computer sys analysis and design

Chapter 4 – Information Gathering: Interactive Methods

Objectives

Understand the value of interactive, (communication), methods in regards to defining a system purpose as well as its detailed needs or requirements.

Review the interview process Client and sponsor buy-in and related ownership is THE key success factor. Their opinions and experiences are critical.

Understand the concept of JAD, (Joint Application Design), and when to use it.

Why use surveys?

Introduction

There are three main interactive methods

Interviewing

JAD

Surveys

Regardless of the approach, the key to success is in the structure/content of the inquiry as well as its documentation and resulting feedback. Always be aware of any personal bias and be prepared.

The #1 skill is to listen. Give the “power” to the person being interviewed and express gratitude. Be humble.

Interviewing

There are five planning steps

Master background material if required educate yourself regarding status quo, important attitudes, important terminology review my Berlin failure.

Establish objectives items such as current usefulness and usability, what works well, what are the gaps, where does information come from and where does it go, frequency and volume of use, scope, what are their priorities for the project.

Decide whom to interview cover all levels.

Prepare always contact the interviewee ahead of time and communicate the objectives pick a time that is convenient to them and begin and end on time if the interviewee is significantly late or is a no show or seems uninterested or even contrary then log the observations and move on do not attempt to squeeze a 30 minute interview into 10 minutes Always notify the chain of command that the interviews are occurring Always provide post session notes for clarity confirmation.

Finalize questions and structure there are open ended, closed and probing questions allow questions to be asked and if time allows vocally summarize what you feel are key findings.

The structure or format of the interview is dependent on the objective.

An objective that is very clear and understood, (e.g. we need to increase web transaction volumes), would favor a pyramid, (figure 4.7 page 109), approach while a broad exploratory objective, (e.g. we need to use technology to reduce manufacturing cycle time), would favor a funnel, (see figure 4.8 page 109) approach.

The diamond structure, (figure 4.9 page 110), is normally reserved when the interviewees represent all chain of command levels and in this regard is rare and normally applies to major corporate entity projects which would require the collaboration of a number of interviewers.

Joint Application Design (JAD)

Created by IBM as a cost and time savings requirement gathering structure.

Think of it as interviewing as a group effort.

Can be used if the objective or problem is not clear or a project effort requires a kick start.

The organizational culture supports group creativity software development, Cerner.

Constraints, (time, travel, and logistics), require a faster more efficient solution.

Those participating must be empowered by their management to make decisions and be accountable, (assume ownership) outside or normal daily tasks must be placed aside during JAD.

Any JAD session requires a leader who not only takes notes but clarifies objectives as well as keeps the session productive.

JAD efforts fail when

Collaboration is not possible

Preparation for involvement, (time, expertise, and understanding of the objective), is not accomplished

Management is not aware or supportive

The “great” session translates to no follow up or related action

Using Questionnaires / Surveys

Questionnaires or surveys use questions to gather details, (closed), and attitudes, (open ended), across a broad range of clients. Results can be quantified and can be used to create an objective statistically correct conclusion.

Strengths

Less overhead to produce

Can reconcile logistical, (volume and distance), opportunities

Can remove interviewee bias

Weaknesses

They are dependent only on those that respond

Assume an honest or sincere response

Often those that should respond do not

Who likes to take surveys?

Is impersonal.

They are best used when they are positioned as an entry point. For example, “What is your satisfaction with the current system?” / “Do you think we need a system?”

However open ended questions should attempt to limit the subjective responses so some value is obtained. For example a response, “I don’t like the system”, is worthless, rather offer the responder a list of reasons why they don’t like the system.

They should always be brief and as clear as possible. Surveys that exceed 5-10 minutes lose their effectiveness unless they become a requirement.

They should always be followed with

Confirmation and gratitude of receipt

Summary of findings

The potential of a follow-up interview.

What do you do once you have the responses?

Scaling

Assigns an interpretive, (interval), or “best” option, (nominal), measurement to the responses.

