1 Data Management for BI Big Data, Bigger Insight, Superior Performance January 2012 Michael Lock
2 Page 2 Executive Summary The term "Big Data" has come to represent the increasing challenges companies face in managing the growth and complexity of organizational data. An ever-increasing number of industries and job roles struggle mightily to drink from the proverbial fire hose of data to produce something meaningful and useful for the business. Best-in-Class companies successfully bridge the gap between their expanding sources of data and the analytical processes and systems they use to transform that data into timely business insight. Supported by a strong foundation of organizational maturity and leveraging the right technologies, these top performers have improved the accessibility, reliability, and timeliness of critical business information. This research is based on feedback from 247 executives across the globe. Research Benchmark Aberdeen s Research Benchmarks provide an in-depth and comprehensive look into process, procedure, methodologies, and technologies with best practice identification and actionable recommendations Best-in-Class Performance Aberdeen used the following four key performance criteria to distinguish Best-in-Class companies: 21 days required to integrate new data sources, compared with 53 days for the Industry Average and 130 days for Laggards 32% year over year increase in accessible business data, compared with a 16% increase for the Industry Average and 6% for Laggards 91% of business data is considered to be accurate, compared with 80% for the Industry Average and 54% for Laggards 92% of key business information is delivered on-time, compared with 75% for the Industry Average and 39% for Laggards Competitive Maturity Assessment Survey results show that the firms enjoying Best-in-Class performance are: 60% more likely than Laggards to report a data-driven culture Twice as likely as all other companies to maintain and update data definitions centrally and continuously 1.6-times more likely than Average companies to use data quality tools Required Actions In addition to the specific recommendations in Chapter Three of this report, to achieve Best-in-Class performance, companies must: Elevate data management initiatives to the executive level Improve ability to measure time-to-information Evaluate data quality technologies This document is the result of primary research performed by Aberdeen Group. Aberdeen Group's methodologies provide for objective fact-based research and represent the best analysis available at the time of publication. Unless otherwise noted, the entire contents of this publication are copyrighted by Aberdeen Group, Inc. and may not be reproduced, distributed, archived, or transmitted in any form or by any means without prior written consent by Aberdeen Group, Inc.
3 Page 3 Table of Contents Executive Summary... 2 Best-in-Class Performance... 2 Competitive Maturity Assessment... 2 Required Actions... 2 Chapter One: Benchmarking the Best-in-Class... 4 Context - Connecting "Big Data" with Analytics... 4 The Maturity Class Framework... 5 The Best-in-Class PACE Model... 6 Best-in-Class Strategies... 7 Chapter Two: Benchmarking Requirements for Success... 9 Competitive Assessment Capabilities and Enablers Chapter Three: Required Actions Laggard Steps to Success Industry Average Steps to Success Best-in-Class Steps to Success Appendix A: Research Methodology Appendix B: Related Aberdeen Research Figures Figure 1: Unique Data Sources Used for Business Analysis... 4 Figure 2: Top Pressures Driving Data Management Initiatives... 5 Figure 3: Top Strategic Actions to Support Data Management... 7 Figure 4: Who Owns Data Management / Governance?... 8 Figure 5: Process and Organizational Capabilities Figure 6: Knowledge and Performance Management Capabilities Figure 7: Best-in-Class Technology Enablers Figure 8: Familiarity with Apache Hadoop Figure 9: Big Data / Advanced Analytics Technologies in Use Figure 10: Best-in-Class Enable User Satisfaction Tables Table 1: Top Performers Earn Best-in-Class Status... 6 Table 2: The Best-in-Class PACE Framework... 6 Table 3: The Competitive Framework Table 4: The PACE Framework Key Table 5: The Competitive Framework Key Table 6: Relationship Between PACE and the Competitive Framework... 20
