13 June 2016


The software requirements are description of features and functionalities of the target system. The requirement gathering process gathers the requirement from client then analyze the requirement document and finally System Requirements Specification document needs to be prepared.

Requirement Elicitation techniques

The requirement elicitation process consists the followings

  • Requirement Gathering: Developers understand from client/end user about the expectation from the software.
  • Organizing Requirements: Developers prioritize the requirements based on the criticality of the requirements.
  • Negotiation and discussion: Ambiguous or unrealistic requirements require discussion with various stakeholders.
  • Documentation: Functional and non functional requirements need to be documented and made available for the next phase processing.

There are different requirement elicitation techniques. Few important techniques are listed below

  • Brainstorming: An informal debate is held among the various stakeholders and all their inputs are recorded for further requirement analysis.
  • Interface Analysis:
  • Interviews, Survey and Questionnaire:
  • Prototyping:
  • Requirement Workshops:
  • Delphi Techniques: This is very famous technique of requirement elicitation. This techniques involves structured interaction among a group of experts on the requirements. The identity of the participating experts are kept confidential to avoid influent on the process. There can be round of inputs from the experts till the process is finalized. The facilitator will have multi round one to one interview with each expert on the questionnaires and collect their opinions and answers.

Brainstorming Techniques

  • An informal debate is held among the various stakeholders and all their inputs are recorded for further requirement analysis.
  • Brainstorming session will have a facilitator and will have time limit for the discussion.
  • Advantages
    • Helps to generate new ideas
    • Helps to encourage involvement of participants
    • Focuses on a topic or problem and coming up with many ideas or solutions
  • Disadvantage
    • During brainstorming session, ideas are not explored.
  • Consideration for Brainstorming techniques
    • Brainstorming is useful early in the requirements elicitation process.
    • Brainstorming can help to generate wide variety of views of the problem and solution.
  • Brainstorming sessions are ideally divided into two phases
    • Idea Generation Phase: It focuses on the gathering all possible ideas within the time limit
    • Idea Consolidation phase: In this phase, the ideas are getting revised, organizing and reviewed. In this phase, each idea gets evaluated, then combine ideas after removing duplicate ideas. Finally, ideas will be rated and ranked after discussing with the team
  • Different types of brainstorming techniques
    • Open Brainstorming: Participants call out the ideas and are capture by the meeting facilitator
    • Individual Brainstorming: Project team member creates a list of features/ideas concerning project issue/risk and shares with the facilitator
    • Structured Brainstorming: Meeting participant silently write down their ideas and facilitator requests

Interface Analysis technique

  • Interfaces would include user interface, interfaces with external applications and interfaces with external hardware devices.
  • Interface analysis helps to clarify the boundaries of the interacting application, identify the functionality, input and output of the each interface.
  • When a system takes input from multiple systems then the interface analysis techniques is preferred for the enhancement/change in that system.
  • Interface analysis techniques can not be used to identify hardware interfaces with software applications.
  • Context Diagram helps in interface analysis because it defines the boundary between a system and its environment, showing the entities that interact with it. This interface analysis tool is used as a part of the Prepare for Interface Analysis phase and NOT as a part of the Conduct Interface Analysis
  • Steps in interface analysis are
    • Prepare for interface analysis
    • Conduct interface analysis
    • Define interface requirement
  • Interface analysis helps in sizing and estimations as well. As far as sizing is concern, the interfaces are considered as non-functional requirements.
  • In interface analysis, the description of each interface, batch feed, transactional files and interfaces to and from the systems (input and output of each interface) are very important factors. Also the validation rules that govern the interfaces and events that might trigger the interaction will be documented.
  • In interface analysis, the user expectation is also documented.
  • The disadvantage for the interface analysis technique is it does not provide an understanding of the underlying business process

