Guest Author: Ramon Chen, VP Marketing Reltio
Article published at Broadcasting & Cable Magazine
A recent Forbes article cited a survey in which 74% of respondents indicated that a data-driven marketing strategy provided a competitive advantage in customer engagement or loyalty. While surveys can of course be misleading, there is no doubt that having better access to the right data leads to deeper insights which most companies hope to tie back to increasing efficiency and revenue.
It’s therefore no surprise that there’s more big data, data-driven analytics and visualization tools being made available every day. These terms have permeated the business lexicon, leading to executives in the C-suite and their marketing teams investing heavily in technology as well making “data scientist” the hottest job du jour, to uncover patterns and insights in a sea of numbers and facts.
What many don’t realize is that their perceived gains in efficiency and revenue could be even greater. Even more shocking, their data may in fact be driving them in the wrong direction.
How trustworthy is your data?
While technology exists today to bring together information from multiple sources to spot and forecast trends such as conversion rate and churn reduction patterns, an often-overlooked ingredient is the reliability and quality of the data.
This is not news—reliable data profiles and correctly identifying customers across data sources has been a 10-year-old IT-only discipline and a $1 billion software category called master data management (MDM). This category is getting new life with cloud and big data technologies, which are helping to deal with ongoing data quality at scale and data volumes that far exceed the capabilities of legacy systems. But IT still needs to control data quality, and it’s traditionally a time consuming process. This flies in the face of the desire of business teams to be empowered to be right faster in their decision-making processes.
With hardware and storage costs more affordable than ever, the temptation might be to just dump everything into so-called “big data lakes,” and let data scientists comb through the information using Hadoop and other big data tools as needed.
As last summer’s debate about “Data Lakes” (or massive data repositories) between Gartner and Infoworld proved, there are always two sides to every story. It really comes down to a balance between governance, security and reliability of the data (the responsibility of IT teams) and deriving relevant insights in a timely manner (the hope of business teams). Enterprises need to plan for this age-old challenge, otherwise they will just end up with the same business time-to-value problem they had before…only bigger.
Don’t forget about your frontline business users
Handling extreme data volumes through back-office data management is actually not the real challenge. When insight is obtained using a standalone analytics tool by a data scientist, the actions that end up being taken by frontline business users are left to chance.
Operational activities that take place in separate siloed applications downstream cannot be controlled. Additionally, analytics are typically based on knowing the questions you want to ask before rummaging through the data for the answers, whereas frontline business users want insight within the context of a person’s job function and goals.
Based on a modern data management platform that blends MDM discipline and big data infrastructure, a new breed of data-driven applications bring together siloed data from across the enterprise, making it reliable in real time with not only insight, but actual recommended actions for business users in the context of their day-to-day activities.
The catch is that these applications should not require significant training. This topic has often been called the Consumerization of IT, making enterprise, data-driven applications as simple to use as popular consumer sites such as LinkedIn and Facebook.
This also leads to another trend known as the democratization of data, providing ubiquitous access to all business teams and giving them the ability to contribute their insights and up-to-date information they may encounter to improve the overall quality of data for the enterprise.
How do you know that insight is resulting in the right action?
With all this technology, why do companies still measure ROI via surveys and conjecture? It’s because direct correlation of ROI requires the ability to tie actual actions taken back to the original data, insights and recommended actions, to form a closed-loop system that proves and justifies outcomes.
Again, separate analytics tools and disconnected execution are the issue. Enterprises are unable to close the loop because their frontline business applications have nothing to do with their macro level analysis.
True measurement eliminates guesswork. A modern data management platform, with data-driven applications integrated, can ensure that their investment in third-party sourced data is good, correlate information with internal sources and audit and clarify how data is used from sourcing all the way to delivered result.
The added bonus of this closed loop is that recommended actions can continuously improve through deep learning, making the next set of predictions and prescriptive actions even more likely to yield the right outcomes.