4 Page 4 Chapter One: Benchmarking the Best-in-Class Context - Connecting "Big Data" with Analytics Today's top organizations flourish on the shoulders of a sustainable business model, a tangible and addressable target market, a quality corps of people to carry out the overall strategy, and on data. Oil grants a car's engine fluidity, engine health, and long-term reliability; clean, timely, and relevant information is the oil that supports efficient business and facilitates longterm market stability. Unfortunately for many organizations in today's fastpaced technology-driven business environment, the influx of raw data in all its forms creates a surplus of data and a shortage of insight--too much crude, not enough refined engine oil. Aberdeen's research shows that data is growing in both volume and complexity. The topic of Data Management for Business Intelligence has been covered extensively in Aberdeen's research. Benchmark reports in 2009 and 2010 demonstrated that companies were seeing an average year over year growth in data volume of 29% and 40% respectively. Aberdeen's 2011 research found that organizations saw an average 38% increase in data volume over the previous 12 months. The average company confronts 2.5- times more data than it did three years ago--an increase that might seem laughably small for particularly data-driven companies. However, data is growing in complexity and variety as well as volume. Between data warehouses, data marts, enterprise applications, spreadsheets and external unstructured or social data, companies are drawing on an increasing number of unique data sources to drive their business analysis. Aberdeen research shows that over the past three years, the number of unique data sources that companies manage is also increasing (Figure 1). Fast Facts Best-in-Class companies achieved: 19% reduction in timeto-information over the previous year Compared with: 5% for the Industry Average 4% for Laggards Figure 1: Unique Data Sources Used for Business Analysis Avg. # of Data Sources SMB (<1,000 Employees) Enterprise (1,000+ Employees) n = 834 Source: Aberdeen Group, December 2009, December 2010, December 2011 The term "Big Data" has recently grown in prominence as a way of describing this phenomenon - growth in data volume, complexity, and disparity. Definitions and thresholds for what actually constitutes "Big Data" have been offered, but for the average employee that relies on data to support business decisions, these definitions are meaningless. Capturing data, storing it, and managing it, are serious challenges for most
5 Page 5 organizations, but where the rubber really meets the road is transforming data into meaningful and usable business insight. Many companies are trying to feed their business intelligence (BI) systems and analytical processes with better data to provide timely insight. However, the influx of data presents many barriers to effective analytics, and to the creation of business insight for most decision makers. Whether their data is inaccessible, fragmented, or simply unwieldy from a volume perspective, companies are seeking formalized data management strategies in response. Late delivery of information is the top pressure driving Aberdeen's respondents to develop their data management initiatives (Figure 2). Finding that "needle in the haystack" of business insight is difficult, but delivering it to a decision maker in time to positively impact a business decision is harder still. "Lack of trust in data drives a downward spiral. No trust in data leads to no time spent updating data, which leads to a degradation in data quality, which leads to even less trust in the data. ~ Quality Management Executive Large U.S. Oil & Gas Company Figure 2: Top Pressures Driving Data Management Initiatives Too much crucial business information is delivered late 53% Growing data is inaccessible or underutilized for business analysis Data is too fragmented / 'siloed' to develop a clear picture of the business The volume of data is growing too rapidly for our current data infrastructure 40% 47% 46% All Respondents The Maturity Class Framework 35% 45% 55% Percentage of Respondents, n = 247 The ability to exploit business data and extract the insights hidden within is predicated on the ability to transform raw data to timely insight. Best-in- Class companies from this study were identified based on their ability to bring new data sources into their analytical infrastructure quickly, make more of these data sources accessible, provide more accurate business data, and deliver insight within the time required to take action. Aberdeen determined Best-in-Class performance based on the following four key performance indicators: Data On-Boarding Efficiency: Measured as an average number of days required to integrate new data sources into the organization s information infrastructure Data Accessibility: Measured as an average year over year increase in the amount of business data that is discoverable / searchable
6 Page 6 Data Reliability: Measured as an average percentage of business data considered to be reliable and accurate On-Time Information Delivery: Measured as an average percentage of actionable business information delivered on-time, or within the required decision window Table 1: Top Performers Earn Best-in-Class Status Definition of Maturity Class Best-in-Class: Top 20% of aggregate performance scorers Industry Average: Middle 50% of aggregate performance scorers Laggard: Bottom 30% of aggregate performance scorers Mean Class Performance 21 days required to integrate new data sources 32% year over year increase in accessible business data 91% of business data is considered to be accurate 92% of key business information delivered on-time 53 days required to integrate new data sources 16% year