Part V

Plan, Stakeholders, Teams & Communication

A project is run through one document and carried out by people. This Part covers the formal Project Management Plan and how much Agile documents, then the people around a project: who the stakeholders are and how to engage them, how teams are built and how they are killed, how communication works and fails, and how to resolve conflict. Expect definitions, lists and case-study "analyse these stakeholders" questions in the exam.

  • You will tell a Project Charter from a Project Management Plan and know what baselining and change control do to the plan.
  • You will run a stakeholder analysis: identify, place on the power/interest grid, assess current → desired engagement, and turn it into a communication matrix (with the Medic case done in full).
  • You will compare team structures, name the seven teamicide practices, and calculate communication channels with n(n−1)/2.
  • You will choose a channel for a purpose, and a conflict-resolution mode for a situation.
5.1

The Project Management Plan exam

Once a project has been approved, dozens of people need to know the same things: what is in scope, when the milestones fall, how much money there is, who decides what, how risks and changes are handled. If each person keeps their own version, the project drifts apart. The Project Management Plan solves this by putting all of it in one approved document that everyone works from.

Definition · Project Management Plan (PMP)

A Project Management Plan is a formal, approved document that defines how the project is executed, monitored and controlled (PMBOK adds "and closed"). It may be a summary or a detailed document. The Project Manager owns and controls it, fills it in and uses it throughout the project, and for everyone involved it is the one source of truth (W04 slides).

Charter vs plan: two different documents

The PMP is easy to confuse with the Project Charter from Part I. A charter is the short proposal written before approval, to win agreement on the project's goals and terms. The PMP comes after approval and explains how those goals will actually be achieved. In one line: the charter wins approval, the plan runs the project.

Project Charter a few pages wins approval and buy-in approved → now mobilise Project Management Plan large, detailed, multi-page runs the project day to day often part of the business case owned by the PM · one source of truth
Short and persuasive, then long and operational. The charter is a summary proposal used to communicate, engage and gain approval; the plan is what the team uses to mobilise and manage the project every day.
Project Charter (Part I)Project Management Plan
PurposeSummary proposal to get approval for the goals and terms; often part of the business caseApproved document showing how to achieve the approved goals and benefits
SizeA few pagesLarge, detailed, multi-page
Used forCommunicating, engaging, gaining buy-in and approvalMobilising the project and managing it day to day

What goes in a PMP

The W04 slides split a typical PMP into two kinds of content. Project information says what the project is: its summary, money, people, scope, approach and dates. Project governance says how it will be controlled: who is responsible for what, and the planning activities every project must carry out. Each of those mandatory activities has its own Part on this site.

Project Management Plan PROJECT INFORMATION Executive summary Financial authority to proceed Key stakeholders Scope Delivery approach / SDLC Resources and people Key milestones Budget Lessons learned applied Constraints SDLC = software development lifecycle: waterfall or Agile (Part II) PROJECT GOVERNANCE Roles and responsibilities MANDATORY PLANNING ACTIVITIES SchedulePart VIII Risk managementPart IV Cost estimationPart VIII Quality assurancePart VII Configuration / change managementPart X
Information on the left, control on the right. The five mandatory planning activities are worth memorising as a list: schedule, risk, cost estimation, quality assurance, configuration/change management. Every one of them is a later Part of this subject.

The PM does not write all of this alone. Subject experts write the specialist parts (a test lead might write the quality section, for example), while the PM coordinates them and stays ultimately accountable. And a PMP is not a uniform template: each organisation tailors it to its own standards (PMBOK, Agile) and to the project in front of it.

What PMBOK adds: integration, baselines and the kick-off

PMBOK (the Project Management Body of Knowledge, the Project Management Institute's standard guide; here the 6th edition, chapter 4) treats the plan under Integration Management, the knowledge area about making all the separate pieces of a project fit together. It adds three ideas to the slides' picture.

1. The PMP integrates all the subsidiary plans and baselines. A subsidiary plan is a smaller plan for one area, such as the risk management plan. The PMP pulls all of them into one document, together with the baselines, the change and configuration management plans, and the chosen development approach.

SUBSIDIARY PLANS (PMBOK) Scope Requirements Schedule Cost Quality Resources Communications Risk Procurement Stakeholder engagement Project Management Plan integrates everything into one document BASELINES Scope · schedule · cost Change & configuration management plans + development approach
The PMP is a hub. Ten subsidiary plans, three baselines (gold), the change and configuration plans and the development approach (predictive, iterative, agile or hybrid) all meet in one document. That is what "integration" means.

2. Baselining. A baseline is an approved, frozen version of part of the plan (the scope, schedule or cost) that actual progress is later measured against. Before the baselines are set, the plan can be updated freely. Once it is baselined, it changes only through integrated change control: someone raises a change request, and a change control board (CCB), a group with the authority to approve or reject changes, decides. The plan is therefore progressively elaborated: it gets more detailed over time, but only through approved updates.

3. The kick-off meeting, held at the end of planning, communicates the objectives, gains the team's commitment and explains each stakeholder's role.

From draft to baseline, and how a baselined plan changes
Draft PMP updated freely Baselined scope · schedule · cost only via change control Kick-off meeting end of planning Executing the plan monitor and control INTEGRATED CHANGE CONTROL Change request raised and logged Impact evaluated effect on the plan CCB decides approve or reject Implemented plan re-baselined progressive elaboration
  1. The PM starts with a draft PMP. Until it is baselined, anyone working on the plan can update it freely (the dashed loop).
  2. The plan is baselined: the scope, schedule and cost versions are approved and frozen. From now on they change only through integrated change control, and progress will be measured against them.
  3. At the end of planning comes the kick-off meeting: it communicates the objectives, gains the team's commitment and explains each stakeholder's role.
  4. The project is executed, monitored and controlled against the plan. Then someone wants a change, so a change request is raised and logged. (This path is the one Template 1's change management plan sets out.)
  5. The change is evaluated for its impact on the plan before anyone decides.
  6. The change control board (CCB) decides whether to approve or reject the change.
  7. An approved change is implemented and the plan re-baselined. Repeating this loop is how the plan becomes progressively more detailed, without ever changing behind people's backs. (Part X covers change control in depth.)

Two sample templates

The subject shows two sample PMP templates, which make the idea concrete. Notice how much of the fuller template is about control (thresholds, change paths, communication rules) rather than about the product.

Template 1 · fuller

ProjectManagementDocs "SmartVoice"

  • Introduction taken from the charter.
  • Management approach: roles, authority, who approves funding.
  • Detailed scope, including what is out of scope.
  • Milestone list; schedule baseline and WBS (work breakdown structure, Part VIII), with escalation thresholds such as CPI/SPI outside 0.8–1.2 (cost and schedule performance indices, Part VIII).
  • Change management plan: requested → logged → impact evaluated → CCB decides → implemented and re-baselined.
  • Communications management plan: a communication matrix (type, frequency, format, participants, deliverable, owner), a team directory and conduct rules.
  • Cost, procurement, scope, schedule, quality and risk management plans; staffing plan and resource calendar; cost and quality baselines.
  • Closes with sponsor acceptance.
Template 2 · shorter

A lean plan

  • Planning basis: scope, milestones, phases, tasks, effort, resources.
  • Schedule, dependencies, assumptions and constraints.
  • Budget, risk log and change log.
Check yourselfMRQ: which statements about the Project Management Plan are correct? (Select all that apply.)
  1. It is a short summary used to win approval for the project's goals.
  2. The Project Manager owns and controls it, though subject experts write specialist parts.
  3. Once baselined, it changes only through integrated change control.
  4. It is a uniform template that every organisation fills in the same way.
  5. Under PMBOK it integrates the subsidiary plans and baselines.

B, C and E. A describes the Project Charter, not the PMP: the charter is the few-page approval document. D is wrong because the PMP is tailored to each organisation's standards and to the project. B, C and E are straight from the slides and PMBOK.

Check yourselfName the five mandatory planning activities that sit under "project governance" in a PMP.

Schedule, risk management, cost estimation, quality assurance, and configuration/change management. Together with roles and responsibilities, they make up the governance half of the plan.

In short
  • The PMP is the formal, approved "how" document: owned by the PM, the one source of truth. The charter is the short "should we?" document that comes before it.
  • It holds project information (summary, scope, budget, milestones…) and governance (roles plus schedule, risk, cost, quality, configuration/change).
  • PMBOK: it integrates all subsidiary plans and baselines; once baselined it changes only via change requests decided by a CCB; a kick-off meeting ends planning.
5.2

Agile: To Document or Not? exam

A formal project leans on a thick plan. Agile projects don't, and that has led to a popular misunderstanding: that Agile teams write nothing down. This section explains what Agile actually says about documents, why Agile teams still document, and what they document.

The myth and the reality

The myth that "Agile means no documentation" comes from misreading the Agile Manifesto's value "working software over comprehensive documentation" (Part III). The Manifesto values working software more; it does not dismiss documents. Some situations always need them: user stories, flowcharts, wireframes (rough screen sketches), notes of client meetings, and regulatory requirements. What Agile does is be "smart" about documentation, in three ways.

1 · how much

Just barely good enough (JBGE)

Enough documentation to support the project, without burdening the team.
2 · when

Just in time (JIT)

Created when it is needed rather than up front, so it does not go stale.
3 · what kind

Living and collaborative

Documents evolve with the product, and the team writes them together.
Up front written once, early Just in time written when needed, living: evolves sprint 1sprint 2sprint 3sprint 4sprint 5 big spec the product keeps changing… stale by release user story wireframe flowchart updated updated updated
Timing is the difference. A document written long before it is needed describes a product that no longer exists. Agile writes each document in the sprint that needs it and keeps it up to date (the green dashed arrows).

