For me the top 2 are:
1) Repository based modeling - This means that drawings in ARIS are composed by selecting objects from a database. This provides reusability of the objects in many diagrams where the attributes of the object are always consistent. This also greatly enhances maintainability of the models over time as they are easier to change. If an object in a model gets changed, you change it once and it is updated automatically on all models. This also means that the models have a much greater chance of staying up to date and being reused across projects rather than starting from scratch. It also means that all of your information is one place and can be maintained consistently by the appropriate data owner. People from other domains can use the objects but don't need to "redefine" them (usually incorrectly or at least using some other name) which greatly improves model quality and communication across the enterprise. This also provides a very powerful impact analysis capability that cannot be done using multiple Visio diagrams spread across the organization.
2) Methodology based modeling - This means that each diagram type in ARIS follows specific rules that are understood by ARIS. This allows you use built in semantic checks (which you can customize to your own particular needs) to increase model quality and consistency. When objects are used consistently across the organization it greatly the communication quality and decreases the time needed for model reviews as you don't have to explain the meaning of the model to the audience (after the initial learning curve of course). I have worked with EA groups who have observed the average time for governance reviews goes from around 4 hours to less then 30 minutes due to standardized models and deliverables. Project teams don't have to spend any time figuring out what objects to use for what as the method is predefined and enforced. Also, it allows the tool to translate correctly between various model types and to generate models based on the content stored in the repository. It also leverages other capabilities like model simulation, integration with deployment environments (SAP SOLMAN, Oracle jDeveloper, etc.), PPM, Balanced Scorecard, etc.
These are the 2 that I stick to although you can see that each has many imbedded benefits.
I am sure others will have more to add.
Hope this helps,
Rick already explained two of the best points where ARIS outperforms Visio, here some more:
1. The ARIS methodology is much broader than the one from Visio
2. You can maintain all content in ARIS in different language, e.g. the name and other attributes of an object can be maintained in English, German, French and other languages at the same time
3. Around modeling ARIS offers a complete product suite for Publishing, Analysis, Optimization, Simulation
4. ARIS offers interfaces or deep integrated product to many platforms of companies like SAP, Oracle, IBM, ...
5. Much ARIS based reference content is available
Hope this helps you, others will surely add more points.
Hi Rick and Uwe,
you both have touched some very important points.
I have started working on compilation of these points....expecting some more replies in the discussion.
I agree in all the above, just...
Don't kill a fly with a shutgun...
Don't use ARIS for a simple task...
Actually I am just trying to figure out some parameters based on which I can decide whether I have to use Shotgun or not.
Who knows the fly might be worth of using Shotgun....why to take chance ?? :)
I like your response.
I believe I am doing the same work what you did three years ago. Can you please share your consolidated report on ARIS Vs. Visio if you have handy?