over year increase in accessible business data 80% of business data is considered to be accurate 75% of key business information delivered on-time 130 days required to integrate new data sources 6% year over year increase in accessible business data 54% of business data is considered to be accurate 39% of key business information delivered on-time Fast Facts 59% of Best-in-Class companies report that they use a significant amount of unstructured data Compared with: 42% of all other companies The Best-in-Class PACE Model Using BI and data management solutions to achieve corporate goals requires a combination of strategic actions, organizational capabilities, and enabling technologies that can be summarized as shown in Table 2. Table 2: The Best-in-Class PACE Framework Pressures Actions Capabilities Enablers Business users do not get the information they need in a timely manner Create a long-term strategy roadmap for managing Big Data Broaden accessibility of data across functions / departments Executive-level support for better data oversight Data analyst / data scientist role to analyze internal business data Formal in-house development of analytical knowledge and skill sets Ability to assess data needs from all different departments / functions Traditional Business Intelligence (BI) platform Data query / discovery tools Data warehouse software Data quality tools (cleansing, enrichment, normalization, etc.) Interactive/drill down reports
7 Page 7 Best-in-Class Strategies To address the business pressures depicted in Figure 2, top performing companies are adopting several strategic actions. First, as data continues to balloon, these companies are developing a long term plan to gather data more effectively, filter out unnecessary noise, and extract business insight faster. They are also collecting feedback and requirements for data access from end users, in order to make that data accessible, transparent and useful to the entire organization. Best-in-Class, Industry Average and Laggard organizations all give the same priority to these strategies (Figure 3). Figure 3: Top Strategic Actions to Support Data Management "Business intelligence is coming of age. It is no longer enough to analyze and report on what happened. BI needs to inform, influence and support decisions on what IS HAPPENING. BI and Big Data are for us an integral line-of-business system" ~ Sales Executive Mid-Sized European IT Consultancy Create a long-term strategy roadmap for managing Big Data Broaden accessibility of data across functions / departments Gather and incorporate end-user requirements for data Identify key data sources required for analysis 37% 40% 37% 36% 36% 45% 50% 51% 25% 35% 45% 55% Best-in-Class Average and Laggard Percentage of Respondents, n = 247 However, underperformers diverge from the Best-in-Class in their focus on identifying data sources required to support business analysis. As new data sources come online and more employees seek access to them, the identification and prioritization of these data sources becomes paramount. Best-in-Class companies place a lower priority on this strategy largely because they have already engaged in this important activity. Aberdeen Insights Strategy The concept of data growth and complexity hits home not just for IT managers, but for other Line-of-Business leaders as well. Between core HR data, customer intelligence, supply chain data, and other forms of swelling corporate data, many functions are looking for ways to better manage big data. Given the wide ranging implications of data growth, who owns the data strategy and activity? Who is ultimately responsible for the management and governance of corporate data? We posed this question to our survey respondents and, overwhelmingly, companies say Continued
8 Page 8 Aberdeen Insights Strategy that IT is ultimately responsible (Figure 4). Figure 4: Who Owns Data Management / Governance? IT 48% Senior Executive 23% Cross-Functional Team Line-of-Business 10% 14% No One 5% All Respondents 0% 10% 20% 30% 40% 50% Percentage of Respondents, n = 247 However, while the majority of companies rely on IT to take ownership of data management, the research showed that an interesting variety of stakeholders may now have a say. Nearly a quarter of companies surveyed report that data management is under the oversight of a non-it senior executive, perhaps the COO or VP of marketing, depending on the organization. Increasingly, though, companies are pushing this responsibility either to the lines-of-business themselves (marketing, customer service, supply chain, etc.) or to a cross functional team that represents line-of-business stakeholders. This is likely due to big data's increasingly broad applicability. To maintain the integrity of customer data, or web-based "clickstream" data, or HR records, whatever the case may be, some companies are decentralizing the burden of data management and putting it squarely on the shoulders of LoB stakeholders. In the next chapter, we will see what the top performers are doing to achieve these gains.
