Planning
Why Planning Matters
Before you open Builder and start creating sections and questions, take time to plan. A well-planned evaluation is easier to build, easier to test, and easier to maintain. When you invest effort upfront in defining the problem and mapping out requirements, you avoid costly rework later in the process.
Define the Problem
Start by clearly articulating the clinical problem your evaluation addresses. Ask yourself:
- What clinical decision does this evaluation support?
- What gap exists in the current workflow that this evaluation fills?
- What outcome do you want to achieve when a user completes the evaluation?
Write down a concise problem statement and refer back to it throughout development. Every feature you add should connect directly to solving this problem.
Identify Your End Users
Determine who will use your evaluation and in what context. Your end users might include clinicians, nurses, patients, researchers, or administrative staff. Each user group brings different expectations, technical comfort levels, and workflow needs.
Consider how your users currently handle the evaluation process. Understanding their existing workflow helps you design an evaluation that fits naturally into their practice rather than disrupting it.
Set Realistic Goals and Objectives
Ground your objectives in industry best practices. Define what success looks like for your evaluation by setting measurable goals:
- How many evaluation paths does the eval need to support?
- What accuracy standards must the scoring meet?
- How quickly should a user be able to complete the evaluation?
Present your perspectives and goals in terms that align with established clinical standards. This grounds your evaluation in recognized methodology and builds trust with your user base.
Think in User Stories
Frame your requirements using the user story format:
As a role, I want feature, so that benefit.
For example:
- As a primary care physician, I want to calculate a cardiovascular risk score, so that I can recommend appropriate preventive interventions.
- As a patient, I want to complete a symptom assessment on my phone, so that I can share results with my provider before my appointment.
- As a clinical researcher, I want to export scoring data, so that I can analyze outcomes across a study cohort.
User stories keep you focused on delivering real value to the people who use your evaluation.
Map Inputs and Outputs
Before you build, map out the data your evaluation needs to collect (inputs) and the results it needs to produce (outputs). Ask yourself:
- What information do you need from the user to reach a clinical conclusion?
- What scores, classifications, or recommendations does the evaluation produce?
- How do inputs connect to outputs through scoring or decision logic?
Sketching this map on paper or in a diagram helps you see the full picture and identify missing pieces before you start building.
Consider the Clinical Workflow
Think through the natural flow of a clinical evaluation. Most evaluations follow a progression through these phases:
- Intake -- Gather demographic and baseline information.
- Assessment -- Collect clinical observations, measurements, and responses.
- Scoring -- Apply algorithms or decision rules to the collected data.
- Recommendation -- Present results, interpretations, or next steps based on the scores.
Align your evaluation's structure with this workflow so users experience a logical, familiar progression.
Choose Your Evaluation Type
Determine which evaluation type best fits your use case:
- Decision Tree -- Present one question at a time, where each answer determines the next question. This approach works well for branching clinical pathways where subsequent questions depend on prior responses.
- Calculator -- Display all questions at once, allowing the user to fill in values in any order. This approach suits scoring tools where you collect a fixed set of measurements and compute a result.
Your choice of evaluation type shapes how you structure sections, questions, and logic throughout the rest of the build process.