Project Management & Agile flashcards
Flashcards for the track: Scrum, Kanban, schedule, earned value, risk, planning, people, governance and careers.
241 cards. Read the question, answer it in your head, then open the card to check.
-
What tells a project apart from operations?
A project is temporary and works in a unique context, with a planned start and end and a result that has not been made in exactly this way before. Operations repeat to keep the organisation running, such as payroll or backups, and have no planned end. Size decides neither.
From the lesson What a project is and what project leads do
-
How does a programme differ from a portfolio?
A programme runs related projects together for a benefit that none would deliver alone; a portfolio is all the work an organisation chooses to fund for its strategy, and a portfolio decision starts, pauses or stops work.
From the lesson What a project is and what project leads do
-
Who does the 2020 Scrum Guide make accountable in a Scrum Team, and does it name a project manager?
It names three accountabilities and no project manager: the Product Owner (maximising product value and ordering the Product Backlog), the Developers (their plan for each Sprint and quality through a Definition of Done) and the Scrum Master (establishing Scrum and the team's effectiveness).
From the lesson What a project is and what project leads do
-
On what does the 2026 PMI outline say a project's success is now judged?
On what stakeholders receive in value and on the outcomes they were after, not only on scope, time and cost. A project ends at hand-over but is judged later, so someone must own measuring the benefit once the team has gone.
From the lesson What a project is and what project leads do
-
What does India's PAIMANA portal count, and how big is the count?
It monitors central government infrastructure projects of ₹150 crore and above. One month's flash report counted 1,981 ongoing projects across 17 ministries, with a revised cost of ₹42.78 lakh crore, of which about 47.59 % had been spent.
From the lesson What a project is and what project leads do
-
How does iterative delivery differ from incremental delivery?
Iterative work repeats cycles that refine the same solution; users may see nothing usable until the end. Incremental work hands over usable pieces one after another. Adaptive approaches such as Scrum combine both: each short cycle refines what was learned and delivers a usable increment.
From the lesson Choosing predictive, agile or hybrid delivery
-
What are the five delivery approaches this track names?
Predictive (full plan up front, result at the end), iterative (cycles that refine), incremental (usable pieces one after another), adaptive (short cycles that do both), and hybrid (fixed spine of milestones and gates with cycles between them).
From the lesson Choosing predictive, agile or hybrid delivery
-
Which two questions pick between predictive and adaptive work?
How clear are the requirements, and how clear is the technology? When both are clear, a full plan up front costs least. When either is unclear, short cycles find the gaps while they are still cheap to fix.
From the lesson Choosing predictive, agile or hybrid delivery
-
Which Cynefin contexts fit plan-driven work, and which fits adaptive?
The 2007 Snowden and Boone article names simple, complicated, complex, chaotic and disorder. Plan-driven work suits simple and complicated problems. Adaptive work suits complex ones, where right answers cannot be worked out in advance and patterns emerge from safe-to-fail experiments.
From the lesson Choosing predictive, agile or hybrid delivery
-
What share of the 2026 PMP exam items represents adaptive and hybrid approaches, and across how many domains?
About 60 % of the items; 40 % represent predictive. The three approaches are not kept in one part of the exam: they appear throughout all three domains (People 33 %, Process 41 %, Business Environment 26 %).
From the lesson Choosing predictive, agile or hybrid delivery
-
What is tailoring in project management?
Deciding, for one project, which phases, gates, cycle lengths and reports it needs, from its risk, its rules and its stakeholders. A step that protects neither a constraint nor learning can be left out.
From the lesson Choosing predictive, agile or hybrid delivery
-
Standard, method, framework, guide: what does each promise?
A standard sets agreed principles that fit any approach (ISO 21502). A method prescribes a way of working with named processes and roles (PRINCE2). A framework gives a minimal structure and leaves the rest to the team (Scrum). A guide is simply the document that describes one of them.
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
How many principles and performance domains does the PMBOK Guide 8th Edition keep?
Six principles (adopt a holistic view, focus on value, embed quality, lead accountably, integrate sustainability, build empowered teams) and seven performance domains (governance, scope, schedule, finance, stakeholders, resources and risk).
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
What does PRINCE2 Version 7 count seven of?
Seven principles, seven practices and seven processes, with a people element and sustainability built into project performance. It is a method, so it prescribes how to run the project step by step.
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
What does the 2020 Scrum Guide define?
Three accountabilities (Product Owner, Developers, Scrum Master), five events (the Sprint plus Sprint Planning, Daily Scrum, Sprint Review and Sprint Retrospective) and three artifacts, each with a commitment. The Guide calls Scrum purposefully incomplete.
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
What does the Kanban Guide v2025.5 define?
Kanban, as a strategy for optimising how value flows through a process, carried out by three practices (define and visualise the workflow, actively manage items in the workflow, improve the workflow) and measured by four flow metrics.