9 Page 9 Chapter Two: Benchmarking Requirements for Success Selecting an appropriate data management solution and integrating it with business process management systems can help an organization turn the above strategies into profit. The following case study illustrates how one social gaming organization was able to apply advanced analytics to its growing volumes of data in order to identify and act upon opportunities for revenue and profit growth. Case Study Zynga In today s business climate, social networking impacts just about every type of organization. Social networking platforms like Facebook and Twitter, and social gaming companies like Zynga, have changed the way many organizations are marketing and positioning their products and services. Zynga is the world s largest social games developer, with a stated mission to connect the world through games. Every day, millions of people interact with their friends and express their unique personalities through Zynga games, which range from harvesting crops to slicing apples to playing poker. From its founding in July of 2007, Zynga has built a user base of over 250 million monthly active users through a combination of vision and the ability to rapidly deploy new games and ingame features. While the concept of social gaming is not new games are generally about being social today we use the term to describe games that run on social networking platforms such as Facebook and MySpace, as well as on mobile devices like smart phones and tablets. There are three key metrics that drive the economics of social gaming; churn rates defined as the loss rate of game players, the viral coefficient a measure of how effective current game players are at drawing new players, and revenue per user an estimate of the lifetime revenue that a game player will generate. Changing these numbers in even a small way on a per-user basis can make the difference between a game which earns $5 profit per user, and $50 profit per user, so understanding these numbers, and influencing them in their favor, is a top priority for social gaming firms. But this is easier said than done. The ability to handle extremely large quantities of rapidly changing data is critical when creating massively multi-user, real-time applications. Zynga realized quickly that a scalable advanced analytics platform was needed to give them the level of performance and user experience called for by their game designers. Zynga ultimately settled upon a solution that utilizes a native columnar storage and execution engine in conjunction with a massively parallel processing (MPP) architecture. In the words of Dan McCaffrey Director of Analytics Engineering at Zynga, Our Continued Fast Facts Best-in-Class companies achieved: 21% reduction in timeto-decision over the previous year Compared with: 4% for the Industry Average 0% for Laggards
10 Page 10 Case Study Zynga analytical platform has MPP and scale solved. We can process data daily to produce an optimized graph. Zynga also saw substantial improvement in processing speed. With respect to performance and scalability, McCaffrey adds being able to run this on tables with tens of billions of rows of data with a fast turnaround is amazing. With this solution, the Zynga team has been able to improve the targeting of items such as gifts an important type of game interaction effectively increasing the level of interaction between the active players while minimizing spam to the passive players. Also, they have begun to build an item of tremendous value a graph of their active social gaming network. Having mastered the basic graph analysis challenge, the Zynga team is already looking towards the next challenge. As they expand their graph cluster, they will begin to enable capabilities to identify clusters of users with like behavior or common paths which will enable even better targeting of game related promotions and activities. Competitive Assessment Aberdeen Group analyzed the aggregated metrics of surveyed companies to determine whether their performance ranked as Best-in-Class, Industry Average, or Laggard. In addition to having common performance levels, each class also shared characteristics in five key categories: (1) process (the approaches they take to execute daily operations); (2) organization (corporate focus and collaboration among stakeholders); (3) knowledge management (contextualizing data and exposing it to key stakeholders); (4) technology (the selection of the appropriate tools and the effective deployment of those tools); and (5) performance management (the ability of the organization to measure its results to improve its business). These characteristics serve as a guideline for best practices, and correlate directly with Best-in-Class performance across the key metrics (Table 3). Table 3: The Competitive Framework Process Organization Best-in-Class Average Laggards Executive-level policies supporting better data oversight 61% 50% 46% Ability to assess data needs across departments/functions 57% 35% 22% Decision culture that values the use of supporting data 69% 46% 43% Dedicated internal data analyst / data scientist role 55% 44% 32%
11 Page 11 Knowledge Performance Technology Best-in-Class Average Laggards Formal in-house development of analytical skill sets 66% 51% 43% Data definitions continuously maintained and updated 54% 32% 23% Data quality issues are continuously captured / monitored 44% 25% 24% Ability to measure 'time-to-information' for end users 42% 21% 10% Traditional business intelligence (BI) platform 81% 61% 60% Data query / discovery tools 78% 63% 55% Data quality tools (cleansing, enrichment, normalization) 59% 37% 34% Capabilities and Enablers Based on the findings of the Competitive Framework and interviews with end-users, Aberdeen s analysis of the Best-in-Class demonstrates that the successful deployment and use of a data management strategy depends on a combination of specific capabilities and technology enablers. Aberdeen's research has identified several capabilities that Best-in-Class companies use to achieve elevated performance. Process Organizations have succeeded at delivering cleaner information to their analytical systems by making these activities a top priority. Cultural shifts don't happen overnight, but with data management, executive sponsorship can push companies in the right direction. 61% of Best-in-Class companies report having executive-level policies for better data oversight. These policies need to filter down to Line-of-Business decision makers, who have more tactical responsibility. To increase the effectiveness and adoption of these policies, companies must understand the data needs of each department. Best-in-Class organizations are more than twice as likely as Laggards to have procedures in place to assess data needs across multiple departments or business functions (Figure 5).