Why document in Agile at all? (four reasons)

The W04 slides give four reasons. Learn them as a list; they are a likely short-answer question, and they are the evidence you need to rebut "we're Agile, so we don't document".

  1. Stakeholder requirements and organisational memory. The software must be used, operated, supported, maintained and extended long after the team moves on, so what the team knows has to be kept.
  2. Communication with external groups, such as outsourced teams. Documents carry the information when stakeholders cannot always be present or the team cannot sit together.
  3. Audit and compliance. Some systems must follow a defined process and keep proof that it was followed. A medical system like Medic (the tutorials' case) is an example.
  4. Presenting, clarifying and solidifying complex ideas. Writing a complex idea down helps you present it, clarify it and make it solid.

What to document

PRODUCT LEVEL Product vision: costs, benefits, risks, milestones Project overview: contacts, technologies (onboarding) BUILDING IT Design decisions Requirements: use cases, user stories, prototypes, executable specs RUNNING AND KEEPING IT Operations docs: dependencies, backups, troubleshooting System docs: the architecture, so knowledge survives a new team SCRUM ARTEFACTS (AS THE SLIDES LIST THEM) Product vision Product Backlog Sprint goal Sprint Backlog Increments Burndown charts Def. of Done gold ring = one of the three artefacts the 2020 Scrum Guide names
Document what outlives the sprint. The top three bands keep the product understandable after the team has gone. The bottom band is the slides' list of Scrum artefacts; the official Scrum Guide recognises only the three ringed ones.
Slides vs Scrum Guide 2020

The slides list seven "Scrum artefacts". The 2020 Scrum Guide names only three: the Product Backlog, the Sprint Backlog and the Increment. Each has a commitment: the Product Goal, the Sprint Goal and the Definition of Done respectively. Burndown charts are a practice, not an official artefact. If a question asks for "the Scrum artefacts" in the Scrum Guide's sense, give the three.

Formal vs Agile: opposite dangers

The two approaches fail in opposite directions. A formal project has a baselined PMP and changes go through the CCB; its danger is too much documentation, which means bureaucracy, one of the team-killing practices called teamicide (§5.6). An Agile project has a vision and a backlog that the Product Owner re-orders; its danger is too little documentation, which means lost knowledge and failed audits.

Too little lost knowledge, failed audits Just barely good enough supports the project without burdening the team Too much bureaucracy, a cause of teamicide amount of documentation → Agile's danger vision + backlog, PO re-orders Formal's danger baselined PMP, changes via CCB
Aim for the green middle from either side. The question is never "documents or not?" but "how much, and when?".
Check yourselfA team lead says: "We're Agile, so we don't document." Discuss. (Self-test Q12)

It is a myth. The Manifesto values working software more than comprehensive documentation, but it does not reject documentation. Agile teams document just barely good enough, just in time, and as living documents. And they have good reasons to: stakeholder needs and organisational memory, communication with external or distributed teams, audit and compliance (a medical simulator such as Medic must prove its process), and clarifying complex ideas.

Check yourselfMCQ: Which is not one of the slides' four reasons to document in Agile?
  1. Audit and compliance
  2. Communication with external groups such as outsourced teams
  3. Fixing the full scope up front so it can be baselined
  4. Organisational memory for maintaining the system later

C. Fixing scope up front is a plan-driven idea; Agile documents just in time. The fourth real reason is presenting, clarifying and solidifying complex ideas.

In short

Agile values working software more than documents but still documents: just barely good enough, just in time, and living. The four reasons are organisational memory, external communication, audit/compliance and clarifying complex ideas. Formal projects risk too much paperwork; Agile ones risk too little.

5.3

Stakeholders and Stakeholder Analysis exam

Projects rarely fail because the code would not compile. They fail because someone with influence was ignored, surprised or turned against the project. Stakeholder management is the discipline of finding everyone who matters to a project, understanding what they want, and deliberately keeping them on side.

Definition · stakeholder, sponsor

A stakeholder is any individual, group or organisation that may affect, be affected by, or have an interest in the project or its outcome, directly or indirectly (W04 slides; PMBOK).

One stakeholder has a special role: the sponsor, who provides the funding and authority and champions the project.

Two opening scenarios (W04 slides)

The lecture opened with two situations every developer meets. Both are really stakeholder problems, and the model responses show the habits this section builds.

Scenario 1: the dashboard demo. At a sprint review (the Scrum meeting where stakeholders inspect what was built, Part III), the client calls a behaviour a "bug", and you are sure it is exactly what they asked for.

Client: "that's a bug!" at the sprint review ✗ don't argue or prove them wrong Listen and record the behaviour they expected Check the agreement user story + acceptance criteria matches what was agreed? yes no It's a change new backlog item; Product Owner prioritises A real defect fix it Lesson record agreements (JBGE documentation) and demo often
A disputed "bug" is settled by the written agreement, not by an argument. If the software matches the agreed user story and acceptance criteria, the client's request is a change, which goes to the backlog for the Product Owner to prioritise. That only works if agreements were recorded.

Scenario 2: the workflow change. Users say the new dashboard approval process is "more complicated than the old way", even though it reduces errors and speeds approvals. Listen first, because some of the complaints may reveal real usability problems. Then explain why the change was made, with evidence, and support people through it with training, quick guides and champion users (respected users who help their colleagues). Pilot the change, gather feedback and iterate.

Unaware Resistant Supportive not engaged early engaged early RECOVERING RESISTANT USERS 1 Listen first: complaints may reveal real usability problems 2 Explain why, with evidence 3 Support them: training, quick guides, champion users 4 Pilot, gather feedback, and iterate
The root cause was late engagement. The users were not involved early, so they went from unaware to resistant instead of to supportive. These are engagement levels, defined in §5.4.

Stakeholder analysis: five activities

Stakeholder analysis is how you get from "who is out there?" to "what do we do about each of them?". The W04 slides break it into five activities. It is a cycle, not a one-off: as the project changes, you re-assess.

The five stakeholder-analysis activities
1 · Identify everyone affected or interested 2 · Understand interests objectives, expectations, concerns 3 · Assess influence power over outcomes, resources 4 · Prioritise engagement importance, influence, impact 5 · Manage relationships anticipate, defuse, keep talking re-assess as the project changes Five activities,one loop Who is affected or interested?directly or indirectly What do they want, expect,and worry about? How much power do they haveover outcomes and resources? Who needs attention first?rank by importance, influence, impact How do we keep them on side?reactions, conflicts, open communication …and round againpositions change during a project
  1. Stakeholder analysis is five activities in a loop. Step through them one at a time.
  2. 1 · Identification. Find everyone who is affected by or interested in the project, directly or indirectly. Missing someone here is the most expensive mistake, because nothing later in the loop can fix it.
  3. 2 · Understanding interests. Work out each stakeholder's objectives, expectations and concerns.
  4. 3 · Assessing influence. Judge their power over outcomes, decisions and resources. Together with step 2 this places them on the power/interest grid (§5.4).
  5. 4 · Prioritising engagement. Rank them by importance, influence and impact, so effort goes where it matters most.
  6. 5 · Managing relationships. Anticipate their reactions, address potential conflicts and keep communication open.
  7. Then re-assess as the project changes. A crisis, a new phase or a handover can move a stakeholder from irrelevant to critical (the Medic case in §5.5 shows both).

Who the stakeholders typically are

A useful first cut is internal (inside the organisation running the project) versus external (outside it). The slides' typical lists are drawn below; an exam answer that names stakeholders from both rings looks thorough.

EXTERNAL INTERNAL The project Shareholders The board Sponsor / business mgrs Management Project manager Project team Employees End users / customers Suppliers Governments Unions Competitors Local communities, public Related institutions
Look in both rings. Inside the organisation: shareholders, the board, the sponsor or business managers, management, the PM, the team and employees. Outside: end users or customers, suppliers, governments, unions, local communities and the public, related institutions, and even competitors.

The stakeholder register

The first output of identification is a stakeholder register: a simple table with one row per stakeholder under the columns Name, Position, Internal/External, Project role, Contact. The slide example has these rows (names and contact details are left out here):

NamePositionInternal / ExternalProject roleContact
…VP of OperationsInternalProject sponsor…
…CFOInternalApproves funds…
…CIOInternalThe PM's boss…
…External consultantExternalProject manager…

Note the last row: the project manager can be an outsider. In the Medic case, too, the PM works for BYTE Consulting, not for the hospital.

Executive roles and what each cares about

Senior executives are usually the most powerful stakeholders, and each one cares about something different. So the same project must be framed differently for each: talk to the CFO about payback, not about clinical safety. The W04 addendum describes these roles.

The board CEO final say; can override other executives talk: strategic value, reputation CFO financial planning, financial risk, capital allocation cost, benefits, payback COO day-to-day operations, efficiency operational impact, rollout CIO IT strategy, infrastructure, security, data governance IT standards, supportability CTO tech stack, R&D, innovation; overlaps the CIO architecture, technology choice CMO Chief Medical Officer: clinical direction, care quality, compliance clinical fidelity, safety, ethics Product manager: owns the product roadmap Project manager: delivers; power delegated by sponsoring execs
Purple text is what to talk to each executive about. The CEO has the final say and is high power, high interest. Note the bottom row: the project manager has no power of their own; it is delegated by the sponsoring executives, which is why keeping the sponsor engaged matters so much.
Check yourselfList the five stakeholder-analysis activities in order.