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
What do Nexus, LeSS, Scrum@Scale and SAFe each build on?
Nexus joins about three to nine Scrum Teams around one Product Backlog; LeSS calls itself Scrum for two to about eight teams; Scrum@Scale extends a single Scrum Team's way of working across networks of teams; SAFe forms Agile Release Trains using its own versions of Scrum and Kanban.
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
Under what licence are the Scrum Guide and Kanban Guide offered, and what does it mean?
Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0). Anyone may share and adapt them, but any adaptation must carry the same licence. That is why this track quotes briefly with credit and writes its own explanations.
From the lesson PMBOK, PRINCE2, Scrum and Kanban: the map
-
What parts make up a business case, and what option is always among the options?
Problem or opportunity, options, costs of each, benefits, risks and a recommendation. Doing nothing is always one of the options, because every other option must beat the cost of leaving things as they are.
From the lesson Business cases and choosing between projects
-
What question does each of payback, NPV and IRR answer?
Payback asks how soon the money spent comes back, ignoring the time value of money. NPV asks how much value an option adds today at the organisation's required rate. IRR is the rate at which NPV would be exactly zero.
From the lesson Business cases and choosing between projects
-
Which measure chooses between options that exclude each other, and why?
NPV at the organisation's required rate: it shows how much value each option adds today. A high IRR on a small sum can look better than a slower option that adds far more value; payback hides everything after the payback year.
From the lesson Business cases and choosing between projects
-
What does weighted scoring compare, and when must the weights be agreed?
Options whose benefits have no honest price: agree the criteria and how much each matters, score every option on each criterion and add the weighted scores. Agree the weights before anyone sees the scores, because the ranking is only as good as the weights, and let each scorer mark alone first, since scores given in an open meeting drift towards the most senior person's view.
From the lesson Business cases and choosing between projects
-
Is money already spent a reason to continue a project?
No. Money already spent, a sunk cost, is no reason to continue: the decision is always about the costs and benefits still to come. The planning fallacy distorts cases too, since Buehler, Griffin and Ross showed in 1994 that people underestimate how long their own tasks will take.
From the lesson Business cases and choosing between projects
-
What does a project charter authorise, and who issues it?
It gives the project lead the right to use the organisation's people and money for the project. The sponsor (or project initiator) issues it, because the sponsor answers for the business case; the lead usually drafts it.
From the lesson Writing a project charter with clear objectives
-
What three parts make an objective measurable?
A target number ("from 15 to 5 minutes"), a deadline ("by week 26") and a source, the place the figure will be read from ("a timed sample of 30 teachers" or "the submission log"). A target nobody can read is no better than no target.
From the lesson Writing a project charter with clear objectives
-
How do an assumption and a constraint differ in a charter?
An assumption is something the plan takes to be true without proof, so it needs a test by a date. A constraint is a fixed limit the project must work within, such as a budget cap or a legal date. A constraint filed as an assumption looks negotiable; an assumption filed as a constraint never gets tested.
From the lesson Writing a project charter with clear objectives
-
Does a Scrum Product Goal replace a project charter?
No. The 2020 Scrum Guide calls the Product Goal the Scrum Team's long-term objective, so it gives direction, but it grants no money or authority and binds nobody outside the team. A charter is still needed when someone must grant money or authority or outsiders rely on the commitment.
From the lesson Writing a project charter with clear objectives
-
What three tests does the Kanban Guide's definition give for a stakeholder?
Anyone responsible for, interested in or affected by the work and what it produces. A first list usually passes only the first test, so a search by group (who pays, uses, builds, permits, lives near or competes with the work) finds the rest.
From the lesson Identifying and analysing stakeholders
-
What does a power-interest grid give you, and what does it miss?
Four approaches by score: manage closely, keep satisfied, keep informed, monitor. It misses whether a stakeholder's claim is legitimate and whether it is urgent, which Mitchell, Agle and Wood's salience model (1997) adds. The grid is a snapshot and must be revisited at every milestone.
From the lesson Identifying and analysing stakeholders
-
Which three yes-or-no attributes does the salience model use?
Power (can the stakeholder impose its will on the project?), legitimacy (is its claim proper, by law, contract or accepted norms?) and urgency (does the claim demand attention now?). A stakeholder with all three is definitive; the model names seven classes in all.
From the lesson Identifying and analysing stakeholders
-
How is stakeholder engagement planned?
As the gap between how engaged each stakeholder is now and the level the work needs them to be, on a five-level scale (unaware, resistant, neutral, supportive, leading). The register records what people said and need, not opinions about them, and is open only to the project's leads.
From the lesson Identifying and analysing stakeholders
-
How does India's DPDP Act affect a stakeholder register?
A register of named people's views is digital personal data under the DPDP Act, 2023, so purpose limitation, data minimisation and security safeguards apply: collect only what the engagement needs and keep it where only the project's leads can read it.
From the lesson Identifying and analysing stakeholders
-
What do R, A, C and I stand for in a RACI chart?