12 Page 12 Figure 5: Process and Organizational Capabilities Percentage of Respondents 75% 50% 25% 0% 57% 35% 22% Ability to assess data needs across multiple departments 69% 46% 43% Decision-making culture that values the use of supporting data Best-in-Class Industry Average Laggard n = 247 Fast Facts 57% of Best-in-Class are using text analytics Compared with: 35% of the Industry Average 33% of Laggards Organization Often, what separates a top performer from the rest of the pack is the ability to escort the right policies and procedures from their starting point as misunderstood corporate mandates into broad cultural adoption. Bestin-Class companies use executive managers and tactical process capabilities to transform one-dimensional policies into ingrained corporate culture. The Best-in-Class are 63% more likely than all others to report a culture that values the use of data to support crucial business decisions (Figure 5, above). Top performing companies are also more likely to have created or hired a dedicated role (or roles) specifically responsible for understanding and using data. Top performers are 84% more likely than Laggards to have such an internal data analyst or data scientist role. Knowledge Management Top performers are also more adept at sharing analytical knowledge to boost the company's overall level of analytical intelligence. Research shows that Best-in-Class companies are 38% more likely to have formal in-house development of analytical skill sets. Many companies, once they have increased their analytical skills, are challenged to use these capabilities in the face of constantly growing and changing corporate data. Best-in-Class companies are twice as likely as all other organizations to create, update, and continuously maintain data definitions centrally, giving their growing user base cleaner and more widely understood information (Figure 6). Figure 6: Knowledge and Performance Management Capabilities Percentage of Respondents 55% 30% 5% 54% 32% Data definitions continuously maintained and updated 44% 25% 23% 24% Data quality issues are continuously captured and monitored Best-in-Class Industry Average Laggard n = 247
13 Page 13 Performance Management The central management and continuous updating of data definitions discussed above puts Best-in-Class companies in a more advantageous position to identify and quickly remedy data quality and usability issues. Top performers are 76% more likely than all others to capture, monitor, and improve upon data quality issues as they arise (Figure 6, above). Given the how important timely information is to organizations today (Figure 2, Page 5), organizations should be able to measure time-to-information for endusers in order to benchmark and improve their performance. While fewer than half of Best-in-Class companies report having this capability, these top performers are three-times more likely than all others to measure time-toinformation. Fast Facts 43% of Best-in-Class are using big data specific consultants Compared with: 23% of all other companies Technology Transforming raw data into deliverable insight requires not only strong organizational capabilities, but also judicious use of the right technologies. At the front end of that process - delivering actionable information to key decision makers - Best-in-Class companies are more likely to have a dedicated business intelligence (BI) platform to formalize the capture and analysis of historical data. Leading companies are more likely to use tools to enable self-discovery for business decision makers. Reports with interactivity or drill-down capability let users ask follow-up questions of the data and pursue their own analysis. Similarly, query / discovery tools allow for self-service questions and answers. Rather than relying only on a prescribed report or chart to deliver information, query tools enable users to ask their own questions of the data and generate a clearer picture of the business. Best-in-Class companies are more likely than Average and Laggards to use both these technologies (Figure 7). Figure 7: Best-in-Class Technology Enablers Percentage of Respondents 85% 65% 45% 25% 5% 81% 78% 61% 60% 63% 55% Traditional BI platform Data query / discovery tools 62% 54% 59% 34% 37% 34% Interactive / drill down reports Data quality tools (e.g. cleansing, enrichment) Best-in-Class Industry Average Laggard n = 247 Technology can also be extremely useful in the improvement and maintenance of data quality. In the business world, poor data quality can rear its ugly head in a variety of disruptive ways. Incomplete, duplicated, or corrupted data leads to distrust in the data, and to unnecessary time spent