Software Prototyping

  • It is an activity of creating prototype of the software i.e. incomplete version of the software program. A prototype typically simulates only a few aspects of, and may be completely different from, the final product.
  • Prototype is a working model of software with some limited functionality
  • The prototype does not always hold the exact logic used in the actual software application and is an extra effort to be considered under effort estimation.
  • Dimensions of prototypes
    • Horizontal Prototype: It provides the broad view of the entire system. It focuses more on user interface than low-level system functionality, such as database access
    • Vertical Prototype: It is more complete elaboration of a single subsystem or function.
  • Types of prototyping
    • Throwaway prototyping: This type of prototype gets discarded rather than becoming part of the final delivered software. When you need to come up with quick hand drawn prototype then the throw-away prototyping is the best choice.
    • Evolutionary Prototyping: The objective of this type of prototyping is to build very robust and structured prototype which can be refined later.
    • Incremental Prototyping: Incremental prototyping refers to building multiple functional prototypes of the various sub systems and then integrating all the available prototypes to form a complete system
    • Extreme prototyping: This type of prototyping is used for developing web application. It breaks down the web development into three phases one after another. In the first phase, the static prototypes i.e. HTML pages gets developed. In the second phase, the screens are integrated with the simulated services layer. In the last phase the actual services are implemented.
  • Prototyping is used to allow the users evaluate developer proposals and try them out before implementation.
  • Advantages
    • Increases user involvement in the product even before the actual implementation
    • User gets a better and visual understanding of the system.
    • Defects get identified in much earlier phase of the development.
  • Disadvantages
    • If the scope of the actual system changes in the later phase of the development then the complexity gets increase since the prototyping gets started with very basic and less understanding abount the system.
    • Developers may try to reuse the existing prototypes to build the actual system, even when its not technically feasible
    • A prototype may create unrealistic expectation in clients and customers about the functionality, reliability or performance.
    • In the early stage of prototyping, one has to make assumption about underlying technologies


  • One of the best ways to gather requirements is to interview with the stakeholders. The interview can be
    • Structured interview: In structured interview
      • Design open and close ended questions
      • Define a logical flow of the questions
      • word the questions consciously and unambiguously
    • Unstructured interview: In unstructured interview
      • Determine the areas to be covered in the interview
      • determine the progression of the interview
  • Interviewer should handle interviewee’s concerns by clarifying them / recording for the future folloup
  • An interview can have close ended questions but interviewer should try to get a precise answer for those.
  • The important activity done while conducting interview is Scribing for better understanding of covered topic - scribing allows for effective documentation of the responses and details.
  • Survey is also another way to improve the software considering the feedback from the stakeholders. Survey can be
    • Open-ended survey: Survey held on overall application
    • Close ended survey: Survey held on a specific area of the application

Below are the follow-up phases of an interview/survey

  • After an interview, the interviewer collates the scribing done during the interview and shares with the interviewee.
  • For open ended surveys, themes that run across the responses are identified
  • Statistical summary is provided for close ended surveys
  • Survey results are summarized and shared with the project sponsor.
  • An interviewee reviews the notes and identified issues in documentation and any uncovered areas.

Requirements Workshop

  • Requirement workshop is a technique of requirement gathering or requirement elicitation.
  • Requirement workshop technique works best when it is led by a trained facilitator and supported by a scribe
  • Successful requirements workshop requires BA’s to carefully select the stakeholders and SMEs to attend this short and intensive meeting.
  • Advantages
    • This technique helps to get the broad view of areas under investigation.
    • It helps to increase productivity
  • Disadvantages
    • Sometimes it is difficult to get the stakeholders available and arrange the required logistics.
    • Also it is very time consuming to organize
  • Activities required to conduct a requirement workshops
    • Facilitate the workshop
    • Identify with stakeholders need to participate in the workshop
    • Set the stage and expectation
    • Clarify the requirements and scope for the workshop
    • Follow-up on the workshop tasks and requirements, to generate final deliverables.