R responsible (does the work; may be several), A accountable (owns the outcome and answers for it; exactly one per row), C consulted (asked beforehand, two-way), I informed (told afterwards, one-way). When one person is both, the cell reads A/R.
From the lesson RACI charts, decision rights and escalation
-
Which four faults appear in most first RACI drafts?
Two A's in a row (nobody knows who decides), no R in a row (nobody prepares or does the work), empty rows (owners forgotten), and one person overloaded with the A's (every decision queues behind them).
From the lesson RACI charts, decision rights and escalation
-
What are decision rights with thresholds, and why do they matter?
Written limits on the impact each level may decide alone, by cost and schedule slip. They stop small decisions from queueing behind big ones. Every decision, whoever takes it, goes into a decision log with its options and record location.
From the lesson RACI charts, decision rights and escalation
-
Where does RACI apply in a Scrum team, and where does it not?
Only to decisions that cross the team's boundary, such as a release, a contract or a compliance check. The 2020 Scrum Guide calls the Scrum Team self-managing, deciding internally who does what, when and how, so a chart that assigns Sprint tasks to a manager contradicts it.
From the lesson RACI charts, decision rights and escalation
-
What three things does a scope statement list?
The deliverables, how each one will be accepted, and what the project will not do (exclusions). Writing exclusions down turns a repeated request into a visible decision rather than a favour.
From the lesson Scope statements and work breakdown structures
-
What is the 100 % rule of a work breakdown structure?
The children of every element hold all of that element's work and nothing more, so the hours at each level add up exactly. Work not in the WBS is not part of the project; two faults break it, missing work and work counted twice.
From the lesson Scope statements and work breakdown structures
-
Why are WBS elements named as deliverables rather than activities?
GAO's schedule guide: the WBS says what is required and the schedule says how it will be produced. Noun names (coding contest, crowd plan) stop two branches from describing the same activity in different words; the project-management work also gets its own branch.
From the lesson Scope statements and work breakdown structures
-
What are a work package and its WBS dictionary entry?
A work package is a lowest WBS element, sized to be estimated and owned (GAO suggests about 4-6 weeks; many teams use 8-80 hours). Its dictionary entry says what it contains, how it is accepted, who owns it, the estimate, dependencies and what it does NOT include.
From the lesson Scope statements and work breakdown structures
-
When do you use a WBS, when a Product Backlog, and when both?
A WBS fixes promised scope early and protects it with change control; a Product Backlog is an emergent, ordered list of what is needed to improve the product. Many projects keep features in a backlog under one branch of a WBS that holds the fixed deliverables.
From the lesson Scope statements and work breakdown structures
-
What are the parts of change control as the 2026 PMI outline lists them?
Log the request, work out the impact (scope, days, money, quality, risks), decide at the right level, communicate the status, carry out the change and update the baselines and documents to match.
From the lesson Change control without bureaucracy
-
How do scope creep and gold plating differ, and what cures both?
Scope creep is scope that grows informally, one small stakeholder request at a time; gold plating is the team adding extras nobody asked for. In both, no one with the right to decide chose the change. The cure for both is to log and decide every change, however small.
From the lesson Change control without bureaucracy
-
Why do change-control thresholds matter, and how do they shift?
They let the project lead settle small changes at once, within a cost limit and the schedule's remaining margin, while the change control board handles larger changes. Thresholds tighten as margin is used up, so what was small last month can be board-level today.
From the lesson Change control without bureaucracy
-
How does change arrive at a Scrum team in the middle of a Sprint?
It goes to the Product Backlog, not into the Sprint. The 2020 Scrum Guide says no changes are made during a Sprint that would endanger the Sprint Goal; the Product Owner orders the backlog, and if a change makes the goal pointless the Product Owner may cancel the Sprint.
From the lesson Change control without bureaucracy
-
What four things does a schedule activity need?
A name with a verb (because it describes work being done), a duration in working days (time it takes, not effort), one owner who reports on it, and a test for done anyone can check so "90 % finished" cannot last for weeks.
From the lesson Activities, dependencies and network diagrams
-
What do the four dependency link types model?
Finish-to-start (most common): successor cannot start until predecessor finishes. Start-to-start and finish-to-finish: model overlap, often with a lag. Start-to-finish: rare and theoretical, almost always rewritable as finish-to-start. GAO expects most links to be finish-to-start.
From the lesson Activities, dependencies and network diagrams
-
What is the difference between a lag, a lead and a milestone?
A lag is a wait between linked activities (publicity starts 2 days after the form starts), and a lead is a negative lag, an overlap. A milestone is an event with no duration that marks a decision or a finished deliverable, with little reason to be linked by anything but finish-to-start. Work must never be hidden inside a lag.
From the lesson Activities, dependencies and network diagrams
-
Which four network faults does a checker look for that dates alone would hide?