14 Page 14 redoubling efforts. Best-in-Class companies recognize the danger posed by poor quality data, and are 62% more likely than all others to use tools to improve and maintain the quality of business data (Figure 7, above). Aberdeen Insights Next Generation Technologies Just about every company, regardless of size, is facing some challenge around the volume, velocity, or variety of business data. But certain organizations and industries have particularly draconian requirements when it comes to their need to handle "Big Data". The open-source database Hadoop is an emerging technology purpose-built to address the challenges associated with data volume and scalability. Supported by the community of the Apache Software Foundation, Hadoop's massive scalability (up to multiple petabytes), and its ability to handle data in various native formats, are gaining it recognition as a weapon in the "battle of the bulge" against big data. Hadoop is also becoming more applicable to the business world. Some of the most visible use cases for Hadoop include logistics applications that involve large volumes of constantly changing location-based data (weather, traffic, GPS), webbased or social media applications with high velocity "clickstream" data, and the creation and connection of machine to machine (M2M) transactional data. However, Hadoop still has yet to achieve a strong level of understanding, or even recognition, among a representative sample of survey respondents (Figure 8) Fast Facts For this survey, Aberdeen segmented respondents into one of two categories, described below: Business Users - Nontechnical consumers of cross-functional business data to oversee performance and influence organizational direction Technical Users - IT / data management professionals and "power users", analysts of business data, with knowledge / expertise of supporting technology and data infrastructure Figure 8: Familiarity with Apache Hadoop Business Users 53% 16% 19% 8% 4% Technical Users 24% 21% 35% 13% 7% 0% 25% 50% 75% 100% Never heard of it Understand it, not exploring it We have implemented it already Heard of it, don't understand it Currently evaluating it Percentage of Respondents, n = 247 Comparing business users to technical users (definition in sideabar) reveals a marked lack of understanding with respect to Hadoop. Almost half of technical users have either never heard of Hadoop or don't understand what it does, compared to almost three-quarters of business users. With the influx of data and expanding number of use cases for Hadoop, both business and technical users could benefit from more education around this technology. Continued
15 Page 15 Aberdeen Insights Next Generation Technologies It should be noted however that despite the buzz around Hadoop and it's purported scalability and power, Hadoop isn't necessarily the answer for all big data problems. Often, challenges of data volume and scalability can be addressed with other next-generation advanced technologies either as stand-alone solutions, or in conjunction with Hadoop. A variety of solutions are available to help technically-inclined users address big data. To mitigate the sheer volume of data, companies are looking for more advanced methods for data compression as well as de-duplication. Some companies are also looking to virtualize their storage environments, not necessarily to reduce the amount of data under management, but to minimize and optimize the hardware assets required control that data (Figure 9). Figure 9: Big Data / Advanced Analytics Technologies in Use Data Compression Storage Virtualization 56% 53% Data De-Duplication 37% Column-Oriented Databases Massively Parallel Processing (MPP) 24% 28% Technical Users 0% 20% 40% 60% Percentage of Respondents, n = 125 From an analytical standpoint, some emerging and advanced technologies are designed to improve the performance of complex query algorithms drastically. One such technology starting to gain a foothold among technical users is the column-oriented or "columnar" database, which offers a unique approach to storing and retrieving data that enables substantial performance gains for certain queries. Additionally, some companies leverage massively parallel processing (MPP) architectures that harness the computing power of many (dozens, if not hundreds, or even thousands) processing cores working in conjunction to perform complex computations against extreme data volumes. For those charged with managing and extracting value from big data while allowing for it to scale beyond the capacity of the organization's data infrastructure, these technologies could become increasingly relevant and valuable.
