At this stage, we are looking for whether the proposed software project will be a suitable candidate for a service! There are two ways to do this:
top-down approach and bottom-up approach
The implementation of the service itself does not refer to any specific platform. If we want to consider a part of a software as a service should have the following criteria:
1. Can this function be defined in such a way that Service concept express correctly?
2. Can this service be used across multiple platforms? Or does it require interaction? And do business partners Organizations will use this service? Does it need to go through a firewall or other barriers? Can from Use a well-known protocol like HTTP or SMTP?

3. Is the implementation of this function as a service to facilitating integration it helps?
4. Does it explicitly refer to one or more business processes or is it as simple as a program?
5. Does the service function cover the scope of work?
6. Are you looking for someone in the business realm Output report Will this service? Is the field of IT to Life cycle or time Will the implementation of the service matter?
7. Does this service require business values? And is such a service with this function already available somewhere? be
8. Is the proposed service at the right stage of scaling?
9. Will servitizing this functionality make it composable?
10. Will our candidate demonstrate a clear life cycle idea?
11. Would it be useful to be able to find the software dynamically?
12. Will we reach the goal of providing "Single version of truth"?
13. Would it be beneficial to use XMl to display data?
14. Will the implementation of this agent as a service lead to a reduction in the cost of integration in the future?

The above items cannot be considered as fixed and main principles, these items are only suitable suggestions that are used in organizations Various are different.
Therefore, the idea of "turning all work into a service" is not an effective idea in organizations, and we must first check that it is a service Is it profitable to do the desired work or not? Therefore, the idea that "the more services, the more successful the company" is completely unfounded. Is.
In an organization, everything cannot be a service and everything should not become a service. There are important and fundamental problems that SOA and Services do not have good solutions for them, however, when you think of a service, you tend to think of any new project Write the company as a service, with this view not only there is no guarantee of return on investment, but it leads to the creation of a large number It will be serviced, which is referred to as Anti-pattern, actually having services for building and structuring SOA before The beginning should be checked "which one can be chosen as a service candidate" and before building the service catalog We must know exactly what should and what should not be serviced.
There are two main approaches at the beginning of service summary work to analyze and determine the service candidate:
Top-down or bottom-up approach
At Top-down approach By examining the macro business areas of the organization, processes are identified and then broken down into services orchestration and basic extraction, and then finer-grained services are identified. On the other hand, with a bottom-up approach Analyzing the facilities and capabilities of the organization's legacy systems is done to create atomic and software-capable services Extracts from these systems are identified.
A top-down approach for start-up businesses that have less disjointed systems and are generally more forward-looking It will be more beneficial. And the bottom-up approach will be appropriate for businesses that have been around for decades This field of work is involved, or has existing systems, or uses different platforms or communications. There are many point by point.