Normally a like or dislike ratio or of the following options, what do you prefer?

Effective methods of administering questionnaires

Web check boxes, radio boxes, third party administration and response crunching.

Physical or email drop box.

Group taking

Individual submission based on a demand schedule.

Often the percentage of responses, regardless of content, tell a story.

Let’s discuss your view on questionnaire effectiveness based on real life.

Have any of you used a survey and if yes what for? Was it effective?

What are your perspectives as a responder?

Provide a real life example.

Chapter 5 – Information Gathering: Unobtrusive Methods

Objectives

Understand the value of unobtrusive, (observational), methods in regards to defining a system purpose as well as its detailed needs or requirements.

Understand the concept of sampling.

How to construct useful samples.

How to create an analyst’s playscript.

Learn the STROBE, (Structured Observation of the Environment), observation technique.

Introduction

Unobtrusive refers to advanced skills of observation and listening. Is normally reserved for an analyst that has good status quo understanding as well as a business orientation. Assumes a higher level of analytical expertise and related accountability.

Can be very effective once a level of expertise is established and acknowledged.

Key unobtrusive methods include

Sampling

Investigation

Observation.

Should always be used in conjunction with interactive methods.

Discuss how a corporate coach begins.

Sampling systemically selecting representative elements

Sampling is used to

Contain costs

Increase speed in data gathering

Improve effectiveness pealing the onion

Reduce bias.

Sampling establishes first impressions, (we need to or don’t need to change), trends, (decisions are made outside of the system or double data entry is always required) planning by spreadsheet.

A good sampling process requires

A consistent decision on what data is collected for example reports or transaction volumes

Determine the population to be sampled for example last financial quarter, last holiday period, month-end reports, Key Performance Indicators (KPIs)

Choose the type of sample

Convenience respond if convenient

Purposive focus only on decision makers or data entry clerks, (doctor’s office example)

Simple random usually not effective, (remember the purpose is to determine a valued observation)

Complex random is systematic, stratified or represents a cluster, for example sample the direct users of data across a reasonable group size

Decide on sample size what is your objective?

To represent 90% of the clients or 20% of system data entry events it is here where industry trends and senior management assistance is required. Let’s discuss

Investigation - the act of discovery “we simply are not getting a return on our investment” or the “competition is leaving us in their wake” discuss speed of FDA drug submissions

Using quantitative documents

Can be high level like market share, stock price, financial reports or low level like order turn-around time, frequency of use look for reports that contradict each other, (we need to keep our inventory levels low but we also need 95% customer demand).

Using qualitative documents

Examples include,

Customer service levels SLAs, (service level agreements), compliance

Percent of staff training compliance

First time quality

Vendor or supplier quality

Audit results

Or can be a combination like product returns discuss why this is both quantitative and qualitative

Observing Decision Makers

Watching to establish a current state

Be consistent in who you watch

Be upfront

Provide feedback undercover boss scenario

Summarize observations using a playscript this is what managers do . . . .

Observing the Physical Environment

Observe how the clients use the technology, tools, reports, and data. Are their ergonomic issues?

Structured Observation of the Environment STROBE

Looking at the “big picture” across these concrete elements.

Office location how do the workers and managers interact? Is it an ivory tower or a bullpen environment?

Desk placement inclusive or exclusive

Stationary equipment where is data stored and why

Props what tools are used in communication and decision making. Let’s discuss

External information sources what is not well known or even hidden? Union negotiations, trade journals, isolated goals . . . .

Office lighting and color let’s discuss from your personal perspective.

Clothing the differences between button down, casual, uniforms

STROBE plays heavy on psychological implications and can work only if the decision makers are honest, open and interested in the results give my example regarding the most important role of senior management during project efforts.

Chapter 6 – Agile Modeling and Prototyping

Objectives

Understand the roots of agile modeling and the four main types of prototyping.

How prototyping is used to gather requirements.

What makes agile modeling different from other developmental methodologies?

