1. Most authoritative statement of rule(s) from high court.
2. Most recent reaffirmation of rule(s) from the binding lower courts.
3. Which case most approximates your case's facts? Do any highly analogous cases exist in other jurisdictions (especially if their overall legal principles are similar)? Typically, real world assignments do not precisely match precedents. The best lawyers are those who see connections between cases with superficial dissimilarities. 
4. Which cases state the principles in the most advantageous way? In which cases does the position you represent win (e.g., the defendant, the movant, etc.)? 
5. What trouble lurks in the cases that your opponent will exploit in attacking your arguments? 
6. What trends can be identified in the cases (e.g., “no court has ever . . .”)? 
7. What overarching principles can you synthesize into a framework to demonstrate to a court, even if the framework is not specifically articulated by courts? Are there related principles that may help support arguments you want to make? 

1. Secondary sources can lead you to the leading cases, plus related propositions that you need to show to bridge to your main claim. 
2. Annotated codes provide most of the key cases that interpret a statute. However, they generally are not 100% comprehensive.
3. One Good Case Method
   • ​​​​Looking back: What cases are cited by the court in your One Good Case? 
   • Looking forward: Using a citator such as Key Cite/Citing References (W) and Shepard’s (L), what later cases interpret the cases you intend to rely on? 
   • Looking out: Topical organizations (headnotes) categorize all published decisions, so a digest will give you good coverage of the universe of cases that exist. Identify Topic + Key Number or Headnote from leading case, or browse the Key Number System Outline on Westlaw. 
4. Keyword Search. Use the keywords you discover by reading the earlier cases to construct an advanced search. Consider conducting terms and connectors searches and narrowing results by using /s, /p, /n, or atleastn(term). 
5. Repeat the above as needed. When you find a new case, Shepardize or KeyCite it. Read it to see if it cites new cases. See if it provides a Key Number that you haven’t seen. Note whether it uses a new term for your issue that would make a worthwhile search. 
6. Final Step: Shepard’s/KeyCite to confirm the cases are still good law. 

Determining this comes with experience and can depend on the issue or the court. The prevailing attitude is that when you keep seeing the same cases over and over again without anything new, you’re done. In general:

1. Use at least one source/search that involves an editor attempting to assemble like cases together. This may be the Westlaw Key Number system, Lexis headnotes, or an annotated code.
2. Use at least one online Boolean search to see if cases exist that were not identified in your first source or were too recent to be indexed in any other system you used.
3. Shepardize or KeyCite every case on point (within reason depending on how many exist).
4. Read every case that might provide new cases, that you might cite or that your opponent might cite, and that is from the highest court (within reason depending on how many exist).