Own the Unknown™ with Donald Farmer

Welcome to the first November issue of Further's Own the Unknown™ LinkedIn newsletter, which means it is time to introduce you to a new thought leader. Twice monthly, we'll share some of the knowledge we've gained from following, reading, and interviewing some of the most insightful and influential thought leaders on LinkedIn.

This month, we will be interviewing Donald Farmer . In this edition, we'll draw both from his book, Embedded Analytics: Integrating Analysis with the Business Workflow, coauthored with Jim Horbury, and his frequent conference speaking. Donald is a member of the faculty of TDWI. Our Data Science Principal Keith McCormick had an opportunity to spend a week with Donald at the recent TDWI Transform event in Orlando. They both offered masterclasses, and they both spoke to the attendees of the Modern Data Leader’s Summit. This issue of the newsletter will also draw heavily upon Donald’s three masterclasses offered in Orlando which covered topics ranging from analytics culture, to data literacy, to predictive analytics. So we have a particularly rich supply of thought leadership to mine this month.

Donald Farmer is a data and analytics strategist with over 35 years of experience in the field. As Principal of TreeHive Strategy and VP of Innovation at Nobody Studios , he advises global clients on innovation, analytics, and AI. Donald's career includes leadership roles at Microsoft and Qlik, where he pioneered significant product designs.

We will be interviewing Donald on November 14th at 2:30 pm. 

Keith and Donald were joined in Orlando by two more members of the Further team who are also faculty members of TDWI, Nicolas Decavel-Bueff and Kristy Hollingshead . Kristy and Further’s head of Data Science and AI, Cal Al-Dhubaib have also spoken to the Summit audience. We’ll have more to say about Further’s participation in upcoming TDWI events at the end of this issue.

An Effective Analytics Culture Needs Data Literacy

Donald likes to sit in on client’s company meetings. He’s often witnessed meetings where the conversation might briefly begin with some reference to data, often a challenge the company is experiencing, but the conversation rarely stays data driven for very long. He encourages a “supportive environment for curiosity”. In order to achieve this kind of effective analytics culture you need a spirit of collaboration between the data teams and others throughout the organization. And that requires some degree of data literacy.

A data driven culture creates a culture where employees feel comfortable asking questions, seeking guidance, and experimenting with new data-driven approaches, supporting continuous learning and improvement.

According to Donald, an analytics culture also includes promoting an experimental mindset. That is certainly something that we also encourage at Further as we have a team with experts like Melanie Hall dedicated to helping clients with using experiments to confirm (or disconfirm) that a strategy is working.

An analytics nurtures an experimental mindset and stresses its benefits. It encourages employees to continually learn and adapt to new information and changing circumstances, supporting a culture of innovation. This understanding inspires employees to embrace this mindset, leading to new ideas and the willingness to take calculated risks.

Donald discusses several criteria for an effective analytics culture, but we’ll mention just one more of them here, data governance, which was also a major theme in our conversation with Jonathan Reichental, Ph.D. in September. Data governance should be enabling, and not a burden. In his masterclass he used the charming metaphor of preparing a home for the arrival of grandchildren. If it is properly prepared his young guests can spiritedly and safely explore the environment. Lacking these preparations everyone, including Donald and his wife, would be forced to closely monitor the situation and frequently intervene. But with data governance in place, along with data that is “well sourced and fit-for-purpose” employees can “feel free, safely, to analyze the data and satisfy their curiosity”. It is reminiscent of a phrase that has become very popular with Cal Al-Dhubaib and other members of our data science and AI team here at Further, “brakes let you go faster”. 

From Prediction to Action

Donald begins his masterclass, Beyond Analytics: From Prediction to Action, with a very simple but powerful classification system. Decisions can be divided into:

• Strategic Decisions: long-term and “involve the overall direction of the business.” They require input from senior management.
• Tactical Decisions: more “short-term in nature, focusing on day-to-day operations that will help meet strategic goals.”
• Operational decisions: each of smaller business value, but that in the aggregate, at scale, can add up to tremendous value.