(1) Identify stakeholders; (2) understand their interests and concerns; (3) assess their influence or power; (4) prioritise engagement; (5) manage relationships. Then re-assess as the project changes, because it is a cycle.

Check yourselfYou are pitching a system upgrade to the CFO and then to the CIO. What do you emphasise to each?

To the CFO: cost, benefits and payback, because the CFO looks after financial planning, financial risk and capital allocation, and holds power over funding. To the CIO: IT standards and supportability, because the CIO is responsible for IT strategy, infrastructure, security and data governance.

In short

A stakeholder is anyone who may affect, be affected by, or have an interest in the project; the sponsor funds and champions it. Analysis is a five-step cycle: identify, understand interests, assess influence, prioritise, manage relationships. Record them in a register, look both inside and outside the organisation, and frame the project in each executive's own terms.

5.4

Engagement Levels and the Power/Interest Grid exam

Once you know who the stakeholders are, you need two judgements about each: how engaged are they now, and how engaged do we need them to be? and how much attention do they deserve? The engagement levels answer the first; the power/interest grid answers the second.

Five levels of engagement (W04 slides; PMBOK)

1 · Unaware doesn't know about the project or its impact 2 · Resistant aware, but resists change 3 · Neutral aware, neither supportive nor resistant 4 · Supportive aware and supports the change 5 · Leading champion: aware and drives the change C D close the gap more engaged →
Each level adds something to the one below. Everyone above "unaware" knows about the project; what differs is their attitude, from resisting it to driving it. The C and D markers show how PMBOK's engagement assessment matrix records a stakeholder: Current level C, Desired level D.

PMBOK's engagement assessment matrix is a table with the stakeholders as rows and these five levels as columns. In each row you mark the current (C) and desired (D) level. The gaps between C and D are what drive the engagement and communication actions: a stakeholder already at their desired level needs maintaining, while a big gap needs a plan. §5.5 fills in the matrix for Medic.

The stakeholder management plan and the stakeholder analysis

The stakeholder management plan can include the current and desired engagement levels, the relationships between stakeholders, their communication requirements, a strategy for each stakeholder, and how the plan itself will be updated.

It is sensitive

A document that says "Mary is resistant; we plan to win her over" would cause real damage if Mary read it. So the stakeholder management plan is prepared by the PM and a few others, kept out of the official project documents, distributed narrowly, and parts of it may never be written down at all.

A stakeholder analysis records each stakeholder's name and organisation, project role, unique facts, interest, influence and management strategies. The slide example shows how it turns into concrete strategies (power / interest given in that order):

StakeholderPower / interestEngagementStrategy
BrianHigh / highLeadingManage closely: ask his advice; give him short in-person updates.
MaryHigh / mediumResistantShe pushes corporate IT standards that the preferred solution breaks. Convince her, and show her that her work is still respected.
Finance teamMedium / highResistantControls the funds: involve them in a detailed review of costs and benefits.
JessicaHigh / lowNeutralKeep her satisfied with financial justification, and ask her to talk to Mary.

The power/interest grid

The grid places each stakeholder by two judgements from the analysis: how much power they have over the project (up the side) and how much interest they take in it (across). The quadrant they land in tells you the strategy.

INTEREST → POWER → HighLow LowHigh Keep satisfied periodic updates, high-level summaries; deal with their needs efficiently; interest may be low, but their decisions can make or break it e.g. delegating executives, regulators, utility company Manage closely engage actively, communicate frequently, ask for input and feedback; they heavily shape the outcomes e.g. CEO, sponsor, major investors, key deciders Monitor keep communication minimal, but watch for emerging issues; ignoring them can cost reputation e.g. vendors, suppliers Keep informed consult them, address their concerns; use their support to influence higher-power stakeholders e.g. end users, customer service staff
The quadrant name is the strategy. Red marks where the most effort goes. Note that "keep informed" stakeholders are useful allies: their support can sway the powerful ones above them.

Placing the slide example on the grid makes the strategies above visible, including one move that uses a relationship between stakeholders.

INTEREST → POWER → KEEP SATISFIEDMANAGE CLOSELYMONITORKEEP INFORMED LowMediumHigh talks to Mary 4 Jessica · neutral 2 Mary · resistant 1 Brian · leading 3 Finance team · resistant
Use the relationships between stakeholders. Jessica (4) is neutral and powerful but not very interested. The plan keeps her satisfied and asks her to talk to Mary (2), the powerful resister. Brian (1) is managed closely; the finance team (3) controls funds, so it is brought into a detailed cost–benefit review.
Pitfall · positions move

A crisis can put a regulator into "manage closely" overnight. Re-assess regularly, and document the assumptions behind each placement.

Pitfall · "Monitor" ≠ ignore

Low-power, low-interest stakeholders still get watched, because ignoring emerging issues can cost reputation.

INTEREST → POWER → KEEP SATISFIEDMANAGE CLOSELYMONITORKEEP INFORMED Regulator Regulator a crisis, overnight
The grid is a snapshot. Re-run the analysis whenever the situation changes; §5.5 shows exactly this happening in the Medic case.

Formal vs Agile: how stakeholder needs reach the team

The slides contrast two cases. In the formal case, an e-commerce platform for a clothing retailer, the internal stakeholders are IT (the PM, developers and testers), marketing, sales and management (the CEO and CFO), and the external ones are the payment processor and the web host. In the agile case, a food-delivery app, the internal stakeholders are the Product Owner, who represents stakeholders' needs, the Developers, and the Scrum Master, who facilitates communication; the external ones are customers and restaurant partners.

The real difference is the route by which stakeholder needs reach the team.

FORMAL PROJECT Stakeholders documentedrequirements gate reviews communicationmatrix Project team PM, developers, testers
Formal: needs travel on paper. Documented requirements, gate reviews and a communication matrix carry stakeholder needs to the team.
SCRUM PROJECT Stakeholders Product Owner the single channel Product Backlog Developers Increment inspected at every sprint review
Scrum: needs flow through one person. The Product Owner is the single channel into the backlog, and stakeholders inspect the increment at every sprint review.

One model that is not in the slides is the salience model (Mitchell et al.), which classifies stakeholders by power, legitimacy and urgency. Know that it exists; the grid is what the course uses.

Check yourselfMCQ: A powerful government regulator currently shows little interest in your project. Which strategy fits?
  1. Manage closely
  2. Keep satisfied
  3. Keep informed
  4. Monitor

B, keep satisfied (high power, low interest): periodic updates and high-level summaries, and deal with their needs efficiently. Their interest is low, but their decisions can make or break the project; and a crisis could move them to "manage closely" overnight.

Check yourselfMRQ: Which are true of the stakeholder management plan? (Select all that apply.)
  1. It records current and desired engagement levels and a strategy per stakeholder.
  2. It should be published with the official project documents for transparency.
  3. It is prepared by the PM and a few others and distributed narrowly.
  4. Parts of it may never be written down.

A, C and D. B is the trap: the plan is sensitive and is kept out of the official project documents.

Check yourselfIn Scrum, who is the single channel for stakeholder needs into the backlog, and when do stakeholders see progress?

The Product Owner represents stakeholders' needs and is the single channel into the Product Backlog. Stakeholders inspect the increment at every sprint review. (Formal projects instead rely on documented requirements, gate reviews and a communication matrix.)

In short
  • Five engagement levels: unaware, resistant, neutral, supportive, leading. Record current (C) and desired (D); the gaps drive the actions.
  • Grid: manage closely (high/high), keep satisfied (high power, low interest), keep informed (low power, high interest), monitor (low/low).
  • Positions move, "monitor" is not "ignore", and the stakeholder plan is sensitive.
5.5

Medic: A Full Stakeholder Analysis exam

The exam's case-study questions ask you to apply all of the above to a scenario. Tutorial 4 does exactly that for the Medic case, the running case of the tutorials, and this section works through its answers.

The case in brief (Tutorial 4 assumptions)

Medic is a surgical-training simulator being built for the hospital RVEEH: trainee surgeons practise on 3D models and feel realistic resistance through haptic (touch-feedback) hardware. You are BYTE Consulting's PM on Project Virtual Vision, a $2M AUD project running for 24 months from 1 Sep 2026. Fatemeh and Craig are taken to be the case's two client surgeons, and the grid is for the delivery phase. The BYTE team has skill gaps in C++, 3D and haptics (Part IV treats these as risks), and the hospital's own IT support team prefers C++.