Requirement Analysis

  • Business analysts uses different models to analysis the gathered requirements.
  • Analysing requirements using models will help to
    • identify gaps during the requirement elicitation phase.
    • define the expected behaviour of the system.
  • Business analysts validate business rule prior to the requirement analysis
  • Business Analysts typically use three Solution Development Methodologies(SDM) during requirement analysis which depends on the types of project, culture and standard of the organization.
    • Business process analysis: focuses on the process improvement. It is done using
      • Activity Diagram: It is a graphical representations of workflows of stepwise activities and actions. An activity diagram will have a starting point, actions, decisions, splits and joins to show concurrent activities, and ending points. An activity diagram is also a type of flow chart.
      • Flow chart: represents an algorithm, workflow or process. The difference between flow chart and activity diagram is that the former is limited to the sequential process with condition checking points whereas the later helps us to understand the concurrent or parallel processes as well.
      • Workflow model: this model depicts the flow of work in a specific business area.
    • Object-Oriented Analysis: This methodologies views a system as a collection of objects that passes messages from one to another. This is achieved using
      • Use case diagram: It depicts the different use cases performed by different types of users.
      • Activity Diagram: It is a graphical representations of workflows of stepwise activities and actions.
      • Class diagram: depicts the elements relevant to the solution. It also depicts the relationship between elements.
      • Sequence diagram: depicts the logic of use case descriptions.
    • Structured Analysis This technique focuses on the functional decomposition and data-flow analysis. It is achieved using one of the following
      • flowchart: represents an algorithm, workflow or process.
      • dataflow diagram: It depicts how information flows through a system. How it is input, processed, stored and output from a system.
      • entity relationship diagram: It depicts the data structre of an application.

Process and flow models

Process and flow models helps BA to understand how system behaves through time. Benefit of knowing the process and flow models

  • helps to understand the information expressed in the model
  • helps you to choose the right model for your project
  • helps you to effectively communicate project requirements
Dataflow diagram
  • The dataflow diagram depicts how information flows through a system. How it is input, processed, stored and output from a system.
  • Dataflow diagram is an ideal choice when solution involves the complex data processing.
  • disadvantages
    • In dataflow diagram decisions and user interactions are not described. It only describes what system does.
    • DFD is not useful for the projects trying to improve processes or create interactive user-based system.
    • If a project requires knowledge of how system works and how people interact with the systems - in such cases can be described using DFD.
  • Key elements of dataflow diagram are
    • External Entity: can send or receive data from data process within the system. External entity is represented using rectangle
    • Data process: processes data that can be used in the system. Data process is represented using circle.
    • data flow: is an arrow headed line represents the data flow from external entity to data process or data process to data store.
    • data store: stores the data. It does not transform/change data further. It is represented by a label inside two parallel lines.
      Symbols are shown below
  • The functions of the key elements in DFD are
    • Sending or receiving data
    • Moving data
    • Transforming data
    • Holding data
  • DFD analysis involves three steps
    • level 0 DFD: referred to as a context diagram, defines the system boundaries. Each level 0 DFD depicts the system as a single, usually complex process (the tasks that a system performs).
    • level 1 DFD: will be created once the system boundaries are defined in level 0 DFD. The process will further broken down into multiple processes. Sometimes data processed by one process need to store before moving the data to another process. This intermediate storage is done in data store.
    • breaking down complex processes A complex process is further broken down into multiple processes. They are sometimes called Level 2 DFD. Leve 2 DFD also can be further broken down to Level 3 DFD. Lowest level DFD may sometimes captures the error messages.
  • A DFD can be broken down into further levels, depending upon the complexity of the process involved. This process is called DFD Levelling
  • Validation rules of a DFD are
    • Each process must have one input and one output
    • Each datastore must have one input and one output
    • All dataflow must start or end at a process
    • No entity should be allowed direct access i.e. read or write to the data store.
    • Entity can not communicate to another entity. A process has to be in between them.
    • Datastore can not communicate to another datastore. A process has to be in between them.
  • Black holes, Grey holes and miracles in DFD
    • Black Holes: A Process may have input flow but no output flow called black holes.
    • Grey Holes: A process may have outputs that are greater than the sum of its inputs are called grey holes
    • Miracles: A Process may have output flow but no input flow called black holes.