A number of insights can be derived from these categories. Despite occasional industry talk of "strategic analytics,” Donald thinks that it is difficult for a data team to provide everything that senior management needs to make a strategic decision. Often they are augmenting the data with information that they have gathered informally in board meetings, conversations, and their own personal reading of trends. Or even useful external data like the weather, which could possibly be integrated into a model by a data team. For Donald, tactical is the area where business intelligence has historically thrived, giving middle management they need, in a timely manner, to drive day-to-day decisions. Also, at the tactical level, full automation is not as common as having a human-in-the-loop approach. Arguably predictive analytics and traditional machine learning has focused instead on operational decisions, sometimes with full automation of these more granular decisions, sometimes called “micro-decisions”. For instance Tom Davenport describes these decisions as being “found deep within key operational processes.” (Tom and his coauthor, Ian Barkin, will be joining us as our interview guests in December.)

Embedded Analytics

In the preface to their book, Embedded Analytics, Donald and his coauthor ask: “how to reach those underserved users (of analytics)?”  Their answer is “that we can embed analytics into the applications that business users work with every day, making analytics part of their regular workflow, not a specialized practice.” We will certainly be asking Donald about this because it will be fascinating to compare and contrast embedded analytics as a solution to this with the idea of the “citizen” data scientists. The citizen movement will be a focus of our discussion with Tom Davenport and Ian Barkin in December.

Another aspect that we’ll ask Donald about is that embedded analytics tries to avoid forcing the business user to switch applications when they need a data-driven answer. 

In addition to being familiar, it is also more efficient to embed analytics in a commonly used application. If we need to switch from our working application to an analytics application, we interrupt our workflow; we have to open and navigate to another application and may lose focus while we’re doing it. How does this work in practice? In a call center, an operator taking customer calls for support can see on one screen not only the customer record, but also (thanks to the embedded analytics) trends and patterns related to this and similar cases. These patterns (perhaps the customer is calling increasingly often over time) may prompt the operator to route the call to a specialist who can better diagnose and resolve the underlying issues. The result should be a better resolution for the customer and more efficient handling by the call center.

Much more about embedded analytics and analytics culture to come in our live conversation. We hope that you will join us. 

Upcoming interviews and events

If you haven't done so, follow Further here on LinkedIn. That's the best way to get the latest news. And click on "attend" so that you won't miss the interview with Donald. You'll also be able to watch the recording in your LinkedIn feed. We are also excited to announce that our December event has already launched. We will be interviewing Ian Barkin , and Tom Davenport about their new book, All Hands on Tech: The AI-Powered Citizen Revolution.

And we have a very exciting event coming up in the Bay Area. Join us in Sunnyvale at the Google Cloud office for Leveraging Google Cloud & AI for Competitive Advantage in Marketplace on November 13th.

Join us at the Ohio AI Summit on November 20th.

We’ve already mentioned the strong partnership between Further’s Data Science and AI team and TDWI. Cal Al-Dhubaib will be presenter for the TDWI Virtual Summit, Powering the Intelligent Enterprise through AI and ML, coming up very soon on November 13. And just around the corner is TDWI Transform West: Las Vegas. Donald, Keith, Nicolas, and Kristy are all presenting masterclasses.

Nicolas Decavel-Bueff and Kristy Hollingshead will present Hands-On Introduction to Customizing Large Language Models (LLMs)

Keith will present the two-day Data Science Bootcamp and AI and Deep Learning: Techniques and Use Cases for Profitable Applications

Donald will present his masterclasses on Teams and Technology: How Your Choices in Technology, Policy, Hiring, and Training Can Create a More Effective Analytics Culture as well as Data Literacy: A Strategy for Building a Culture of Analytics.

Finally, Cal, Keith and Nicolas have all been on the TDWI podcast.