Placing the Medic stakeholders on the grid

Medic stakeholders, quadrant by quadrant
KEEP SATISFIEDMANAGE CLOSELYMONITORKEEP INFORMED INTEREST → POWER → HighLow LowHigh STAKEHOLDERS (DELIVERY PHASE) 2 3 4 5 2Tom Hanks, CMO 3Jane Smith, CTO (effective sponsor) 4Fatemeh Hashemi, haptic liaison 5Craig McMillan, Head of ENT, UI liaison 1 8 1RVEEH Board (approved the budget) 8BYTE management 6 9 6Trainee surgeons (end users) 7Mengyao Li + 25 support engineers 9Project team (Venkat, Rameez, Kimberly…) 7 10 11 10Haptic / 3D hardware vendors 11Indemnifiers, implant makers, patients, media, government 7 7 at handover
  1. The empty grid. Power runs up, interest across. We fill it one quadrant at a time, for the delivery phase.
  2. Manage closely. 2 Tom Hanks, the CMO: clinical authority, champions haptic fidelity. 3 Jane Smith, the CTO, accountable for delivery and effectively the sponsor: controls technology direction and acceptance. 4 Fatemeh Hashemi, Haptic Touch Liaison (medium–high power): signs off haptic realism and is contracted to work day to day. 5 Craig McMillan, Head of ENT and UI liaison: the UI must be minimal and non-intrusive.
  3. Keep satisfied. 1 The RVEEH Board approved the budget (high power, low–medium interest): it can cancel or re-fund the project and wants milestone and budget status. 8 BYTE management (high power, medium interest) cares about the commercial outcome and reputation.
  4. Keep informed. 6 Trainee surgeons, the end users (low power, high interest): usability matters, so involve them in usability tests. 7 Mengyao Li and 25 support engineers from RVEEH IT (medium power, medium and rising interest; prefer C++): they can refuse to accept a system they cannot support. 9 The project team, Venkat (UI), Rameez (3D modelling), Kimberly and 12 inexperienced developers, Anthony and 5 testers: low–medium power (their power is expert power), high interest, involved daily.
  5. Monitor. 10 Haptic/3D hardware vendors (low–medium power, low interest), though they are a critical dependency. 11 Indemnifiers, implant makers, patients, media and government: low/low until a crisis; they are indirect beneficiaries.
  6. The grid moves. As handover approaches, the support engineers' interest rises and they move into manage closely: they will run the system after BYTE leaves.

Current → desired engagement

The tutorial also gives each stakeholder's current and desired engagement level (C → D). The picture below is the engagement assessment matrix drawn as a chart. Notice how many of the client's people need to move up to Leading.

UNAWARERESISTANTNEUTRALSUPPORTIVELEADING RVEEH Board Tom Hanks, CMO Jane Smith, CTO Fatemeh Hashemi Craig McMillan Trainee surgeons Mengyao Li + support engineers BYTE management Project team Hardware vendors Indemnifiers, patients, media… current (C) desired (D) already at the desired level
The arrows are the work. Jane, Fatemeh, Craig and the project team must move from supportive to leading; trainees from unaware to supportive (involve them in usability tests); the support engineers from neutral to supportive. Grey dots need maintaining, not moving.
StakeholderPower / InterestQuadrantC → DRationale
RVEEH Board (approved budget)High / Low–MedKeep satisfiedSupportiveCan cancel or re-fund; wants milestone and budget status
Tom Hanks, CMOHigh / HighManage closelyLeadingClinical authority; champions haptic fidelity
Jane Smith, CTO (accountable for delivery, effectively sponsor)High / HighManage closelySupportive → LeadingControls technology direction and acceptance
Fatemeh Hashemi, Haptic Touch LiaisonMed–High / HighManage closelySupportive → LeadingSigns off haptic realism; contracted to work day to day
Craig McMillan, Head of ENT, UI LiaisonHigh / HighManage closelySupportive → LeadingUI must be minimal and non-intrusive
Trainee surgeons (end users)Low / HighKeep informedUnaware → SupportiveUsability; involve them in usability tests
Mengyao Li and 25 support engineers (RVEEH IT; prefer C++)Med / Med, risingKeep informed, then manage closely at handoverNeutral → SupportiveCan refuse to accept an unsupportable system
BYTE managementHigh / MedKeep satisfiedSupportiveCommercial outcome, reputation
Project team: Venkat (UI), Rameez (3D modelling), Kimberly and 12 inexperienced devs, Anthony and 5 testersLow–Med (expert power) / HighKeep informed, involved dailySupportive → LeadingSkill gaps in C++, 3D and haptics
Haptic/3D hardware vendorsLow–Med / LowMonitorNeutralCritical dependency
Indemnifiers, implant makers, patients, media, governmentLow / Low (until a crisis)MonitorUnawareIndirect beneficiaries

Justifying a placement (tutorial check questions)

The tutorial tested whether students could justify a position rather than just state it. The pattern in the answers: interest depends on your assumptions, while power, here, comes with the position.

QuestionAnswerWhy
Can CMO Tom Hanks be low interest?YesIf he delegates to his liaison.
Can he be high interest?YesBecause he champions the haptic requirement.
Can he be low power?NoHis power comes with the CMO position.
Can Craig McMillan be low power?NoHe is head of department and the UI acceptance liaison.
How this is examined

A stakeholder question on a case wants each placement justified with case facts ("high power: as Head of ENT he signs off UI acceptance"), the strategy that follows from the quadrant, and the engagement gap. State your assumptions when interest is ambiguous, exactly as the tutorial did for Tom Hanks.

From analysis to a communication matrix

The analysis then becomes a plan of who is told what, how, how often and by whom: the communication matrix. These are the tutorial's columns and answers.

Stakeholder / meetingObjectiveFormatFrequencyOwnerImportance
Jane Smith: status meetingProgress, risks, change approvalsFace to face or online, plus status reportWeeklyPMHigh
Tom HanksHaptic progress, clinical decisionsBriefing plus one-page summaryFortnightlyPMHigh
Fatemeh and CraigRequirements, demos, acceptanceWorking sessions; sprint reviewDaily; each sprintPM / Product OwnerHigh
RVEEH BoardMilestones, budget, go/no-goFormal report and presentationQuarterly and at milestonesJane Smith with PMHigh
Trainee surgeonsUsability tests, trainingWorkshops, web portalEach releaseFatemeh/CraigMedium
Mengyao Li's teamSupportability, handover, knowledge transferMeetings, documentation portalMonthly; weekly near handoverPM, KimberlyMedium → High
Project teamCoordinationStand-up, chat, task boardDailyKimberly / Scrum MasterHigh
Daily Eachsprint Weekly Fortnightly Monthly Quarterly +milestones Eachrelease Jane Smith: status meeting Tom Hanks Fatemeh and Craig RVEEH Board Trainee surgeons Mengyao Li's team Project team near handover high importance medium
Frequency follows the grid. The people managed closely hear from the project daily to fortnightly; the Board, which is kept satisfied, gets a formal report quarterly and at milestones. The support team's meetings speed up as handover nears, matching its move on the grid.

A new challenge: the crisis

The Medic crisis: re-running the analysis
KEEP SATISFIEDMANAGE CLOSELYMONITORKEEP INFORMED INTEREST → POWER → Tom Hanks, CMO RVEEH Board RVEEH Board Patients Media and public Government Indemnifiers Patients Media and public Government Indemnifiers Surgeons · supportive Surgeons · may turn resistant THE EVENT Patients complain of pain after surgery bynewly simulator-trained surgeons. The pressreports it, and the government opens anenquiry. Government enquiry, regulators Patients and families Media and public Indemnifiers NEW HIGH-POWER, HIGH-INTEREST enquiry, regulators, patients, media, indemnifiers RVEEH Board Hospital legal and PR teams EXISTING STAKEHOLDERS SHIFT Board and CMO intensely interested; project may pause; legal and PR join; surgeons may resist ACTIONS 1 Re-run the stakeholder analysis2 Route all external comms via PR and legal3 Cooperate transparently with the enquiry4 Preserve requirements, test, haptic evidence5 Add the new risks to the risk register6 Raise the reporting frequency
  1. Before the crisis. The CMO is managed closely, the Board kept satisfied, surgeons are supportive, and patients, media, government and indemnifiers sit in "monitor": low power, low interest.
  2. The event. During delivery, patients complain of pain after surgery by newly simulator-trained surgeons, the press reports it, and the government opens an enquiry.
  3. New high-power, high-interest stakeholders appear in "manage closely": the government enquiry and regulators; patients and families, whose power comes through complaints, the media and the law; the media and public; and indemnifiers. The "monitor" quadrant has emptied upwards.
  4. Existing stakeholders shift. The Board and the CMO become intensely interested and the project may be paused; the hospital's legal and PR teams join; and surgeons may become resistant.
  5. The actions follow. Re-run the analysis; route all external communication through PR and legal; cooperate transparently with the enquiry (Part IX, ethics); preserve the requirements, test and haptic-validation evidence; add the new risks to the register (Part IV); and raise the reporting frequency. This is exactly why audit documentation matters even in Agile (§5.2).
Check yourselfWhy do the Medic support engineers (Mengyao Li's team) move into "manage closely" at handover, and what does that change in the communication matrix?

They will operate and maintain the system after BYTE leaves, and they can refuse to accept a system they cannot support (they prefer C++). Their interest rises as handover nears, so they move up the grid. In the matrix their meetings go from monthly to weekly near handover, and their importance rises from medium to high.

Check yourselfAirline quiz (tutorial): an airline system issues automatic refunds exactly as specified, but customer service reviews refunds manually and finance fears revenue loss. What is the situation?
  1. A software defect the developers should fix
  2. A requirement gap: key stakeholders were never consulted
  3. A change that needs a formal change request
  4. Both B and C

D, both B and C. There is no defect: the system does what was specified. It is a requirement gap, because key stakeholders (customer service, finance) were never consulted, and fixing it needs a formal change request. It is a lesson in step 1 of stakeholder analysis: identification.

Check yourselfSelf-test Q13: Medic patients complain after surgery by simulator-trained surgeons; the press reports it and the government opens an enquiry. Comment on the stakeholders.

The enquiry, regulators, the media and patients become new high-power, high-interest stakeholders. The Board and the CMO become intensely interested, and surgeons may resist. The PM should redo the analysis, communicate externally only via PR and legal, cooperate transparently and preserve the evidence, add the new risks, and report more often.