Loops (a cycle so no order satisfies the links), open ends (no successor, so slip never reaches the finish), a lag standing in for a link (dates match today but hide the real waiter), and dangling logic (FF predecessor needs an FS or SS; SS successor needs an FS or FF).
From the lesson Activities, dependencies and network diagrams
-
What does the forward pass compute, and what rule applies at a merge?
Each activity's early start (ES) and early finish (EF = ES + duration), walking from the start to the finish. At a merge of several predecessors, the successor's ES is the LATEST predecessor's EF, the step most hand calculations get wrong.
From the lesson The critical path method, step by step
-
What does the backward pass give, and how are total and free float defined?
Each activity's late finish (LF) and late start (LS = LF - duration), walking back from the finish. Total float = LS − ES: slip before the finish moves. Free float: slip before any successor's early start moves. Free float is never larger than total float.
From the lesson The critical path method, step by step
-
What is the critical path, and does it stay fixed?
The longest continuous chain of activities and the one that sets the earliest finish; its activities have the lowest total float (often zero). GAO: "critical" means on this path. It is not fixed: a near-critical chain slipping by more than its float becomes the critical path.
From the lesson The critical path method, step by step
-
What makes a critical path valid (GAO 6th best practice)?
It runs continuously from the status date to the finish milestone, carries no lags or leads, includes no level-of-effort or summary activities, and is not made critical by constraints on unimportant work. Negative float means the plan already misses a constraint date.
From the lesson The critical path method, step by step
-
What is a Gantt chart, and what does it hide?
A view of the schedule: each activity drawn as a bar on a calendar, from the dates the network and calendar give. It shows slips and crowded weeks but hides the logic, so read it with the links or a network check and never fix a date by dragging a bar.
From the lesson Gantt charts, calendars and schedule baselines
-
What is a schedule baseline, and what is statusing?
A baseline is a saved copy of the agreed schedule, kept unchanged for comparison (GAO's 10th best practice). Statusing is the update at each status date with actual starts and finishes and remaining durations, then recalculating through the links. Re-baseline only for new/removed scope or a formal replan.
From the lesson Gantt charts, calendars and schedule baselines
-
What are GAO's four characteristics of a reliable schedule?
Comprehensive (all work, with resources and durations), well constructed (simple complete logic; real critical path and float), credible (horizontally and vertically traceable; backed by a risk analysis) and controlled (updated with actual progress and compared with a baseline). Ten best practices group under these four.
From the lesson Gantt charts, calendars and schedule baselines
-
Why does the project calendar matter so much on an Indian team?
The Central Government, states and the RBI all publish different holiday lists, and India Standard Time has no daylight saving (IANA Asia/Kolkata since 1945). Teams spread across states need one calendar per place or a project calendar everyone agrees to; partners abroad move clocks twice a year.
From the lesson Gantt charts, calendars and schedule baselines
-
What does a resource histogram show, and what does levelling do?
It shows how many activities each person has on each day; a 2 is one person booked twice, a date that cannot be kept. Resource levelling delays or reassigns activities to remove the double booking. When every delay fits inside float and moves no other date, it is called resource smoothing.
From the lesson Resource levelling, crashing and fast-tracking
-
What is a cost slope, and how does crashing use it?
Cost slope = (crash cost − normal cost) / (normal duration − crash duration): the cost per day saved by shortening an activity. Crashing shortens the critical path one day at a time at the cheapest slope, recomputing paths after every day because a cheaper path can become critical.
From the lesson Resource levelling, crashing and fast-tracking
-
What does fast-tracking spend compared with crashing?
Risk, instead of money. Fast-tracking changes finish-to-start logic to start-to-start, overlapping work; it saves days without crashing cost, but work that starts before its predecessor finishes may have to be redone. Crashing spends money.
From the lesson Resource levelling, crashing and fast-tracking
-
Why does adding people to a late project often make it later?
Fred Brooks (The Mythical Man-Month, 1975): new people need time to learn the work and someone who knows it has to teach them, and the count of pairs that might have to talk rises as n(n−1)/2 (3→3, 6→15, 12→66). Only work that splits well, such as labelling cables, crashes well.
From the lesson Resource levelling, crashing and fast-tracking
-
What three numbers make a three-point estimate, and why skew to the right?
Optimistic (O), most likely (M) and pessimistic (P), gathered from the people who know the work (anonymously where that helps). Most schedule ranges skew right because things can go much more wrong than right.
From the lesson Three-point estimates and schedule risk analysis
-
What does PERT compute for an activity?
Expected time t_e = (O + 4M + P) / 6, standard deviation σ = (P − O) / 6. Along an independent critical path, means add and variances add (not standard deviations), so σ for the path is the square root of the summed variances.
From the lesson Three-point estimates and schedule risk analysis
-
What do P50 and P80 mean, and how is contingency sized?