Companies still are struggling to get the right data in the hands of the right users at the right time. A new breed of enterprise data-driven applications built on a modern data management foundation can bring together reliable data, relevant insights and recommended actions under one umbrella.
Rather than IT spending time managing data lakes, companies can deliver immediately consumable, purified insights direct to their business teams.
What can you do today?
If you are in IT and looking at or invested in the following technologies, this is what you can do:
• Master Data Management (MDM): These tools have been around for 10+ years and continue to offer MDM as a separate siloed discipline, requiring complex IT infrastructure and processes, leading to months and years of design and implementation, before reliable data can be made available to business users. Because the delivery latency of the information leads to stale data, business users typically have access to more up-to-date information as a course of their day-to-day operations. These legacy MDM tools do not allow them to provide feedback in a collaborative and controlled manner, effectively wasting valuable intelligence and competitive advantage. Business users and IT teams become equally frustrated, viewing MDM as a promise unfulfilled.
• Big Data Infrastructure and Tools: Ones such as Hadoop, HBase, Cassandra, Graph Databases, MapReduce, Spark and the continuous stream of new technologies designed to handle ever increasing volume, variety and velocity, have changed the way data can and should be managed. However, stitching together all of the pieces required for a complete end-to-end offering is a complex undertaking. Creating big data lakes, and integrating master data management, relationship discovery and other disciplines to support a wide variety of business needs across an enterprise is a multi-year endeavor. Not all organizations have the appetite, resources, skills or budget to integrate, deploy and maintain such infrastructure, while keeping up with an evolving technology landscape.
Action: Look beyond MDM and use modern data management platforms that can handle not just master data but transactional/activity data, third-party, social and public data. Ensure that business users can contribute their insights and add value to the quality of data through everyday activities. Combine that with the benefits of elastic pay-as-you-go via the cloud. Business users should not just be the only constituents to benefit from easy-to-use interfaces such as LinkedIn and Facebook. IT should have a platform that makes managing data as easy as click-and-drag, without having to pop the hood and continuously fine-tune the engine and other lower-level moving parts. Data-as-a-service, which provides on-demand access to third-party data, should also be a prerequisite. Moving forward real-time procurement and refresh of data should be as easy as if you were shopping for data on Amazon.
If you are in marketing or responsible for giving frontline teams applications to help them do their jobs better, this is what you can do:
• BI and Analytics tools are powerful and sophisticated. They empower certain data scientists and a select group of business users to get faster access to information on their own. Convergence of data at scale from multiple sources is now possible, however the data is not guaranteed to be clean and accurate prior to analysis. Looking at unreliable data may lead to wrong conclusions. Although many of these tools claim to be for non-data scientists, they are still beyond the skill of frontline field teams as they go about their day-to-day operations. As previously discussed, these tools still require action to be taken in separate siloed operational applications, and do not automatically correlate results back to the analysis performed.
• Horizontal Packaged Business Applications such as CRM, ERP, HR and financials are widely deployed for core business processes. However, legacy design and the process-driven nature of such applications provide a ridged structure that is hindering business agility. Many end-users question why in the age of consumer apps such as Facebook and LinkedIn, they are still stuck with enterprise applications that required labor intensive manual data entry, jumping between applications to get the complete view they need, and having to sift through complex patterns of information. A new breed of enterprise data-driven applications can offer relevant insights and recommended actions specific to their daily operations, allowing them to significantly improve productivity and outcomes.
Action: Look for data-driven applications that include the visualization and analytics that are tailored for your business users. These applications should complement your process-driven legacy CRM and ERP apps. Failing that, determine how you can close the loop and integrate your analytics with your downstream applications so that the guesswork is eliminated.
Ramon Chen is responsible for worldwide marketing a Reltio, a provider of enterprise data-driven applications with modern data management. Prior to Reltio, he was VP of Product Marketing for Veeva Network at Veeva Systems. He has over 25 years of experience running marketing and product management teams and is well versed in the challenges facing cable and media companies having been at MetaTV, one of the companies that pioneered interactive television, prior to their acquisition by Comcast.