In short

In Medic, the four client decision-makers (Tom, Jane, Fatemeh, Craig) are managed closely, the Board and BYTE management kept satisfied, trainees, support engineers and the team kept informed, and vendors and the wider public monitored. Most client people must move up to leading. The analysis feeds a communication matrix, and a crisis forces you to re-run it.

5.6

Teams exam

Software is built by teams, and how a team is put together and treated has a large effect on whether the project succeeds. This section defines a team, compares the ways a team can be structured, and lists the management habits known to destroy one.

Group vs team

Definition · group, team

A group is a collection of people. A team is a group that is (or is expected to be) cohesive and focused on a common purpose or goal. So all teams are groups, but not all groups are teams.

GROUP a collection of people groups teams every team is a group; not every group is a team TEAM goal cohesive, focused on a common goal
What turns a group into a team is the shared goal and cohesion, not the number of people or the fact that they sit together.

Why teams? (six reasons from the slides)

1 · sizeMost software is too large for one programmer to build in the time available.
2 · complexityLarge, complex or mission-critical systems need many people working together effectively.
3 · diversityComponents written by people from different backgrounds better meet the needs of a broad user population.
4 · learningMembers argue, but they also learn from each other.
5 · employabilityTeamwork is an Engineers Australia Stage 1 competency, and employers consistently rate it.
6 · successTeam structure affects project success.

The problems students fear in group work, such as coordination, uneven workload and unreliable teammates, also occur in industry. They are handled the same way there: through risk management, task and time tracking, and peer feedback.

Four team structures

A team can be structured in four ways, which differ in who makes the decisions. The first three each have a characteristic problem; the fourth, the collaborative structure, avoids them, and it is the one Agile teams use.

AUTOCRACY boss Boss gives orders single point of failure; little collaboration, morale suffers ANARCHY ? ? ? ? ? No leader nobody knows what to do DEMOCRATIC vote Every decision collective bogged down in decisions; diluted responsibility ("nobody's fault") COLLABORATIVE consult UI 3D code test each decides in their own area Flat; decide within your expertise Agile teams work this way
Who decides? One boss (autocracy), nobody (anarchy), everybody about everything (democracy), or whoever has the relevant expertise, consulting only those who share it (collaborative). The collaborative structure avoids both the boss as a single point of failure and the democratic team's paralysis.
StructureDecisionsProblems / benefits
AutocracyBoss gives ordersLittle collaboration; morale suffers; boss is a single point of failure
AnarchyNo leaderNobody knows what to do
DemocraticEvery decision made collectivelyBogged down in decisions; diluted responsibility ("nobody's fault")
Collaborative examFlat structure; members decide within their area of expertise and consult only those with relevant expertiseAvoids the problems of both autocracy and democracy. Agile teams are collaborative

XP teams and Scrum teams

In XP (Extreme Programming, Part III), all code is written by pair programming: the driver writes the code while the navigator reviews it, and one member writes the test cases while the other does the testing. Because two people know every piece of code, knowledge is not lost when someone leaves, and juniors and seniors learn from each other.

Driver writes the code Navigator reviews it one writes the test cases, the other tests knowledge isn't lost; juniors and seniors learn
XP: two people, one keyboard. Pairing spreads knowledge across the team.
SCRUM TEAM · 7 ± 2 (SLIDES) Product Owner one person; decides what is built, in what order Scrum Master ensures Scrum is followed; protects from outside barriers Developers decide how to deliver the Sprint Backlog fixed-length sprints, preferably 2–4 weeks
Scrum: three accountabilities. Collaborative, with a slightly democratic flavour.

A Scrum team is a collaborative and slightly democratic variant, and the slides give its size as 7 ± 2. Work happens in fixed-length sprints, preferably 2–4 weeks long. The team agrees how much work to take on, and members sign up for tasks. Several mechanisms keep everyone aligned: the client sees progress every sprint, everyone sees everyone's work, and burndown charts track the work remaining (Part VI).

Slides vs Scrum Guide 2020

The slides say 7 ± 2 people and sprints of preferably 2–4 weeks. The 2020 Scrum Guide says "typically 10 or fewer people" and sprints of at most one month.

Teamicide: seven ways to kill a team exam

DeMarco and Lister (Peopleware, 1987) could not write a recipe for a team that works well, but they could list the things that kill one. They called this teamicide.

the team enthusiasm, focus 1 · Defensive managementsignals distrust of the team 2 · Bureaucracymindless paperwork 3 · Physical separationkills casual, creative contact 4 · Fragmented timeproject switching kills focus 5 · Quality-reduced productshipping poor work demoralises 6 · Phony deadlinesartificially tight dates 7 · Clique controlbreaking teams up (rotation)
Seven practices, all done by management. Notice the links between them: phony deadlines are a form of defensive management, and bureaucracy is the "too much documentation" danger from §5.2.
  1. Defensive management. Management shows the team that it isn't trusted.
  2. Bureaucracy. Mindless paperwork kills enthusiasm.
  3. Physical separation. It prevents spontaneous, casual interaction, which is what drives creativity.
  4. Fragmentation of time. Switching between projects prevents concentration.
  5. Quality-reduced product. Shipping poor work to cut cost or time is demoralising.
  6. Phony deadlines. Artificially tight dates; a form of defensive management.
  7. Clique control. Policies that break up teams, for example regularly rotating people.
Check yourselfSelf-test Q16: Management makes developers log every 15 minutes of work, sets a "deadline" two weeks before the real one, and splits each person across three projects. Name three teamicide practices.
  • Bureaucracy and defensive management: the 15-minute logging is paperwork that signals distrust.
  • Phony deadlines: the fake "deadline" two weeks before the real one.
  • Fragmentation of time: splitting each person across three projects.

Distributed (virtual) teams

A distributed team is spread across locations and relies on communication technology; such teams are also called geographically dispersed or remote teams. Organisations use them to serve a region, because staff live there, because people work from home, or to get access to high-level or lower-cost talent.

Costs (W04 slides)

Communication becomes less frequent and less rich, because there are fewer visual and behavioural cues. There are cultural and time-zone overheads, and visual work such as design is especially hard.

Benefits (W04 slides)

Fewer biases; it is easier to contact local stakeholders; and the team can work a 24-hour cycle, handing work across time zones.

The Wikipedia reading adds structure. It gives four defining features: geographic dispersion, electronic dependence, national diversity and dynamic structure. It gives three enabling conditions: shared understanding, integration across organisational systems, and mutual trust. And it describes a lifecycle:

The lifecycle of a distributed team (Wikipedia reading)
Preparation Launch Performancemanagement Teamdevelopment Disbanding Five stages from forming the team to winding it upStep through to see what the reading says matters at each stage. Preparation: deciding to work distributedReasons the slides give: to serve a region, because staff live there, because people work from home,or to reach high-level or lower-cost talent. Launch: ideally a face-to-face kick-offMeeting in person once, at the start, clarifies the team's goals, roles and rules,the same job the kick-off meeting does for a PMP. Performance management: practices that help a remote team work • agreed norms for technology use• agendas sent in advance• regular calls• rotating presenters• shared repositories• "virtual water-cooler" chat Team development: the three enabling conditions • shared understanding• integration across organisational systems• mutual trust DisbandingThe team is wound up when its work is done; the lifecycle ends.
  1. The reading describes a distributed team's life as five stages, from preparation to disbanding.
  2. Preparation. The organisation decides to use a distributed team, for one of the reasons the slides list.
  3. Launch, ideally with a face-to-face kick-off to clarify goals, roles and rules.
  4. Performance management. The reading's mitigations keep remote work effective: agreed norms for technology use, agendas in advance, regular calls, rotating presenters, shared repositories and "virtual water-cooler" chat.
  5. Team development. The team grows on the three enabling conditions: shared understanding, integration across organisational systems, and above all mutual trust.
  6. Disbanding ends the lifecycle.

Finally, an HBR article (Ferrazzi 2014) reports that only 44% of virtual-team workers find virtual communication as productive as face to face.

Not covered: Tuckman's stages

You may have met Tuckman's team-development stages elsewhere. The guide notes that they are not in the 2026 slides, so they are not taught here; the team topics to know are the ones above.

Check yourselfMCQ: In which team structure do members decide within their own area of expertise and consult only those with relevant expertise?
  1. Autocracy
  2. Anarchy
  3. Democratic
  4. Collaborative

D, collaborative. It avoids the autocracy's single point of failure and the democracy's diluted responsibility, and it is how Agile teams work. Democratic is the near miss: there every decision is collective.

Check yourselfSelf-test Q9 (predecessor exam, 433-443 2002 Q2): give at least two reasons teams are necessary today.
  1. Systems are too large and complex for one person in the time available, and parallel work shortens schedules.
  2. Systems need specialist skills that no single person has.
  3. Diverse backgrounds give better coverage of users' needs, and members review each other's work and learn from each other.
  4. Knowledge survives when someone leaves.
In short
  • A team is a cohesive group with a common goal. Structures: autocracy, anarchy, democratic, collaborative; Agile teams are collaborative.
  • XP pairs a driver with a navigator; a Scrum team is 7 ± 2 (slides) with a Product Owner, a Scrum Master and Developers.
  • Teamicide: defensive management, bureaucracy, physical separation, fragmented time, quality-reduced product, phony deadlines, clique control.
  • Distributed teams lose rich communication but gain reach; mutual trust is the key enabling condition.
5.7

How Communication Works and Fails exam

Everything in this Part so far, from stakeholder engagement to team cohesion, runs on communication, and communication goes wrong far more often than people expect. This section explains why it matters so much, how a message travels from one person to another, and where it gets distorted on the way.

