Most firms are swimming in data but starving for insight. My passion is making sense of the complex and finding signal through the noise: creating actionable insights that drive business results.
My name is Tim Anderson, and my career has been focused on hardware and software product management, but I also greatly enjoy corporate strategy, insight development, data wrangling/visualization and financial analysis. I have built my career around asking questions, sifting through data, building hypotheses, finding the critical information, and developing insights.
For many years I used Excel for financial modeling and business forecasting. About six years ago I stumbled onto R and learned a term called reproducibility. I realized that in Excel I could accomplish a goal quickly, but six months later if someone asked me how I got a result I was hard pressed to describe the steps I took to get an answer. A long time ago I used VBA to automate some of my Excel work, so R seemed liked a logical step
I’ve been using R for the past five years in my workflow analyzing markets and customers. I recognize that R is heavily used in universities and scientific fields, the purpose of this blog is to describe some of the business methods built in R. You won’t find much analysis of iris data here, but I will share some thoughts about customer acquisition, pricing, product feature analysis, churn, profitability, etc.
My objective with this blog is to capture some examples of my work mostly for my own reference. Originally, I built these pages with the R Bookdown package, but updated it to Quarto recently.
Below is a work in process description of how I think about data analytics. It’s mainly for my own edification as a template to follow, or things to think about as I begin projects.
The QUEST Framework: A Virtuous Cycle of Discovery
Introduction
In any complex endeavor—be it scientific research, data analysis, or strategic business planning—a structured approach can mean the difference between simply gathering information and generating real, actionable insights. I’ve adopted a five-step cycle, I call QUEST, which guides teams through a purposeful sequence of Question, Understand, Execute, Synthesize, and Transmit. By cycling through these steps, we ensure that each new round of inquiry builds on the lessons learned from the last.
I’ve unapologetically used the acronym QUEST to evoke the vision of an epic adventure…because in my estimation that should be our aim…embarking on a data project should be a quest towards the answers we seek, but also we can keep it a little fun.
1. QUESTION
Every impactful project starts with a compelling question. This step ensures we’re aiming at the right target. Here, we pinpoint the specific problem or gap that demands our attention—be it improving a business process, testing a scientific hypothesis, or exploring an untapped market opportunity. A well-crafted question sets the stage for clear objectives and measurable outcomes.
Key activities
Defining the core problem or question
Identifying the various stakeholders
Setting success criteria
Linking to relevant frameworks or theories
Outcome: A sharp, focused question that drives the rest of the process.
2. UNDERSTAND
Once the question is clear, it’s time to explore and prepare. We start by delving into existing knowledge: reviewing relevant literature, historical data, or prior work. Next we review what data we have and create a list of data we need to procure. During the data understanding phase it’s important to focus on our data dictionary to make sure we know what data we have. Then we formalize our approach, determining which methods or tools best fit the problem. Along the way, we assemble the right team—complete with domain experts, data analysts, or whichever skill sets the project demands—and ensure that ethical and regulatory standards are met.
Key activities
Conducting thorough research or market scans
Data acquisition/warehousing strategy
Outline how data will be obtained, including APIs, internal data warehouses, or data-sharing agreements.
Plan how often data will be refreshed (one-time vs. ongoing).
Data governance and security policies
Choosing methods (qualitative, quantitative, or mixed approaches)
Assembling expertise (and ensuring all ethical boxes are checked)
Outcome: A solid foundation of background knowledge, a clear plan, and a well-prepared, ethically aligned team.
3. EXECUTE
With a plan in hand, we move into action. This stage is where data is collected and analyzed, experiments are run, or surveys are administered. The key is rigorous data management—everything from ensuring the quality of inputs to adhering to privacy and governance standards. As findings begin to emerge, we share initial results with stakeholders, inviting feedback to refine our direction as needed.
Key activities
Gathering and cleaning data
Data preprocessing, and exploratory data analysis (EDA)
Applying analytical or experimental methods
Engaging and validating insights with stakeholders
Outcome: A set of verified, initial results and insights—backed by real data and stakeholder input.
4. SYNTHESIZE
After the execution phase, it’s tempting to draw quick conclusions, but the Synthesize step encourages deeper reflection. We compare our findings against the original question and the broader context, exploring any surprises or limitations. Crucially, we invite feedback from the team, peers and other stakeholders, giving us a chance to refine our understanding before finalizing recommendations. This feedback loop often reveals new angles or clarifies ambiguities, resulting in a richer, more accurate interpretation.
Key activities
Evaluating findings in relation to the initial question
Refining conclusions based on domain insights or unexpected patterns
Holding constructive discussions with stakeholders
Seek peer review and solicit cross-disciplinary input
Outcome: A mature, thoroughly vetted set of conclusions and next-step recommendations.
5. TRANSMIT
Finally, we share our findings with broader audiences. This might mean publishing an academic paper, presenting to senior leadership, or creating a user-friendly dashboard for frontline teams. By tailoring communication to each audience, we ensure that insights are not only heard but actually acted upon. In doing so, we close the loop on our initial question—but the process doesn’t end here.
Key activities
Preparing clear, targeted reports and presentations
Providing actionable recommendations or open data for reproducibility
Identifying new questions or avenues for future exploration
Outcome: Findings are disseminated effectively, sparking new questions and continuous learning.
The Virtuous Cycle
The beauty of the QUEST approach lies in its cyclical nature. Each new piece of knowledge, once transmitted, naturally leads to better questions. These refined questions launch a fresh round of the process—Question, Understand, Execute, Synthesize, and Transmit—where each iteration builds on the successes (and failures) of the last. Over time, the team’s collective expertise deepens, and the quality of insights grows exponentially.
The QUEST framework provides a disciplined yet flexible roadmap to help the team stay focused on the right objectives, gather and analyze information rigorously, and communicate outcomes in a way that drives real-world impact. By asking the right question and following it through each step, the teams cultivates not just answers, but a powerful cycle of continuous learning and innovation.