P50 and P80 are the percentile dates of the Monte Carlo finish distribution: the dates by which 50 % and 80 % of simulated runs had finished. Schedule contingency is the gap between the chosen percentile (often P80) and the plan's finish, held as one visible block rather than hidden in each estimate.
From the lesson Three-point estimates and schedule risk analysis
-
What is merge bias in a schedule?
Where several paths meet, the event waits for the latest of them, so the probability of being on time multiplies. Two paths each 60 % likely to be on time are both on time only 36 % of the time. Parallel chains of similar length make a milestone later than any one of them.
From the lesson Three-point estimates and schedule risk analysis
-
How should a schedule risk analysis be communicated?
As a range with a confidence ("80 % likely to be over by working day 43"), not a single date. Record the seed and number of runs so it can be reproduced, and re-run as actual durations arrive so the finish range narrows.
From the lesson Three-point estimates and schedule risk analysis
-
Which three estimating methods does GAO's cost guide name?
Analogous (adjust a similar past project's cost), parametric (a statistical cost-estimating relationship fitted to past projects), and bottom-up / engineering build-up (price every work package). Early on only the top-down methods work; later the bottom-up becomes the budget.
From the lesson Cost estimates, budgets and reserves
-
Why is a bottom-up estimate often the lowest, and how is it checked?
Anything missing from its package list costs nothing in it. GAO's best practices ask for major cost elements to be cross-checked with another method; a gap between methods is a first finding, not an accident.
From the lesson Cost estimates, budgets and reserves
-
What is a time-phased budget, and what is the cost baseline?
Each package's cost placed in the months its work is planned in. The running total is the cost baseline, drawn as a cumulative curve. It is what earned value is measured against; the sponsor's project budget sits one layer above it.
From the lesson Cost estimates, budgets and reserves
-
What are the contingency and management reserves, and where do they sit?
Contingency reserve covers identified risks, sized from the register (sum of probability × cost), sits INSIDE the cost baseline, released by the project manager when a named risk occurs. Management reserve covers unknown problems, sits OUTSIDE the baseline, released by the sponsor or steering group.
From the lesson Cost estimates, budgets and reserves
-
GAO's four characteristics of a reliable estimate?
Comprehensive (the whole life cycle, nothing omitted or double-counted), well documented (sources and calculations repeatable by someone else), accurate (best method for each element, inflation-adjusted), credible (limits stated, major assumptions varied, a risk analysis).
From the lesson Cost estimates, budgets and reserves
-
What three numbers does earned value take at a status date?
Planned value (PV, budget of work scheduled), earned value (EV, budget of work actually done), actual cost (AC, what that work cost). All three are in money, so cost and schedule can be judged apart. BAC is the total budget.
From the lesson Earned value management on a worked project
-
What are CV, SV, CPI and SPI, and which values are unfavourable?
CV = EV − AC; SV = EV − PV; CPI = EV / AC; SPI = EV / PV. Negative variances and indices below 1.0 are unfavourable. CPI 0.887 means each 1.00 spent bought 0.89 of planned work; SPI 0.893 means 89.3 % of the planned work is done.
From the lesson Earned value management on a worked project
-
Why does an EVM index travel across projects while a variance does not?
A variance of −1,280 is serious in a 24,000 budget and trivial in a 2,400,000 one. CPI 0.887 means the same in both. A finished account always has SV 0 whenever it finished, so cost and schedule must be read together.
From the lesson Earned value management on a worked project
-
On an EVM S-curve, what do the two gaps mean?
PV is the cumulative baseline, with EV and AC drawn against it up to the status date. The vertical gap between AC and EV is the cost problem; the gap between PV and EV is the schedule problem, measured in money.
From the lesson Earned value management on a worked project
-
Why is a forecast for a sponsor a range rather than a single number?
One number invites false precision. The warehouse's earned value forecasts ran from 25,280 if the overrun stopped today to 28,962 if catching up on the delay costs extra, about 27,100 at the current efficiency. A forecast a sponsor can act on states such a range, how confident the team is in it and what would move it.
-
What is the planning fallacy, and what corrects it?
The finding that people underestimate their own completion times even when they know similar tasks overran before. Reference-class forecasting corrects it by budgeting from the actual outcomes of similar past projects instead of from the plan's own details.
-
What five recovery options exist, and what does each give up?
Descoping gives up scope (with the client's agreement), crashing gives up money, fast-tracking gives up safety from rework, releasing management reserve gives up cover for unnamed problems, and re-baselining gives up the old targets. Releasing reserve funds an overrun but moves no date.
-
What is a watermelon status, and when does it appear?
A status report that is green outside and red inside. It appears when the report's colours have no written thresholds, so the report hides the gap it should show honestly.
-
What is AgileEVM, and why does a Scrum release need it?
AgileEVM applies the same three earned value numbers to a release measured in story points. Scrum itself defines no way to track cost, so story points alone cannot show whether a release will run out of money before its work is done.