Why it is in this subject

Communication management is an ACS accreditation requirement (the Australian Computer Society accredits the degree), and employers treat communication and teamwork as the two pillars of graduate employability; one survey ranks communication 1st and teamwork 4th of 14 skills. Yet graduates often lack them. Fewer than half of employers think graduates are adequately prepared (McKinsey), and Deloitte and PwC retrained Covid-era graduates in 2023.

Challenges at three levels (W04 slides)

ORGANISATION (PMI) political cultural linguistic: each discipline's jargon TEAM status silos information overload lack of communication protocol (the rules) INDIVIDUAL semantics perception channel feedback anxiety culture semantics = meaning of the words · perception = how they are interpreted
Problems nest. A message can be lost to one person's interpretation, to a team's silos or overload, or to the organisation's politics and jargon, and often to several at once.

The communication model (Shannon 1948)

Claude Shannon's 1948 model of a communication system breaks the journey of a message into five parts, with noise corrupting the signal on the way. The project-management version adds feedback and context.

A message's journey, then the PM version
Information sourceproduces the message Transmitterencodes it as a signal Channelthe medium Receiverdecodes it Destinationthe intended person Noise jargon · culture · emotion · distraction senderencodechannel / mediumdecodereceiver feedback: check what was understood CONTEXT: TIME · PLACE · STRUCTURE
  1. Shannon's system has five parts. Follow one message through them.
  2. The information source produces the message: say, a PM who wants to tell the client the release will slip a week.
  3. The transmitter encodes it into a signal: the PM chooses words and writes an email.
  4. The channel is the medium that carries it: here email (other channels are meetings, phone, a web portal).
  5. On the way, noise corrupts the signal. In projects, noise is jargon, culture, emotion and distraction: the client skims the email between meetings.
  6. The receiver decodes the signal back into a meaning, which may not be the one intended.
  7. The destination is the person the message is meant for.
  8. The project-management version renames the parts (sender, encode, channel, decode, receiver) and adds feedback, so the sender can check what was understood, and context (time, place, structure). The key point: the receiver rarely interprets a message exactly as the sender intended, so check understanding through feedback. Shannon set meaning aside; project communication is all about meaning.

Listening

Half of communication is receiving it. There are two kinds of listening, and the slides illustrate them with a lecture and a tutorial.

PASSIVE · A LECTURE speaker takes information in without processing or reacting
Passive listening. Information flows one way, and nobody checks it landed.
ACTIVE (EMPATHETIC) · A TUTORIAL message questions, feedback ? shows interest, asks questions, keeps eye contact, doesn't interrupt, avoids distractions, reads verbal and nonverbal messages
Active listening. The listener works: questions and feedback close the loop that the model above needs.
Benefits of active listening

It shows concern, gets you better information and encourages further communication. It also strengthens relationships, calms upset people, invites others to listen to you, and improves cooperation and problem solving.

Barriers to listening

Physiological limits; missing background information; selective memory or expectations; fear of being persuaded; bias, boredom or emotion; distractions such as phones and noise; the physical environment; cultural differences; past experiences; jargon and acronyms.

Key PM communication skills (W04 slides)

The skills a PM needs are: conveying their view, motivating and influencing, delegating, solving problems, presenting updates, setting goals and a vision, managing conflict, networking and negotiating. In practice, a successful PM must be able to read the client, run a meeting, communicate accurately in writing and speech, manage the team, influence the environment, and secure alignment and buy-in.

Check yourselfName the five parts of Shannon's communication system in order, and say what the project-management version adds.

Information source → transmitter (encodes) → channel (medium) → receiver (decodes) → destination, with noise corrupting the signal. The PM version adds feedback (to check understanding, because the receiver rarely decodes exactly what was intended) and context (time, place, structure).

Check yourselfMRQ: Which are communication challenges the slides place at the team level? (Select all that apply.)
  1. Silos
  2. Semantics
  3. Information overload
  4. Political obstacles
  5. Status

A, C and E (team level also includes lack of communication and protocol). Semantics is an individual-level challenge; political obstacles are at the organisation level (PMI).

In short

Communication fails at individual, team and organisational levels. Shannon's model runs source → transmitter → channel → receiver → destination with noise; PM adds feedback and context, because the receiver rarely decodes what was meant. Active listening closes the loop.

5.8

Communication Plans, Channels and Methods exam

A large share of a PM's time goes on communication; the PMI paper puts it at about 90%. Left to chance, that time is wasted on the wrong people, the wrong medium and the wrong moment. A communication plan makes it deliberate, and some simple arithmetic shows why it gets harder as a team grows.

The communication plan

A good plan lets the PM be proactive, sets a common understanding of what happens when, and clarifies who is responsible and who delivers what. You build it from the stakeholder analysis, using the 5 Ws and 1 H checklist from PMI: who, what, when, where, why and how. According to the W04 slides it defines eleven things.

Stakeholder analysis register, grid, C → D 5 Ws + 1 H who what when where why how THE COMMUNICATION PLAN DEFINES 1 what is communicated (detail, format) 2 the channel (meetings, email, phone, portal) 3 when: frequency, formal and informal 4 who is responsible 5 stakeholders' communication needs 6 resources allocated to communication 7 sensitive info: handling, who authorises 8 the flow of communications 9 constraints 10 standard templates and formats 11 escalation process for conflicts, issues
Analysis in, plan out. The stakeholder analysis says who matters and how much; the 5 Ws and 1 H turn that into the eleven decisions the plan records. Item 11, an escalation process, says what happens when communication itself breaks down.

Template 1 (§5.1) also adds rules for "communications conduct": agendas go out 2 days ahead, minutes within 24 hours, and emails are kept short and professional.

Communication channels: n(n−1)/2 exam

Why does communication get harder as a team grows? Because every pair of people is a potential line of communication, a channel, that can carry a misunderstanding. The number of potential channels between n people is a measure of communication complexity:

channels = n(n − 1) / 2

The reasoning: each of the n people can talk to n − 1 others, giving n(n − 1) ordered pairs, but "A talks to B" and "B talks to A" are the same channel, so divide by 2.

Adding people one at a time
P1 P2 P1 P2 P3 P1 P2 P3 P4 P1 P2 P3 P4 P5 P1 P2 P3 P4 P5 P6 n = 2 → 1 channel n = 3 → 3 channels n = 4 → 6 channels n = 5 → 10 channels n = 6 → 15 channels PEOPLE nNEW CHANNELSTOTAL 2+11 3+23 4+36 5+410 6+515 channels = n(n − 1) / 2 adding one person to a team of n adds exactly n new channels so the total grows roughly with n²
  1. Two people, one channel. P1 and P2 can only talk to each other: 2 × 1 / 2 = 1.
  2. Add P3 (highlighted). P3 needs a line to each of the 2 people already there, so +2 channels: 3 × 2 / 2 = 3.
  3. Add P4: +3 new channels, total 4 × 3 / 2 = 6.
  4. Add P5: +4, total 5 × 4 / 2 = 10. This is the reading's example: 5 people, 10 channels.
  5. Add P6: +5, total 6 × 5 / 2 = 15. Headcount went from 2 to 6 (×3); channels went from 1 to 15 (×15).
  6. The pattern: adding one person to a team of n adds exactly n channels (the newcomer links to everyone already there), so the total is 1 + 2 + … + (n − 1) = n(n − 1)/2, which grows roughly with n².

Worked examples

5 people:  5 × 4 / 2 = 10 channels        (the reading's example)
6 people:  6 × 5 / 2 = 15 channels
9 people:  9 × 8 / 2 = 36 channels

Team grows 6 → 9:
  headcount  6 → 9      +3   = +50%
  channels  15 → 36     +21  = 21/15 = +140%
  check: the 7th, 8th and 9th people add 6 + 7 + 8 = 21 channels ✓
TEAM GROWS FROM 6 TO 9 Headcount6 → 9 +50% Channels15 → 36 +140% same three new people, nearly three times the relative growth in channels (+21)
Channels outrun headcount. A 50% bigger team has 140% more channels.
channels 0100200300 1510152025 people in the team (n) 9 → 36 25 → 300 headcount n
The curve bends up. Channels grow with roughly n², while headcount (dashed) grows in a straight line.

The same growth explains a famous observation: because channels grow roughly with n², adding people to a late project can make it later. Each newcomer adds communication work for everyone already there.

Medic: 25 people, 300 channels

The Medic core team is the PM (1) + Venkat and Rameez (2) + Kimberly and 12 developers (13) + Anthony and 5 testers (6) + Fatemeh, Craig and Mengyao (3) = 25 people. That gives 25 × 24 / 2 = 300 channels. Nobody can manage 300 conversations, so the PM must structure communication through leads, stand-ups and the communication matrix.

25 people, 300 potential channels
Unstructured. Every pair is a potential channel: 300 places for a message to go wrong.
PM FatemehCraigMengyao KimberlyAnthonyVenkatRameez 12 developers 5 testers structured through leads + daily stand-ups and the communication matrix
Structured. The PM talks to the client liaisons and the team leads; leads talk to their people; stand-ups and the matrix carry the rest. The 300 channels still exist, but the PM no longer has to manage them one by one.
Check yourselfSelf-test Q11: A team grows from 7 to 10 people. By how much do the communication channels change?
C(7)  = 7 × 6 / 2 = 21
C(10) = 10 × 9 / 2 = 45
increase = 45 − 21 = 24   →  24/21 = +114%
headcount: 7 → 10 = +43%
check: the 8th, 9th and 10th people add 7 + 8 + 9 = 24 ✓

Channels increase by 24 (+114%) for a +43% increase in headcount.

