In my first Working Out Loud post for this course, I shared the story of remodeling my master closet. Hidden behind one section of drywall was mold that I never expected to find. That experience became my first lesson in needs assessment: what we see on the surface is not always the real problem.
At the time, I thought that was the lesson. By the end of the course, I realized it was only the beginning.
Over the past six modules, I learned that uncovering the problem is only the beginning. Learning designers also need to gather evidence, distinguish training needs from non-training needs, and build recommendations stakeholders can trust. Looking back now, I can see just how much my understanding of needs assessment has changed.
My Biggest Shift: From Confirming to Investigating
The first time my thinking was challenged was during the Bitloom Technologies scenario in Module 1. Like Bitloom’s leadership, I assumed that if someone requested training, there was probably a training problem to solve. As we worked through the Five Steps of Needs Assessment, I realized that assumption could easily lead organizations to invest time and money solving the wrong problem.
Instead of confirming what stakeholders already believe, a needs assessment asks us to investigate what is actually happening. It begins with questions rather than solutions. Who is experiencing the problem? What evidence already exists? Who should be interviewed? What additional information is needed before making recommendations?
That shift changed how I view the role of a learning designer. Our responsibility is not to validate assumptions. It is to uncover the evidence before deciding what kind of solution, if any, is actually needed.
Training Isn’t Always the Answer
That lesson became even clearer during the Forbin International case. Leadership believed inconsistent production, rising accidents, and low morale stemmed from inadequate training on newly installed equipment. At the beginning of the course, I probably would have agreed.
Instead of jumping directly to a solution, I developed a needs assessment plan, created a data collection strategy, and analyzed four employee interviews using Braun and Clarke’s thematic analysis process. The evidence revealed a much more complicated picture. While employees needed additional training on the equipment, they also described inconsistent safety enforcement, poor communication between leadership and employees, and limited opportunities for professional growth.
Only one of those issues could realistically be solved through training. The others required organizational changes that training alone could not fix. That realization fundamentally changed my thinking. Learning designers are often asked to create instruction, but our greatest value may be recognizing when instruction is only one piece of a much larger solution. Anyone can recommend a course. The real expertise is knowing when not to.
Evidence Changed the Way I Think
This course also changed the way I see evidence. Like looking into a mirror, evidence can reveal details we might otherwise overlook, but only if we’re willing to look closely.
One of my biggest lessons came from a mistake I made while analyzing interview data. I summarized themes using what I believed were participant quotations, only to realize that some of those phrases were actually my own coding labels. Codes represent my interpretation. Quotes represent the participant’s exact words. Quotation marks are a promise of accuracy, and that experience reminded me that the credibility of an analysis depends on getting the small details right.
The course also connected directly to my own workplace. We complete employee feedback surveys that ask us to identify our department and specialized role. Although they are technically anonymous, those details often make it possible to identify who submitted a response. I’ve watched colleagues become hesitant to provide honest feedback because they believed it affected how they were treated afterward. Whether that perception is accurate or not, it changes the quality of the data.
When the Forbin case described survey results that revealed very little, I immediately understood why. If people do not feel psychologically safe, they protect themselves instead of telling the truth. That realization changed how I think about ethical data collection. Protecting participants is not simply about following research guidelines. It is essential for collecting evidence that accurately reflects reality.
I also came to appreciate the value of triangulation. Comparing interviews, surveys, observations, and organizational documents provides a more complete picture than relying on a single source. Sometimes the most meaningful insights come from conflicting evidence. Those inconsistencies often point directly to the underlying performance problem.
Finally, I learned to clearly separate findings, conclusions, and recommendations. Findings describe what the evidence shows. Conclusions explain what those findings mean. Recommendations identify the actions most likely to improve performance. Maintaining that sequence keeps recommendations grounded in evidence instead of assumptions.
How This Will Change My Practice
Whether I continue supporting teachers in K–12 education or move into instructional design and learning and development, this course has changed how I will approach performance challenges. I will begin by asking better questions, gathering evidence from multiple sources, protecting participants throughout the process, and resisting the temptation to recommend solutions before fully understanding the problem.
More importantly, I now understand that learning designers do far more than create engaging instruction. We help organizations solve performance problems by identifying their true causes. Sometimes training is the answer. Sometimes it is only one part of a much larger solution.
When I look at my remodeled closet today, I don’t think about the shelves, the wallpaper, or the finished design. I think about what was hidden behind the wall and how easy it would have been to ignore it. The same is true in learning design. Performance problems often look obvious until we investigate them. Our responsibility isn’t simply to build learning solutions. It’s to uncover what’s beneath the surface so our solutions address the real problem.
The problem is rarely what it first looks like. The best learning designers know that understanding the problem is the first step toward solving it.
I used Chat GPT to review my own written work against the assignment requirements and suggest revisions while preserving my voice and ideas. I critically reviewed all suggestions, made my own decisions about what to include or revise, verified course-specific content against my own work, and take full responsibility for the final submission.