-
How does ISO 31000 define risk, and what two kinds does a project have?
Risk is the "effect of uncertainty on objectives"; an effect is any deviation from what was expected, good or bad. A project has threats (would harm an objective) and opportunities (would help one). Both are uncertain — they may or may not happen.
From the lesson Identifying risks and keeping a risk register
-
What four columns does a RAID log keep apart?
Risks (uncertain events that would affect objectives), Assumptions (believed true for planning, not yet verified), Issues (problems happening now) and Dependencies (work or decisions needed from others). Entries can move: an uncertain assumption becomes a risk; a risk that happens becomes an issue.
From the lesson Identifying risks and keeping a risk register
-
What pattern does a good risk statement follow?
It follows one pattern: "Because [cause], [event] may happen, which would [effect]." The cause is a fact true today, the event is the uncertain thing and the effect is what it would do to an objective, so the cause points at responses, the event at what to watch and the effect at the size of the impact.
From the lesson Identifying risks and keeping a risk register
-
Why use a risk breakdown structure instead of "what could go wrong?"
An open question finds only risks people already worry about. A risk breakdown structure sorts risk sources into categories, each with prompts, so the team walks it branch by branch. Opportunities need the same treatment as threats, or nobody acts on them.
From the lesson Identifying risks and keeping a risk register
-
What three things stop a risk register from rotting?
An owner who watches it and acts, a trigger (written in advance, such as "contract unsigned by week 8") and a next review date. A risk that can no longer happen is closed; a risk that has happened is closed in the register and carried on in the issue log.
From the lesson Identifying risks and keeping a risk register
-
Why must a probability-impact matrix have written scales?
Without them, "high impact" means two weeks to one person and two months to another, so scores become opinions. Cox showed matrices can give identical ratings to very different risks and even rate smaller risks higher.
-
What is expected monetary value, and when does it mislead?
EMV = probability × impact, negative for threats and positive for opportunities. Summed over a register, it sizes a reserve for the whole set. It misleads for a single risk, which costs either nothing or its full impact, never its EMV, so a rare very large threat needs its own response.
-
Which tools go beyond one EMV per risk?
A decision tree lays out choices and chance outcomes in sequence and picks the option with the best expected value. A sensitivity analysis (tornado chart) shows which uncertain input moves the result most. A Monte Carlo simulation combines many uncertain inputs into a distribution of outcomes.
-
Why does a probability-impact matrix need written scales BEFORE anyone scores?
Without them, one person's "high" and another's differ, so scores become opinions. On the hospital app's 5 × 5 matrix, two risks banded by feel in the workshop moved once the scales were written down, the shared Android developer from 12 to 8 and the portal-login opportunity from 12 to 9; one cell can hide a tenfold gap in money.
-
What five threat responses does the PMBOK Guide name, and on what does each act?
Avoid (changes the plan so the risk cannot happen), mitigate (lowers probability or impact), transfer (moves who pays if it strikes), escalate (hands the decision to its proper owner outside the project), accept (does nothing until a trigger fires, actively with a plan or passively).
-
What five opportunity responses mirror the threat responses?
Exploiting makes an opportunity certain, enhancing raises its probability or its benefit, and sharing gives part of it to a partner better placed to capture it. Escalating and accepting work as they do for threats.
-
What are a residual risk and a secondary risk?
Residual risk is what remains after the response; a secondary risk is a new one the response itself creates. Every response leaves something behind.
-
What is the difference between a contingency plan and a fallback plan?
A contingency plan runs when a risk's trigger fires; a fallback plan runs if the contingency plan fails.
-
What do an issue log and a risk burn-down chart each track?
An issue log drives problems that have already happened to resolution. A risk burn-down chart shows the total exposure of the open risks falling over time.
-
What do the Agile Manifesto's four values actually claim?
Each value puts two useful things side by side and says which to prefer when they conflict (people and interactions, working software, customer collaboration, responding to change). The follow-up sentence makes this explicit: items on the right have value, items on the left have more. So "over" is a tie-break, not a ban.
From the lesson The Agile Manifesto and its twelve principles
-
What four themes group the Agile Manifesto's twelve principles?
Customer value (1, 2, 3, 7), people (4, 5, 6, 11), ability to keep going (8 sustainable pace, 9 technical excellence, 10 simplicity), and the loop that improves everything (12 reflection and tuning). Principle 3 prefers the shorter release timescale rather than naming one length.
From the lesson The Agile Manifesto and its twelve principles
-
What does shortening the release rhythm do, with identical work?
One invented 48-week project with 12 features, released every 2 weeks instead of every 12, turned 4 releases into 24, brought first customer use forward from week 12 to week 2, cut mean wait of finished work from 6.8 weeks to 0.6, and gained 74 extra weeks of customer use.
From the lesson The Agile Manifesto and its twelve principles
-
Which three Manifesto misreadings cost teams the most, and why are they wrong?