Check yourselfMCQ: A team of 12 hires one more developer. How many new potential channels does that create?
  1. 1
  2. 12
  3. 13
  4. 78

B, 12. Adding one person to a team of n adds exactly n channels (the newcomer links to each of the 12). Check: C(13) − C(12) = 78 − 66 = 12. D (78) is the new total, the near miss.

Face to face matters, but read Mehrabian carefully

The slides quote Mehrabian's 7–38–55 rule: 7% of meaning comes from the words, 38% from tone and 55% from body language. The lesson is to pay attention to more than the words, because nonverbal cues shape how others feel about us and how we feel about ourselves.

7% 38% 55% wordstone of voicebody language Misuse warning (UTPB reading) Only for inferring a speaker's feelings or attitude when words and nonverbal cues conflict. It does not mean 93% of all information is nonverbal. Read cues with the 3 Cs: context, clusters, congruence.
A famous number with a narrow meaning. The defensible claim is that tone and body language strongly shape interpretation, not that words barely matter.

The slides also list some items to remember. The receiver rarely decodes exactly what was intended. Geography and culture add complexity, through working hours, language and norms. Communication helps manage conflict. Practise, and seek feedback. And choose the channel appropriately.

Channel suitability (W04 slides) exam

The slides rate how well each channel suits each purpose, from 1 = most suited to 3 = least suited. The pattern to take away: relational or emotional matters go face to face; anything that needs a record goes in writing.

Hard copyPhoneVoicemailEmailMeeting /face to faceWebsite SUITED TO… Confirming commitments 1 3 3 1 2 1 Building consensus 3 2 3 3 1 3 Mediating conflict 3 2 3 3 1 3 Resolving misunderstanding 3 1 3 3 1 3 Addressing negative behaviour 3 2 3 2 1 3 Expressing support / appreciation 1 1 2 1 1 1 Encouraging creative thinking 2 3 3 2 1 3
Read down the gold column. Face to face scores 1 on every relational purpose (consensus, conflict, misunderstanding, negative behaviour, creativity) and only 2 on confirming commitments. For confirming commitments, written channels with a record (hard copy, email, website) score 1. Expressing appreciation works almost anywhere.
Check yourselfSelf-test Q14: A manager says "only 7% of communication is words, so email is pointless." Evaluate.

This misapplies Mehrabian: 7–38–55 concerns judging feelings when words and cues conflict, not all information. Email is in fact among the best channels for confirming commitments and keeping records (it scores 1), while face to face is best for consensus, conflict and misunderstandings. The point is to match the channel to the purpose.

Check yourselfMCQ: Two team members have fallen out over a code review. According to the slides' channel-suitability table, which channel is best for mediating this conflict?
  1. Email, so there is a record
  2. Phone
  3. A face-to-face meeting
  4. A post on the project website

C. Mediating conflict scores 1 for a meeting/face to face, 2 for phone and 3 for email, hard copy, voicemail and website. Email is the trap: a record matters for commitments, not for emotions.

The communications charter

A communications charter is a collectively designed, mutually agreed summary of a group's communication norms and practices. It is "guidelines for best practice, not a rule book" (UMD), and it is a living document. It sets out which tools to use for which tasks (synchronous, like a call, or asynchronous, like chat), formats, response times, meeting rules (attendance, cameras, minutes, time zones), how decisions are recorded, and when to escalate.

A charter saves time and cuts message volume. It also includes remote members and helps across cultures, for example with people who are reluctant to report problems to a senior. Its limitations: it can stifle informal interaction and creativity, and it takes effort. Forbes's advice is to build it with the team, so that they own it. Mindtools gives six steps for creating one:

  1. Define the purpose.
  2. Decide how to create it.
  3. Define what it covers.
  4. Set guidelines.
  5. Incorporate feedback.
  6. Implement and update it.
Check yourselfHow does a communication plan differ from a communications charter?

The communication plan is the PM's plan, built from the stakeholder analysis with the 5 Ws and 1 H: what is communicated, through which channel, when, by whom, to meet which stakeholders' needs, with an escalation process. The communications charter is a team's own collectively agreed norms (tools, response times, meeting rules, decision records), "guidelines, not a rule book", built with the team so they own it.

In short
  • The communication plan defines eleven things, built from the stakeholder analysis with the 5 Ws and 1 H.
  • Channels = n(n − 1)/2; adding a person to a team of n adds n channels. 6 → 9 people: 15 → 36 channels (+140%). Medic's 25 people: 300 channels, so structure through leads.
  • Mehrabian's 7–38–55 applies only to feelings when cues conflict. Relational matters go face to face; records go in writing.
5.9

Conflict and Conflict Resolution exam

Put people with different goals, opinions and pressures on one project and they will disagree. The question is not how to avoid conflict but how to resolve it so that the project gains rather than suffers.

Definition · conflict

Conflict results from differences in perception, opinion or beliefs: incompatible goals, thoughts or emotions (Villax & Anantatmula, PMI 2010). The slides call it "the single most undermanaged activity in projects", and left unresolved, it will destroy a project.

Common causes (W04 slides). Conflict on projects commonly arises over the schedule, intellectual disagreements, personalities, project priorities, manpower, technical opinions, administration and cost.

Views of conflict, and which kinds help

THREE VIEWS OF CONFLICT Traditionalavoid it Behaviouralit is natural, so manage it Interactionistsome is needed for performance WHETHER IT HELPS DEPENDS ON ITS KIND Too little: groupthink pressure for unanimity and self-censorship; a factor in NASA's Challenger and Columbia counter: a safe climate and a devil's advocate Task conflict about the work can improve decisions where trust and psychological safety are high Relationship conflict emotional, personal almost always harmful
Aim for the green middle. The interactionist view says some conflict is needed, but only the right kind: disagreement about the work in a team that trusts itself. Too little conflict is as dangerous as the personal kind.

Five conflict-resolution modes (PMI reading; PMBOK names in brackets)

The PMI reading describes five ways of resolving a conflict. Each produces a different outcome and fits a different situation. Collaborating is preferred whenever possible, although it takes time.

Forcing(direct) ✓ ✗ youthem win–lose use authority toimpose your solution emergencies,safety or legal issues Accommodating(smoothing) ✗ ✓ youthem lose–win let the other partyhave its way matters more to them;preserve relationship Avoiding(withdrawing) ? ? youthem problem festers ignore it;take no action trivial issues;cooling off Compromising(reconciling) ½ ½ youthem partial win, both partly satisfies both,fully satisfies neither equal power,time pressure Collaborating(problem solving) ✓ ✓ youthem win–win understand both sides;fully satisfy both preferred wheneverpossible; takes time
Match the mode to the situation. Each card shows the outcome for each side and, in purple, when to use it. Forcing is right in an emergency; avoiding is right for trivia; collaborating is the default whenever there is time.
ModeWhat it isOutcomeUse when
Forcing (direct)Use authority to impose your solution regardless of the other sideWin–loseEmergencies, safety or legal issues
Accommodating (smoothing)Let the other party have its wayLose–winThe issue matters more to them; to preserve the relationship
Avoiding (withdrawing)Ignore it; take no actionProblem festersTrivial issues; cooling off
Compromising (reconciling)A solution partly satisfying both sides, fully satisfying neitherPartial win for bothEqual power, time pressure
Collaborating (problem solving)Understand both sides' concerns and find a solution that fully satisfies bothWin–winPreferred whenever possible; takes time
The first step (lecture poll)

When a conflict arises, the first step is to discuss the problems and the reasoning causing the conflict, openly, with the people involved, before escalating to a mentor. Never attack each other or lobby behind people's backs.

Matching strategy to source (PMI table, condensed)

The right strategy also depends on where the conflict comes from.

SOURCE OF CONFLICT STRATEGY Relationships collaborating, building trust, active listening, team building Goals and priorities SMART goals, clear roles, appropriate PM authority Authority or role ambiguity define responsibilities at the start Interface (outside groups) involve them in planning early Culture training and awareness
Diagnose before you treat. Several of these are prevention rather than cure: defining responsibilities and involving outside groups early stop the conflict before it starts. (SMART goals are specific, measurable, achievable, relevant and time-bound.)

PM skills for conflict (projectmanagement.com): communication, active listening, negotiation, collaboration (aiming for win–win), stress management and emotional awareness.

Check yourselfSelf-test Q15: Which conflict mode fits? (a) Two senior developers disagree on Medic's rendering architecture. (b) A production security breach needs an immediate decision. (c) A trivial naming argument the day before a release.
  • (a) Collaborating. Explore both designs, perhaps spike them (build a quick experiment), and decide on evidence, because task conflict is worth the time.
  • (b) Forcing. The PM decides now: it is an emergency.
  • (c) Avoiding or compromising for now, then revisit it in the retrospective.
Check yourselfMCQ: Which conflict-resolution mode produces a "lose–win" outcome?
  1. Forcing
  2. Accommodating
  3. Avoiding
  4. Compromising

B, accommodating (smoothing): you let the other party have its way. Forcing is win–lose, avoiding lets the problem fester, compromising is a partial win for both.

Check yourselfWhat is groupthink, and how do you counter it?

Groupthink is the pressure for unanimity and self-censorship that comes from too little conflict; it was a factor in NASA's Challenger and Columbia failures. Counter it with a safe climate (psychological safety, so people voice doubts) and a devil's advocate who deliberately argues the other side.

