Conventions, Rules

Analysis is a highly creative, almost investigative activity that is often more of an art than a science. Because every analysis is unique, it is difficult to predict exactly what outputs it will produce, making it hard to standardize. This differs from software development, where the primary output is code—a format for which formal rules and conventions can easily prescribe how it should look and be processed. Analysis, by contrast, produces everything from plain text to pseudo-code, making strict output rules harder to enforce. But it is definitely not impossible...

Standardized Outputs

As modern information systems grow more complex, typical organizations use far more applications than they did in the past. Consequently, analysts need to quickly gather and understand information from a wider variety of sources. This becomes a difficult task if systems are documented inconsistently and the quality of documentation varies wildly. Whether the documentation describes what needs to be built (development specifications) or what has already been built (post-release documentation), the goal is to present information in a unified format to keep it manageable, maintainable, and searchable.

The sheer volume of information describing modern systems makes it impossible to manage without proper standards. Standardizing outputs allows anyone in the organization to quickly find information about any system.

Conventions

Analysts can also learn a lot from how software development teams use conventions. Conventions are rules that enforce standardization. For example, naming conventions define how various entities should be labeled. When documented components of a system or domain follow a standardized pattern, the overall documentation becomes more consistent and easier to search. If one use case is labeled "Customer searches meals - use case" and another is called "Show meal list," it confuses readers and disrupts search results. Naming should always follow consistent, predictable patterns. Here are a few examples of well-structured use case and screen labels: "UC Search Meals", "UC Add Contact", "SCR Show Meal List", "SCR Update Client".

Another example is modeling conventions, which define how specific diagrams should look to keep them as unified as possible. This supports collective ownership of models and makes maintenance much easier. Some modeling conventions also improve readability, such as the basic rule to avoid crossing lines wherever possible.