"No documentation" (text keeps documents with a named reader), "no planning" (principle 2 welcomes change to a plan, so there must be one), "no contracts" (collaboration is preferred, the agreement stays). All three come from reading the left-hand item as the only item.
From the lesson The Agile Manifesto and its twelve principles
-
Does the Agile Manifesto describe a method?
No. It names no roles, meetings, boards or estimates. Scrum, Extreme Programming, Kanban and the rest are methods built on the Manifesto's values. Two teams can both follow it and look nothing alike.
From the lesson The Agile Manifesto and its twelve principles
-
Where did the Agile Manifesto come from, and what was it written about?
Seventeen people agreed it in 2001, after meetings of people who each already had a method that worked, among them Extreme Programming, Scrum, Crystal, DSDM, Feature-Driven Development, Adaptive Software Development and Pragmatic Programming. They wrote about software, so applying it elsewhere takes a translation the text does not provide.
From the lesson The Agile Manifesto and its twelve principles
-
What two pillars does Toyota rest its production system on?
Jidoka (a machine or person stops the moment something is wrong, so faults are not built into later work) and Just-in-Time (make only what is needed, when it is needed, in the quantity needed). Toyota also names three kinds of waste together: muda (waste), mura (unevenness), muri (overburden).
From the lesson Lean flow: waste, Little's law and constraints
-
What does a value-stream map record on each step?
Every step from the request to the delivery, with the hours of real touch time and the hours of waiting before each step. It measures the item, not the people, so the useful question is where the time went, not how busy anyone was.
From the lesson Lean flow: waste, Little's law and constraints
-
What is flow efficiency, and what does a low value say?
Touch time divided by elapsed time: 44 hours of work inside 220 hours is 20 %. A low figure means the time is in queues, handoffs and partly done work — the cheapest improvements usually sit there. It is a direction finder, not a target.
From the lesson Lean flow: waste, Little's law and constraints
-
Which four tools act on a value-stream map, and what does each do?
Little's law ties WIP, throughput and cycle time together; the theory of constraints names the slowest step (the bottleneck) and says improving any other step adds nothing; batch size decides how long finished work waits; cost of delay prices one week of lateness so the queue can be ordered by money.
From the lesson Lean flow: waste, Little's law and constraints
-
What are Extreme Programming's five values, and what do they ask of a lead?
Communication (work requirements face to face, customer in the room weekly), simplicity (ask for what is needed now; treat extra scope as a cost), feedback (short iterations, running software, honest measures), respect (let people doing the work size it), courage (tell the truth about estimates and progress).
-
What eight XP practices hold the values up?
Small releases, planning with stories (a card, a conversation, a confirmation), test-first work, pair programming (two people, one story, one screen), continuous integration, refactoring in small safe steps, collective ownership under agreed standards, and sustainable pace.
-
Why does a project lead care about XP practices, not just the developers?
Frequent delivery without an automated check moves risk to users; Done without a written definition moves work into the next quarter; one owner per module turns one person's leave into a slip. The practices decide whether a reported date can be trusted.
-
What does Extreme Programming say that Scrum and Kanban do not?
Scrum and Kanban say how work is chosen and pulled; XP says how it is built, meaning how a requirement is agreed, how code is written and checked and how often it reaches a user.
-
What did quarterly releases cost the bus-pass app team before XP practices?
Each release took about fourteen days of checking by hand, users reported forty-one defects over the year and three releases had to be taken back out. The plan was never the problem; the way the work was built was.
-
What two foundations does the 2020 Scrum Guide name for Scrum?
Empiricism (knowledge comes from experience and decisions are made from what has been observed) and lean thinking (reduce waste and focus on essentials). Together they trade a detailed plan for a short build-inspect-adapt loop.
From the lesson Scrum theory, values and accountabilities
-
What are Scrum's three pillars, and how does each one fail?
Transparency (fails when "done" is false), inspection (fails when the check becomes a status readout nobody acts on) and adaptation (fails when the team is not empowered to change the next slice). The Guide: "Inspection without adaptation is considered pointless."
From the lesson Scrum theory, values and accountabilities
-
What are the five Scrum values?
Commitment, Focus, Openness, Respect and Courage. They are distinct behaviours, not five words for "try hard". The Guide says decisions should reinforce these values, not diminish them, so the test is behavioural — not a poster on the wall.
From the lesson Scrum theory, values and accountabilities
-
What did the 2020 Scrum Guide change about team self-direction?
"Self-organising" became "self-managing": the team also chooses what to work on, not only who does it and how. The Scrum Team internally decides who does what, when and how, and the Developers are accountable for the plan for each Sprint.
From the lesson Scrum theory, values and accountabilities
-
Which common Scrum words are NOT in the Scrum Guide?
Velocity, story points, user stories, "Development Team" (one unit now), "roles" (now accountabilities), "self-organizing" (now self-managing), "servant-leader" (now "true leaders who serve"), and the three Daily Scrum questions. These are techniques teams add, not framework.
