03.03.02 — Review Model: OMG → Try Again → Close Enough → Good Job
Lesson goal
This lesson introduces a practical review model for evaluating the quality of test cases in Salesforce projects.
After this lesson, you should be able to:
- quickly assess test case quality without overanalyzing,
- identify the main reason why a test case is weak,
- guide improvements in a structured and repeatable way.
Why test case reviews need a model
Reviewing test cases without a shared framework often leads to:
- subjective feedback,
- inconsistent quality standards,
- long review cycles with little improvement.
In Salesforce projects, where test cases are heavily context-dependent, reviews must focus on fitness for execution, not stylistic preferences.
A simple review model helps:
- align expectations,
- speed up feedback,
- improve quality incrementally.
The four review levels
The review model uses four quality levels:
- OMG
- Try Again
- Close Enough
- Good Job
Each level answers one question:
How usable is this test case right now?
Level 1 — OMG
An “OMG” test case cannot be executed reliably.
Typical characteristics:
- missing or vague Preconditions,
- unclear user context,
- steps that rely on assumptions,
- Expected Results that verify nothing meaningful.
Symptoms:
- different testers get different results,
- execution depends on tribal knowledge,
- failures cannot be reproduced.
Action:
Do not execute. Redesign required.
Level 2 — Try Again
A “Try Again” test case has a visible structure but is still unsafe to execute.
Typical characteristics:
- partial Preconditions,
- mixed setup and execution,
- UI-driven steps,
- incomplete verification.
The test case:
- might work for the author,
- fails during Dry Runs,
- exposes hidden assumptions.
Action:
Return for revision with targeted feedback.
Level 3 — Close Enough
A “Close Enough” test case is executable and mostly correct.
Typical characteristics:
- clear user and data context,
- stable High-Level Steps,
- meaningful Expected Results,
- minor clarity or maintenance issues.
This level is often sufficient for:
- internal testing,
- sprint validation,
- exploratory support.
Action:
Execute, but log improvement notes.
Level 4 — Good Job
A “Good Job” test case is production-quality.
Typical characteristics:
- explicit and complete Preconditions,
- clear separation of intent, context, and execution,
- durable step instructions,
- precise Expected Results,
- easy to review and maintain.
These test cases:
- survive system changes,
- support delegation,
- build long-term test assets.
Action:
Approve and reuse.
Using the model during reviews
The purpose of the model is not to label people.
It is to:
- quickly classify test cases,
- focus feedback on the biggest risk,
- avoid micro-optimizing weak foundations.
A review should answer:
What is the single biggest issue preventing this test case from being “Good Job”?
Common review traps
Avoid:
- debating formatting,
- enforcing personal writing style,
- over-focusing on step wording while Preconditions are weak.
Reviews should prioritize:
- execution safety,
- clarity of context,
- verification quality.
Improving test cases iteratively
Most test cases will not be “Good Job” on the first pass.
The model supports:
- fast feedback cycles,
- visible progress,
- shared quality standards.
A “Try Again” today can become “Good Job” tomorrow with focused changes.
Key takeaway
- Test case reviews require structure.
- The OMG → Try Again → Close Enough → Good Job model provides fast, practical guidance.
- Focus reviews on usability and risk, not personal preferences.
Consistent reviews create consistent quality.