Harvard Business Review, Article #3, Why IT Fumbles Analytics

HBR  
Why IT Fumbles Analytics
by Donald A. Marchand and Joe Peppard Jan 2013
http://hbr.org/2013/01/why-it-fumbles-analytics/ar/1

Preface

• Massive amounts of data now available from internal and external sources
• Companies are spending heavily on IT tools and hiring data scientists
• They treat their big data and analytics projects the same way they treat all IT projects 
• Not realizing that the two are completely different animals

Conventional approach to an IT project

• Focuses on building and deploying the technology :
• On time 
• To plan
• Within budget
• The information requirements and technology specifications are established up front, at the design stage
• This approach works fine if the goal is to improve business processes
• Such projects improve efficiency, lower costs, and increase productivity,executives are still dissatisfied. 
• The reason: Once the system goes live, no one pays any attention to figuring out how to use the information it generates to make better decisions or gain deeper

Where is the problem?

• It’s crucial to understand how people create and use information
• Project teams need members well versed in the cognitive and behavioral sciences, not just in engineering, computer science, and math
• Deploying analytical IT tools is relatively easy. Understanding how they might be used is much less clear
• No one knows the decisions the tools will be asked to support 
• No one knows the questions they will be expected to help answer
• Therefore, a big data or analytics project can’t be treated like a 
• conventional
• large IT project 
• with its defined outcomes 
• required tasks
• detailed plans    

Five guidelines for taking this voyage of discovery

• Place People at the Heart of the Initiative

• WRONG : The logic behind many investments in IT tools and big data initiatives is that giving managers more high-quality information :
• More rapidly will improve their decisions 
• Help them solve problems 
• Gain valuable insights
• Because: It ignores the fact that
• Managers might discard information no matter how good it is
• They have various biases
• They might not have the cognitive ability to use information effectively
• The reality is that many people, including managers, are uncomfortable working with data
• It must place users, the people who will create meaning from the information, at its heart
• How they do or do not use data in reaching conclusions and making decisions

• Emphasize Information Use as the Way to Unlock Value from IT

• People don’t think in a vacuum; they make sense of situations on the basis of their own knowledge,mental models, and experiences
• They also use information in different ways, depending on the context
• An organization’s culture, can frame how people make decisions, collaborate, and share knowledge
• People use information dynamically and iteratively
• The design of most IT systems takes into account the data that have been identified as important and controllable
• That approach is fine for activities that are highly structured and whose tasks can be precisely described
• It is ideal for moving information from the human domain into the technology domain
• The problem is that many organizations mistakenly apply same design philosophy to the task of getting data out of the technology domain and into the human domain 
• Analytics projects succeed by challenging and improving the way: 
• Information is used
• Questions are answered
• Decisions are made

• Here are some ways to do this:
• Ask second-order questions, that is, questions about questions
• Discover what data you do and do not have
• Give IT project teams the freedom to reframe business problems

• Equip IT Project Teams With Cognitive and Behavioral Scientists
• Most IT professionals have engineering, computer science, and math backgrounds
• They are generally very logical and are strong process thinkers
• Big Data needs people who understand how people:
• Perceive problems 
• Use information 
• Analyze data in developing solutions ideas and knowledge
• This shift  economics to behavioral economics
• Organizations that want employees to be more data oriented in their thinking and decision making must train them to know 
• When to draw on data 
• How to frame questions
• How Build hypotheses 
• How conduct experiments
• How interpret results

• Focus on Learning
• Big data and other analytics projects are more akin to scientific research and clinical trials than to IT initiatives
• In short, they are opportunities for discovery
• How make learning a central focus of big data and analytics projects:
• Promote and facilitate a culture of information sharing
• Expose your assumptions, biases, and blind spots
• Strive to demonstrate cause and effect
• Identify the appropriate techniques and tools

• Worry More About Solving Business Problems than About Deploying Technology
• Conventional IT project management is risk-averse
• Big data should focus less on managing the risks of deploying technology and more on solving business problems
• The paradox is that the technologies that were supposed to help manage data are now causing a massive deluge