Nearly every operator is exploring how to capitalize on the promise of big data. Functional groups, ranging from customer support to network operations to marketing, have identified strategic opportunities for using big data to drive better decision-making.
CMOs in particular have focused their sights on using big data to drive better results. Making an investment in better understanding customers means the ability to leverage insights to more effectively differentiate the mobile experience by delivering better products and services and more value to customers.
Although the end goal is clear, many CMOs are asking: How do we get there and how long will it take? And this has IT asking: Do we build or do we buy?
There are several factors that contribute to the build vs. buy decision. Being able to handle the voluminous amount of data is part of the equation. Collecting, storing, and crunching the data – structured, unstructured, and semi-structured – is a critical capability, and one that is neither easy nor inexpensive when you’re talking about infrastructure that leverages the latest big data technologies to harness data captured across the organization.
Another part of the equation is determining how to extract the information so that it brings value to the company. For the marketer, this means having the capability to first understand who a customer is and how he or she behaves, and then the ability to act on insights in an automated fashion on an individual customer basis.
There is also the question of scaling and being able to monitor the dynamic behaviors of millions of customers so that you can actually influence the decisions that they make each and every minute.
Many marketing and IT organizations, driven by top-down imperatives, are addressing the need for big data solutions to transform how they engage with and market to their customers to drive better results. In working with mobile operators across the globe, we’ve identified the key questions being asked as it relates to the build versus buy decision on big data marketing platforms:
Should marketing be a starting point for operators making an investment in big data?
The two groups leading the way with big data initiatives in the majority of operators are marketing and network engineering. Facing highly competitive environments, marketing organizations are consistently looking for every advantage possible for retaining existing customers and maximizing revenue opportunities with those customers. We have seen big data initiatives consistently result in quick hits as marketers have been able to act upon behavioral insights that have been lost in the aggregated customer metrics that are the cornerstone of traditional approaches to marketing.
How are marketing groups determining whether the value of a big data analytics platform justifies the investment?
Marketers have an innate confidence that the better you understand a customer and the more you are able to personalize your interactions with that customer, the more successful you will be in achieving your KPIs, whether they be around revenue, churn rates, usage levels, promoter scores, or other metrics. They also have increasing confidence, supported by case studies from other operators pursuing similar initiatives, that a big data solution is a path to success. The question is how quickly the investment will realize the return. Projected returns are being established from both pilot projects and industry case studies. Determining costs has been easier for operators pursuing a buy strategy, particularly when working with vendors with success-based business models. On the other hand, the ROI calculation associated with a build strategy has been more problematic as the returns have been less consistent, the time frames considerably longer, and the costs subject to more unknowns and risk.
Is there a trend toward building or buying big data marketing solutions?
The early trend has clearly been toward buying solutions. A considerable factor in this trend relates to the business case around the solutions. The buy business case is generally able to show much less risk, faster returns, and greater certainty around returns and costs. The risk associated with a build decision is amplified by the state of the underlying technology required to construct an end-to-end big data marketing solution. Much of the vendor investment focus has been on the underlying infrastructure for storing and managing big data, while the technology to build applications to take advantage of big data have lagged, making applications extremely challenging to build. As a result, most operators have looked toward vendors that can provide complete, out-of-the-box solutions instead of embarking on the journey alone.
What are the primary considerations when making this decision?
First and foremost, operators are looking at ROI and risk. They are also looking at the strategic nature of big data across their organizations and at their own core competencies. Those that are confident in big data as a method to drive long-term returns are often starting with solutions they buy but keeping an eye toward ultimately bringing the solution completely in house and driving further development internally. Others will recognize they lack the core data science and engineering capabilities required to get the most out of big data marketing solutions and will instead strategically partner with one or more vendors with leading solutions and visionary roadmaps.
What are the challenges that operators building this type of platform are facing today?
A considerable challenge has been in finding the staff with the skills and experience required to build a big data marketing solution. Individuals with the requisite skills sets are in high demand and often quickly hired by the vendors providing the underlying big data technology and the applications, including marketing solutions, built on top of that technology. For those operators that have been successful in recruiting and developing their own staff, they still face significant challenges in building a solution. With limited tools, experience, and examples of successful big data marketing initiatives, efforts to build the solutions internally have been fraught with time and cost overruns and challenges with scaling, implementing data science, and achieving desired results.
Duane Edwards is senior vice president of product Development for Globys (www.globys.com).
Edited by Stefania Viscusi