly with more strategies for delivering customer insights through advanced analytics. To review more advice from the team at Cloudera and TDWI be sure to access the full research.