In short

Conflict is the most undermanaged activity in projects. Task conflict can help where trust is high; relationship conflict harms; too little breeds groupthink. The five modes are forcing (win–lose), accommodating (lose–win), avoiding (festers), compromising (partial) and collaborating (win–win, preferred). Start by discussing the problem openly with the people involved.

5.10

Kinds of Power

A PM must get things done through people who often do not report to them. Which kinds of power actually work? This is an older exam topic, not in the 2026 slides, but it appears in the predecessor subjects' past papers (Self-test Q8).

Definition · power

"Power is the potential ability to influence behaviour." French and Raven describe five bases of power, judged here from a project manager's point of view.

HIGHLY EFFECTIVE MODERATE / LIMITED LEAST EFFECTIVE Expertrespected knowledge, skill; esp. technical staff Referentliked, respected, identified with; commitment RewardPMs rarely control pay; recognition motivates Legitimateformal position; compliance, not commitment Coercivepunishment; resentment breeds hidden problems and turnover; a last resort
The powers a PM can earn beat the powers a PM is given. Expertise and being respected build commitment; formal position gets mere compliance; punishment backfires.
PowerSourceEffectiveness for a PM
LegitimateFormal positionWorks for routine direction, but PMs often have little formal authority (in matrix organisations, where staff report to functional managers). Gets compliance, not commitment
RewardGiving pay, assignments, recognitionModerate. PMs rarely control pay, but challenging work and recognition motivate
CoercivePunishmentLeast effective. Breeds resentment, hidden problems and turnover. Last resort
ExpertRespected knowledge and skillHighly effective, especially with technical staff
ReferentBeing liked, respected, identified withHighly effective; builds commitment

Studies of project managers (Thamhain & Wilemon) link expertise and work challenge most strongly to success, and penalty power least. Since a PM's formal power comes from the sponsoring executives, keep the sponsor engaged. (This is also why the Medic project team's power is called expert power in §5.5.)

Check yourselfSelf-test Q8 (predecessor exam, 433-443 2001 Q10): list three kinds of power and how effective each is for a PM.
  • Expert power is highly effective and sustainable.
  • Referent power is highly effective and builds commitment.
  • Coercive power has low effectiveness: it breeds resentment and is a last resort.

Legitimate power is often weak for a PM in a matrix organisation.

In short

Five bases: legitimate, reward, coercive, expert, referent. Expert and referent are highly effective; coercive is least effective; legitimate gets compliance, not commitment. A PM's formal power is borrowed from the sponsor.

5.11

Self-Test: Past-Exam Questions exam

The questions tagged "predecessor exam" are real questions from the exams of SWEN90016's predecessor subjects (433-443 and 433-341). No SWEN90016 paper is public; the model answers are the study guide's own, not official. Questions Q8, Q9 and Q11–Q16 appear earlier in this Part, next to the material they test.

Answering people questions

These questions reward structure: name the concept (a teamicide practice, an engagement level, a conflict mode), then apply it to the scenario's facts, then say what you would do. For "how would you run the meeting" questions, give example quotations, as the model answers below do.

Q1 · predecessor exam 433-443 1998 Q4Describe four people-management activities and a risk for each.
  1. Staff acquisition. Risk: key skills are unavailable. Medic's developers, for example, lack C++ and haptics skills.
  2. Team organisation and role definition. Risk: role ambiguity causes conflict, and an autocratic leader is a single point of failure.
  3. Motivation and performance management (feedback, recognition, appraisal). Risk: demotivation and turnover, made worse by teamicide.
  4. Communication and conflict management. Risk: misunderstandings, or unresolved conflict that wrecks the project.

(Training is also acceptable; its risk is that the time it takes is underestimated.)

Q2 · predecessor exam 433-443 1998 Q10Who are a software project's stakeholders? Give two problems each, then the two most significant.
  • The sponsor or senior management may withdraw support, or impose unrealistic deadlines.
  • The client may give vague or changing requirements, or be unavailable to approve.
  • End users may resist change, or not be consulted, so that the requirements have gaps.
  • The PM may have little authority, and can become a communication bottleneck.
  • The development team may have skill gaps, and may suffer turnover and give optimistic estimates.
  • IT support may be consulted too late, or impose technology constraints.
  • Suppliers and regulators may deliver late, or change the compliance rules.

Most significant: (1) unclear or changing requirements from poor client and user involvement, a leading cause of failure, and every estimate depends on the requirements; (2) weak or divided executive support: without it, resources, decisions and conflict resolution fail, and the project may be cancelled.

Q3 · predecessor exam 433-443 1999 Q6A graduate tells you a senior team member calls him "Dumbo" and "stupid". How do you run the meeting, and what follow-up do you take? Give example quotations.

The meeting should be private and unhurried, and you should listen actively.

  • "Thank you for telling me. You were right to come to me."
  • "What was said, when, and who was there?"
  • "This isn't acceptable here. I'll handle it discreetly and get back to you by Friday."
  • Don't promise secrecy you can't keep.

Follow-up:

  • Meet the senior member privately, and describe the behaviour and its impact: "Calling a colleague 'Dumbo' and 'stupid' is bullying and breaches our policy and professional code (Part IX). It stops now. If you have a performance concern about him, bring it to me."
  • Document both meetings, and escalate to HR if it repeats.
  • Check back with the graduate ("Has anything changed?").
  • Reinforce respectful team norms, and consider a mentor for the graduate.

Note the channel choice: addressing negative behaviour scores 1 for face to face in the channel-suitability table (§5.8).

Q4 · predecessor exams 433-443 1999 Q7 and 2001 Q2Identify the sponsor and the stakeholders of a university assignment and exam results system that users access directly.

The sponsor is the head of the administration department (for example the Academic Registrar), who owns the cost-saving business case and funds it.

The stakeholders are:

  • administration staff, whose work changes;
  • lecturers and tutors, who enter marks;
  • heads of department and exam boards, who approve results;
  • students, who view results directly and care about privacy and accuracy;
  • the university executive;
  • IT development, operations, security and help desk;
  • privacy officers and auditors;
  • external stakeholders: government reporting bodies, scholarship providers and vendors.
Q5 · predecessor exam 433-443 2000 Q3Mark, a good coder, recently got a pay rise and now attends client meetings. He looks unhappy and arrives later. (a) Give reasons related to teamwork. (b) What would people-focused managers do?

(a) Several teamwork reasons could explain it.

  • He has been pulled away from his team, so he suffers isolation and a loss of cohesion.
  • He suffers fragmentation of time between coding and client work (a teamicide practice).
  • Teammates may resent his raise, so he is excluded.
  • There is role ambiguity, and his workload grew without his coding being reduced.
  • Client work is stressful, and he has had no training or recognition for it.

(b) People-focused managers would:

  • talk to him privately and listen actively to find the cause;
  • clarify his role and rebalance his workload;
  • offer client-skills training or a mentor;
  • give him recognition and discuss his career;
  • reconnect him with the team and keep checking in. This also manages the turnover risk.
Q6 · predecessor exam 433-443 2000 Q5(a) What are stakeholders, and why analyse them? (b) Senior management is divided and the project may be cancelled. How does this affect planning, risk, scheduling and the process model?

(a) A stakeholder is any individual, group or organisation that may affect, be affected by, or have an interest in the project or its outcome (§5.3). Analysis finds everyone with an interest or influence and captures their needs, so that the requirements are complete. It also prioritises effort, plans communication, anticipates resistance and secures support.

(b) A divided senior management affects each of the four areas:

  • Planning: map supporters and opponents and strengthen the sponsor. Manage sceptical executives closely, with a strong business case, and communicate more often.
  • Risk: put cancellation and resource withdrawal at the top of the register.
  • Scheduling: front-load visible, high-value deliverables, use short milestones and go/no-go points, and avoid long-lead commitments.
  • Process model: choose an incremental, Agile or spiral model. Early usable increments demonstrate value and build support, and they leave something useful if the project is cancelled, whereas waterfall delivers value only at the end (Part II).
Q7 · predecessor exam 433-443 2001 Q9Frank, newly promoted to team leader, picked team members who are clones of himself and based task estimates on his own past task times, to be used for everyone. What feedback do you give him?
  • Praise his enthusiasm and early planning.
  • On the clone team: with no diversity of skills or views, groupthink is likely. There will be gaps in testing, documentation and deployment, which he needs in order to put the application into service, and everyone will want to do design and coding. He should build the team from the skills the work requires.
  • On estimates from his own times: he is the top performer, so the plan will systematically underestimate others' tasks, which means overtime, missed dates and low morale (like phony deadlines). He should estimate with the people doing the work (consensus or three-point estimates, Part VI and Part VIII), use team history, add contingency and re-calibrate from actuals.
  • His job is now leading, so he should delegate and communicate.
Q10 · predecessor exam 433-341 2008 Q5Name two stakeholders of the regional library system and say why.
  • Librarians are the main users: they check items out and in, impose fines and run monthly reports, and their manual workflow changes completely.
  • Members (borrowers) register with two forms of ID, borrow books, videos and games, and receive overdue and fine notices. Under-18 restrictions apply, and their personal data must be protected.

(Also accept administrators, or the library's management as sponsor.)

In short

People questions are answered with named concepts applied to the scenario's facts: stakeholders with their problems, teamicide practices behind a demotivated employee, conflict and communication choices with example wording. Divided executive support favours incremental, Agile or spiral delivery, because early increments build support and survive cancellation.