16 Page 16 Chapter Three: Required Actions Whether a company is trying to improve its data management performance from Laggard to Industry Average, or Industry Average to Best-in-Class, the following actions will help spur the necessary performance improvements: Laggard Steps to Success Elevate data quality to the executive level. Time and again, Aberdeen's research finds a strong correlation between having a strong analytical or data-driven culture, and achieving better business performance. Culture is not easy to improve. A reasonable place to start is by proving the value of clean and timely data to the most influential people in the organization - the senior executives - and having those business leaders facilitate the growth of that culture. Best-in-Class companies are 1.3-times more likely than Laggards to create executive level policies supporting clean and relevant data. Finding or creating a champion at the executive level will give more weight to data management initiatives and help Laggards move closer to that data-driven analytical culture. Implement tools for end-user data discovery. A significant amount of value is associated with historical static business reports, charts, and data views. These offer a perspective into how the business is performing. The trouble with many of such reports is that they are static, stifling any follow-up curiosity. Best-in-Class companies nurture business decision makers' creativity and desire for discovery. Only 34% of Laggards have interactive or drill-down capability within their existing reporting tools. By investing in technologies focused on discovery, rather than static consumption of data, Laggards will enable curious business users to ask better questions and identify salient business opportunities. Industry Average Steps to Success Take stock of end-user needs. Most organizations have a wide range of information needs across the company. Requirements vary by data type - structured, unstructured, raw text, image, video - and access speed. Distribute data management technology and organizational capability appropriate to the needs of each department. Whatever investment is made in people and technology should be delivered judiciously, based on need. Without understanding those needs, companies can burn through money quickly and never achieve the ROI they seek. Best-in-Class companies are 63% more likely than the Average to assess data requirements across different departments within the company. Evaluate data cleansing/hygiene technologies. Given the significant investment many companies make in BI / analytical tools, Fast Facts 39% of Best-in-Class companies report that they need actionable information within real time or nearreal time Compared with: 20% of all other companies "The biggest concern falls around issues of trusting the data. The need to manually manipulate the data in order to achieve the desired results does not allow for an easily repeatable process and once the business users change, the new users do not trust what those before them did to arrive at the results being reported out." ~ Procurement Manager Large U.S. Based Pharmaceutical Company
17 Page 17 it stands to reason that they should invest proportional time, effort and resources in the quality of the data those systems use. Dirty or corrupted data can skew business analyses to the point of leading a company down the wrong path. Data integration and cleansing technologies are instrumental to improving data quality. Only 29% of Industry Average companies are using data cleansing to improve data hygiene. Combined with the right processes and organizational capabilities, data cleansing tools will increase the value of organizational data and pave the way for informed business decisions. Best-in-Class Steps to Success Prioritize access to high-demand data. Best-in-Class companies are typically strong at prioritizing data for user access. But with the acceleration of data growth and the increasing difficulty of capturing the right data for analysis, Best-in-Class companies have fallen behind. Whether it's end-of-period financial reporting data, customer intelligence data, or external unstructured data, many organizations face time-sensitive situations where certain data will be in high-demand. Aberdeen's research shows that only a third of Best-in-Class companies prioritize access to data. The ability to prioritize that data and enable role-based access reduces unnecessary use of the data infrastructure, and paves a smooth path for crucial data-driven activities to be completed in a timely way. Improve ability to measure time-to-information. Best-in- Class companies have also been superior, historically, in their insistence on measuring and improving time-to-information. The time factor one of the most important elements playing into analysis. The most insightful business information is useless if isn't delivered in time to align resources and exploit the opportunity. Delivering insight within the decision window is key to performance--and it all starts with measuring information delivery time. In this study, only 42% of Best-in-Class companies have the ability to measure time-to-information for end-users. By returning to their roots and developing this capability, top performers will be able to deliver more insight within the required window, and exploit more business opportunities as a result.
18 Page 18 Aberdeen Insights Summary Unlike gold, frozen orange juice or crude oil, information has no inherent value. It isn't traded on any exchange. Its value depends on an organization's ability to transform it into business insight. A faster transformation of data to insight results in more satisfied users, and the research shows that Best-in-Class companies have a high level of satisfaction in three crucial aspects of data management (Figure 10). Figure 10: Best-in-Class Enable User Satisfaction Satisfied or Very Satisfied Response time to self-service queries Quality / relevance of accessible info Information system ease of use 19% 17% 13% 52% 45% 63% 37% 77% 77% 0% 20% 40% 60% 80% Best-in-Class Industry Average Laggard Percentage of Respondents, n = 247 By maturing their organization's approach to data management, and creating an environment in which end-user curiosity can flourish, Best-in- Class companies can make information more accessible, relevant, intuitive, and timely. Best-in-Class organizations don't capture data just for the purpose of compliance, or to hop on the bandwagon of "Big Data", but rather to create useful business insight, positioning themselves for enhanced performance.
