Data Strategy

Data Strategy

By: Bertnard Marr

Allison’s Rating:

Data Strategy by Bernard Marr provides a high-level data strategy summary that is that is accessible to large corporations and small businesses alike. Marr concedes that data collection is not a new concept, but suggests that with the genesis of big data (structured and unstructured) has ushered in a new era of analytics. Using the familiar 4Vs of big data we learned in KMBI (volume, velocity, variety and veracity) Marr introduces his idea of the 5th “V” – Value. The book concentrates on how to maximize this value by turning data into insights. When reading Data Strategy there was one clear theme throughout the entire book – a solid data strategy is paramount when using data to drive decisions. Regardless of the industry (or the size of the company) if there is not a culture of acceptance, the benefits from data will not be realized. Marr explains that a responsible data strategy will lead to business decisions that are rooted in data, and that the ability for a company to succeed is “increasingly driven by how well it can leverage its data” (pg. 16).

A high-quality data strategy is comprised of two parts: a culture of data and a focused business goal. A culture of data includes providing employees with tangible benefits of data, allowing for them to see for themselves how data will better their job. Marr claims that communication is critical and this will help grow a “business understanding [of] the value of data and how it can help the business success” (pg 172). This means that the strategy must be clear in its expectations and also on the timeline of execution. A focused business goal will help companies find value in the immense amount of data being collected. By defining business critical questions that should be prioritized, drive a focused work effort that is aligned from the CEO to the individual contributors performing the analysis. This corporate narrative helps to set the tone and encourages employees to deliver key insights and trends.

After a company establishes a strong data strategy and has the ability to incorporate data into decisions they gain competitive advantage in decision making, operations and monetization.

Decision Making

Marr defines strategic decision making as, “anything that moves the organization closer to achieving its strategic goals” (pg. 37) and suggests that data provides the insights necessary to do just that. The value of data within companies is gleaned from the visualization and communication of the data. It is critical that the data collected is shared with the appropriate audience, and that they have access to the most up to date information. An example of this was shared with Dickey’s BBQ providing a real-time dashboard allowing for adoption to changes in the environment. Taking this idea to my workplace I can see this concept in action. Within the e-commerce group we have a large TV display that displays a dashboard from Google Analytics that is updated in real-time – showing the predicted e-commerce budget and a comparison for the day’s web traffic. This leads to better business decisions within our development team, providing easy insight into question such as “what time is best for a release today?”. With access to data employees are armed with the ability to make more confident decisions that drive real business value.

Operations

Incorporating data based operations into daily activities helps a business to run more smoothly. Marr classifies data based operations into two major categories:

1) Day to Day Business Optimization: Using information gained from sensors, machine output or customer input to make every day processes more efficient. The area that stood out to me the most in this area was the discussion regarding big data analytics and forecasting in retail. The ability to efficiently optimize prices is critical, and helps with both conversion rate and inventory management. The example provided by Macy’s removing their end of year sale had me reflect on Brooks’ sale practices and how we might incorporate sale optimization from the data gained from Google Analytics. Additionally, at Brooks we are currently implementing a recommendation engine on our website. It was interesting to read about how Amazon has built their recommendation engine and the amount of data that goes into making one recommendation! While Brooks will leverage a third party to analyze the data and produce recommendations I am interested in learning more about what factors will be important for Brooks’ recommendations.

2) Improving Customer Offerings: The ability to use data to improve customer service or products will help a company gain a competitive advantage. The John Deere customer portal example provided by Marr demonstrates how a company can think creatively, seeking trends from the data own - even if it is outside of their main business operations. In the section of data driven operations, this was almost exactly in line with my research paper for the KMBI class in regard to data-driven product development. Marr states that “all businesses are becoming data businesses” which could not be more in line with the research that I conducted. Even at Brooks we have a Business Intelligence team and are creatively seeking solutions to the question of “who are the runners in different communities?”. While Brooks isn’t quite creating shoes with artificial intelligence, there are development discussions about using data to deliver an individualized runner shoe. These actions solidify for me Marr’s claims that there is a “huge potential to improve product offerings”, even in industries that I might least suspect.

Data Monetization Data for datas sake will not increase a company’s bottom line. However, when data is collected with a key purpose it will increase the overall value of a company. Additionally, companies that collect large amounts of data have the option of selling that data to companies that are seeking more data. An example provided is the credit card industry, and how American Express provides online analysis for companies with the data generated by end users. One that hit particularly home for me is how Uber sells data to interested parties such as hotels and airlines. As a frequent Uber rider, it makes me think twice about how the data is being used. However, this quarter I heard a podcast from Freakonomics, “Why Uber Is an Economist’s Dream” and talked about how the data from Uber is helping to teach economic students about the supply & demand curve. With that massive amount of information stored in a database (probably has some major partitioning happening there!) it is able to be used across industries and help to discover key insights, apart from just how to optimize an Uber trip.

Final Thoughts

I enjoyed reading Data Strategy, it was an easy read and the topics were very relevant to those discussed in class. I was happy to be able to cite the book for parts of my research paper in KMBI for concepts around wearable technology and factor analysis. I did not care for the latter portion of the book – discussions on technical solutions (Hadoop, AWS etc). It didn’t feel like Marr was providing any insights, just simply listing research he had completed around the topic. I am interested in learning more about what Brooks’ specific data strategy is and if it seems to fit Marr’s requirements of specificity. Working in e-commerce there is data surrounding every aspect of my job. I look forward to using the information gleaned from Data Strategy to help foster a “data-driven culture” and help the Digital Products team make smarter, more effective decisions on the website.