Summary of different types of diagrams

  • Flow Chart: depicts according to ISO standard, the sequence and logic of a set of related activities
  • Activity diagram: depicts according to the UML standards, a sequence of related activities and the decision that controls the flow.
  • State machine diagram depicts all possible states an object may be in
  • Data flow diagram: depicts how information flows through the system
  • Workflow model: depicts the flow of work in a specific business area.
  • Sequence diagram: uses a class diagram to depict how a usage scenario is executed.

Data and behaviour model

  • Data and behaviour model provides a way of understanding data and the relationship between data.
  • This model does NOT show the data and process flow. Also does not depicts the user interaction
  • Common techniques used to generate data and behaviour model are
    • Entity relationship diagram(ERD): From ERD, a business analysts get the associated data with the entities and conducts structural analysis on their relationship
    • Class diagram(ERD): This diagram helps software engineers to build a system. It gives simplified views of business classes and the data associated with them. This is very popular technique of object oriented analysis.
    • Data dictionary: The glossary of the technical terms. So all stake holders understand the terms consistently.
    • Metadata definition: describes the context in which data is used.
    • Data transformation and mapping: helps business analysts to reuse data from legacy system to the target system
    • CRUD matrix: helps business analysts to verify and track access rights to data stores within a system.

Entity Relationship Diagram

  • ERD is one of the most commonly used data modelling technique.
  • To create ERD, following components are very important
    • Entity:
    • Relationship:
    • Cardinality:
    • Attributes:

  • In the below example, there are two entities called Artists and Songs. One song will have one and only artist but one artist may perform zero or more songs.
  • 4 stages to complete an ERD are
    • Identify the entities
    • define the entity relationships
    • define cardinality
    • assign attributes to the entities

Class diagram

  • The class diagram is a static diagram. It represents the static view of an application.
  • The class relationship can be depicted in a class diagram are
    • Association: explains how one class is associated with another class via some business rules.
    • Generalization: represents IS-A relationship. Savings account and Credit accounts are generalized by Account
    • Aggregation: represents HAS-A relationship. Suppose a class HAS-A students so Class is an aggregator and the lifetime of the students does not depends on the aggregator i.e. class. Another example is A school HAS-A classes. If school shuts down then all classes will also get closed. so the lifttime of the school depends on school.
    • Multiplicity: is similar to the cardinality in ERD. It represents one-to-many or many-to-one or many-to-many or many-to-one relationships.

Usage models

  • Usage models focuses on solution from the users prospectives.
  • This models describes only what is visible to the end user
  • The most common usage models are
    • prototypes: is a simplified representation of proposed software
    • user interface design: helps users to navigate specific system features
    • user profile: describes user requirement based on the user type
    • user stories: describes functional requirements from the user’s perspective. End users write these stories. It includes Description, Estimates, Priority, Unique Identifier and story card
    • use case descriptions: describe the behaviour - who (actor), what (behavioural requirements), how (events and conditions) and why(use case goals) - of the system being modeled. It includes Actor, Post Conditions, Flow of events, Pre-conditions and Name
    • use case diagrams: show the actors, use cases and the relationship that exists between the key elements of the system being modelled.
  • Following are the important steps to write usage models
    • Identify actors
    • Identify use cases: There are three types of use cases as shown below
      • Tasks: are the goals an actor accomplishes through the interaction with the system. Each task is a system use case.
      • Events: Systems responds to the events. Each response is a use case
      • Automated requests: get originated from the system and each request is a use case.
    • Create use case diagram: Add actors and use cases into a diagram to create use case diagram.
    • Describe Context: Context includes the information about the actors, pre-conditions and post-conditions.
    • Describe the basic flow: Focus is on the success scenarios.
    • Describe alternate and exception flow: Focus is on the alternate scenarios (variation of the successful scenario) and exceptions scenarios.
  • Below example clarify some of the important understanding. In the below image, the make appointment use case includes check patient record use case. Which means the later use case gets executed along with the former use case. Here both the cases are the part of the basic flow. Defer payment use case is the extended use case of pay bill use cases. So Defer payment does not occur always. In some cases extended use cases gets executed. Bill insurance use case depends on the more treatment use case. Hence these generalized use cases are part of the alternate flow as described above.