19 Page 19 Appendix A: Research Methodology Between November and December 2011, Aberdeen examined the use, the experiences, and the intentions of 247 enterprises using data management tools and strategies in a diverse set of industries and use cases. Aberdeen supplemented this online survey effort with interviews with select survey respondents, gathering additional information on data management strategies, experiences, and results. Responding enterprises included the following: Job title: The research sample included respondents with the following job titles: CEO / President (23%); EVP / SVP / VP (17%); Director (21%); Manager (20%); Consultant (14%); and other (5%). Department / function: The research sample included respondents from the following departments or functions: senior management (11%); sales and marketing (23%); IT manager or staff (35%); operations manager (8%); logistics and procurement (6%) and other (17%). Industry: The research sample included respondents from a variety of industries. The largest segments represented were: IT consulting/services (22%); high tech/software (16%); financial services (11%); and education/public sector (8%). Geography: The majority of respondents (60%) were from North America. Remaining respondents were from the Asia-Pacific region (12%) and EMEA (28%). Company size: Thirty percent (30%) of respondents were from large enterprises (annual revenues above US $1 billion); 27% were from midsize enterprises (annual revenues between $50 million and $1 billion); and 43% of respondents were from small businesses (annual revenues of $50 million or less). Headcount: Forty-three percent (43%) of respondents were from large enterprises (headcount greater than 1,000 employees); 25% were from midsize enterprises (headcount between 100 and 999 employees); and 32% of respondents were from small businesses (headcount between 1 and 99 employees). Study Focus Responding executives completed an online survey that included questions designed to determine the following: The degree to which data management tools and strategies are deployed in their operations and the financial implications of the technology The structure and effectiveness of existing data management implementations Current and planned use of data management to aid operational and promotional activities The benefits, if any, that have been derived from data management initiatives The study aimed to identify emerging best practices for data management, and to provide a framework by which readers could assess their own capabilities.
20 Page 20 Table 4: The PACE Framework Key Overview Aberdeen applies a methodology to benchmark research that evaluates the business pressures, actions, capabilities, and enablers (PACE) that indicate corporate behavior in specific business processes. These terms are defined as follows: Pressures external forces that impact an organization s market position, competitiveness, or business operations (e.g., economic, political and regulatory, technology, changing customer preferences, competitive) Actions the strategic approaches that an organization takes in response to industry pressures (e.g., align the corporate business model to leverage industry opportunities, such as product / service strategy, target markets, financial strategy, go-to-market, and sales strategy) Capabilities the business process competencies required to execute corporate strategy (e.g., skilled people, brand, market positioning, viable products / services, ecosystem partners, financing) Enablers the key functionality of technology solutions required to support the organization s enabling business practices (e.g., development platform, applications, network connectivity, user interface, training and support, partner interfaces, data cleansing, and management) Table 5: The Competitive Framework Key Overview The Aberdeen Competitive Framework defines enterprises as falling into one of the following three levels of practices and performance: Best-in-Class (20%) Practices that are the best currently being employed and are significantly superior to the Industry Average, and result in the top industry performance. Industry Average (50%) Practices that represent the average or norm, and result in average industry performance. Laggards (30%) Practices that are significantly behind the average of the industry, and result in below average performance. In the following categories: Process What is the scope of process standardization? What is the efficiency and effectiveness of this process? Organization How is your company currently organized to manage and optimize this particular process? Knowledge What visibility do you have into key data and intelligence required to manage this process? Technology What level of automation have you used to support this process? How is this automation integrated and aligned? Performance What do you measure? How frequently? What s your actual performance? Table 6: Relationship Between PACE and the Competitive Framework PACE and the Competitive Framework How They Interact Aberdeen research indicates that companies that identify the most influential pressures and take the most transformational and effective actions are most likely to achieve superior performance. The level of competitive performance that a company achieves is strongly determined by the PACE choices that they make and how well they execute those decisions.