The critical values to agile modeling.

Understand how to improve efficiency using either structured or agile methods.

Introduction

From Chapter 1, Agile Modeling

Is based on values, principles and core practices of an organization

Success is based on communication, simplicity, feedback and courage.

The approach focusses on the critical resources of time, cost, quality and scope.

Development is interactive and incremental.

The stages are

Exploration see if it’s possible

Planning an entity wide game played by the rules

Iterations to the initial release

Productionizing changing on the fly

Maintenance

Prototyping

Simply stated, is a method of sharing an evolving solution in an effort to surface opinions, observations, agreement and direction is further used in the detailed project stages.

Think of it as starting with a ball of clay . . . .

There are four prototype methods

Patched-Up building a system based on “parts” it may work but most likely is inefficient a new sports car should contain . . .

Nonoperational a non-working model used to test aspects of the system requirements usually focusses only on data input and outputs a clay figure of the proposed sports car that offers clarity on wind coefficients and look.

First-of-a-series represents a full scale model or a pilot a concept car.

Select Features operational model that is closer to the final solution there will be three sport models of our car.

As an alternative to the traditional SDLC (system development life cycle)

SDLC takes more time.

Is subjective to changes in requirements.

SDLC lacks flexibility you normally get to a point of no return.

Yet prototyping runs a risk of:

Rushing a solution to the market prior to understanding its impact thus resulting in a waste of effort or a poorly designed solution.

Discuss the realities of new product submissions.

The final approach is normally a combination.

Developing a Prototype

Guidelines

Work in manageable modules work on the engine and drive train independent from body and interior.

Build rapidly speed or progress can only be delivered if there is something to review look at a two door versus four door model.

Modify via iterations keep forming the car model.

Stress user interface the system or new car will be successful only if the clients can effectively interact with proposed changes or options in this regard a self-fulfilling prophecy is established.

COTS, (Commercial off the Shelf), Software typical Microsoft tools

A classic example is Microsoft Access, (an interactive data base design tool).

The benefits are obvious:

Avoids significant software expense

Uses already understood tools

Support data sharing across other tools.

The key in effective prototyping is user honestly and involvement.

Agile Modeling

The agile approach is not just based on results.

It is also based on values, principles and practices.

The four values of agile modeling

Communication lock down terminology

Simplicity begin with the obvious using process flows or pictures and build from that pealing the onion

Feedback concrete observations within an accountable time frame why did I place emphasis on accountable?

Courage the team must trust each other the process must be viewed as a collaboration Why?

Is there room within agile modeling for a resident expert?

Common ground in regard to shared principles is key in agile modeling.

Customers come first

Change must be embraced

Continuous improvement concept is a corner stone

Work daily during the modeling process everyone must get their hands dirty

Listen and trust

Keep things simple

Always focus on timely delivery

Activities, Resources and Practices

The four basic activities

Coding

Testing

Listening

Designing

Let’s discuss

Four resource control variables

Time there must be a starting and ending point think of a model ship that is never finished or a jigsaw puzzle that remains a set of scattered pieces.

Cost time is money do we need new tools just to begin like think pads, laptops, conference rooms.

Quality should we aspire for an “A” or “B” paper? Let’s discuss.

Scope offer a reasonable starting point and related alternatives scope creep can doom agile or any other method lets discuss.

Four core agile practices

Short releases as opposed to a major revision. Why?

Work within acceptable time and resource commitments should overtime be assumed? Refer to my failed Milan example of requirement gather where everything went wrong.

Hands on is a requirement can a key member contribute long distance across time zones or as a part time player?

Pair programming remember the trust or holistic strategy.

Agile Development as it relates to requirement gathering

Listen for user stories.

Transcribe stories into a process flow(s) that hopefully imply sequence and dependencies.

Create new user stories using the process flow(s).

Develop prototypes.

Begin the feedback cycle.

Comparing agile to structured methods

Let’s review diagram 6.8 page 169.