Non-Functional Requirements (NFR)

  • Quality of a software i.e. accuracy, performance, security and modifiability are the important consideration in NFR
  • SNAP(Software Non-functional Assessment Process) provides the quantifiable measure for NFR. SNAP helps to determine the NFR size and effort to address NFRs
  • If NFRs are not given proper importance then the amount of rework in the later stage would be huge.
  • FP GSC: When only function points are used for sizing, Functional Point(FP) General System Characteristics (GSC) are expected to cover NFRs.
  • SNAP always give better quantifiable measures for NFR sizing than FP GSC.
  • Usability is a metric to measure percentage of users who find it comfortable to use the system.
  • Changibility is a metric to measure the reliability factor of a system.

Design consideration and NFRs

  • Extensibility:
    • Can the database schema flex to accommodate changes?
    • Can end user extends the system - scripts, user defined fields etc.?
  • Portability:
    • For web applications, which browsers are needed to be supported?
    • In which OS the system will work?
  • Availability: How long the system will run without failure?
  • Scalability: What is the peak load that system must be able to handle?
  • Security: Prevents unauthorized access and ensures system integrity from the malicious damages. E.g. all external communications between server and client should be encrypted. Another example is backup should be taken in every 24 hrs and backup copy should be stored in another location.
  • Compatibility: Compatible to run the application when sharing resources with other applications running in the same system.

Requirements Traceability Matrix (RTM)

  • Requirements Traceability Matrix (RTM) is a tool used to trace each requirements of a given business case.
  • RTM captures all requirements and their tracibility in a single document
  • RTM also is used to record the relationship of the requirements.
  • RTM documents are created during the analyze phase of the project and updated and maintained throughout the lifecycle of the project.
  • RTM ensures
    • requirements are prioritized
    • requirements are examined for the system impact
    • requirements are uniquely numbered
    • requirements can be traced forward and backword.
    • coverage between requirements, features and use cases
  • RTM keeps track of the connection between
    • user requirements
    • functional specification
    • design specifications
    • test cases and
    • use cases
  • RTM work flow includes
    • Business analysts talk to the end user and gather the requirements
    • Solution architects creates the design specification
    • Developers starts implementing the requirements
  • RTM stakeholders
    • Business analysts creates the RTM documents
    • Solution architect updates the design specification details during the design phase
    • Test analysts updates the test case information and developers update the source code details
  • Types of RTM
    • Forward tracibility matrix: maps requirements into test cases
    • Backword tracibility matrix: maps test cases into requirements

Requirement prioritization technique using MoSCoW

  • MoSCoW analysis is a technique to prioritize requirements
  • M - Must Have, S - Should Have, C - Could have, W - Won’t/Would have
  • MoSCoW analysis is helpful when the priority of the requirements from multiple stakeholders varies. Steps of this analysis includes
    • Gather all requirements from all stakeholders
    • Call out a meeting with all stakeholders
    • All requirements will be labelled with M,S,C,W
    • In case of disagreement in setting priority then voting can be used to arrive at some decision
    • Finalize on the priorities
    • Notify the prioritized list to all stakeholders.

blog comments powered by Disqus