From the lesson Scrum theory, values and accountabilities
-
What are the five Scrum events, and what contains the other four?
The Sprint is a fixed-length container for the other four: Sprint Planning (goal, items, plan), the Daily Scrum (adapt the plan toward the goal), the Sprint Review (what to build next) and the Sprint Retrospective (how to work better). Cancelling a Sprint is for the Product Owner alone.
From the lesson The Sprint and the four Scrum events
-
Are Scrum timeboxes targets or maxima?
Maxima. For a one-month Sprint, Sprint Planning has 8 hours, the Review 4 and the Retrospective 3, and shorter Sprints usually have shorter events. The Daily Scrum is 15 minutes at any Sprint length, and an event that reaches its outcome early ends early.
From the lesson The Sprint and the four Scrum events
-
What outcome must each Scrum event end with?
Sprint Planning: a Sprint Goal, items and a plan (not task assignment). Daily Scrum: a plan for the next day (not a status round). Sprint Review: decisions and backlog adjustments (not just a presentation). Sprint Retrospective: a change with an owner (not a complaints round).
From the lesson The Sprint and the four Scrum events
-
Is Product Backlog refinement a Scrum event?
No. The Product Backlog is refined as needed, with no timebox, while the Sprint contains the four events of Sprint Planning, the Daily Scrum, the Sprint Review and the Sprint Retrospective.
From the lesson The Sprint and the four Scrum events
-
What are Scrum's three artifacts and their commitments?
The Product Backlog, whose commitment is the Product Goal (the long-term product objective); the Sprint Backlog, whose commitment is the Sprint Goal (one objective for the Sprint); and the Increment, whose commitment is the Definition of Done (the quality bar work must clear to count).
-
Why does each Scrum artifact carry a commitment?
The 2020 Scrum Guide says a commitment is there "to ensure it provides information that enhances transparency". It is what everyone checks the artifact against, so the team inspects its work from the same picture; without it an artifact becomes a passive record instead of an active guide.
-
What is a Sprint Goal, and what is it not?
The single objective a Sprint serves, written so that it can be shown met at the Sprint Review. Of three teams' drafts for live seat counts, only one passed; the other two were a list of build tasks and a feature description that never said why the Sprint matters.
-
What are Scrum's artifacts, in one sentence?
The three containers the framework defines for holding work and its plans, the Product Backlog, the Sprint Backlog and the Increment, each built to make its key information transparent so everyone who inspects it can adapt from the same starting point.
-
Who orders the Product Backlog, and may it be a committee?
The Product Owner is accountable for ordering the Product Backlog, and the Scrum Guide is blunt that this is "one person, not a committee". Stakeholders change the order by convincing the Product Owner.
-
What does each of four backlog-ordering methods give you?
MoSCoW gives a promise of what is guaranteed in a fixed timeframe. WSJF gives a sequence, best economic outcome first. The Kano model shows the kind of satisfaction an item buys. A story map shows the users' journey and what a release leaves out.
-
What does WSJF compute, and what rises in the order?
WSJF, from the Scaled Agile knowledge base, divides each item's relative cost of delay by its job size, so small work that is urgent or removes a risk rises in the order. It cannot see guarantees already made.
-
What does a story map lay out, and what can it not show?
It lays the users' activities and steps out left to right and slices a release across them, so it shows the journey and what a release leaves out. It cannot show the value of any one step.
-
What does Sprint Planning subtract from the naive capacity?
Three public holidays, one developer's leave, one developer on support duty and the hours the Scrum events take, then a reserve for interruptions. The Chennai team's sheet starts from the naive 480 person-hours (six people, ten working days, eight hours) and what is left is the real number.
-
What should a Sprint forecast look like, and where does it come from?
A range, read from how many items the team finished in each of its last few Sprints (its recent throughput), showing what it can select with high confidence, with medium confidence and at a stretch. One average hides the spread.
-
What two questions does Sprint Planning answer before anyone commits?
How much work these people can really finish, which needs an honest team capacity, and which outcome is worth committing to, which needs a forecast and a Sprint Goal.
-
Why did six people and ten working days not give the Chennai team sixty person-days?
A harvest-festival holiday cluster took three days, one person went to the support rota and the Scrum events took a slice of every day. Planned on the naive figure, the team carried a third of its Sprint Backlog into the next Sprint.
-
What is the Scrum Master accountable for?
For establishing Scrum as the Guide defines it, and for the Scrum Team's effectiveness. The role holds no power to hire, assign work or appraise anyone; the Guide calls Scrum Masters "true leaders who serve", serving the Scrum Team and the organisation around it.
-
Which three parties does the Scrum Master serve, and with what?
The Developers (coach self-management, keep events productive, cause impediments to be removed), the Product Owner (help with the Product Goal and backlog techniques, facilitate stakeholders) and the organisation (lead Scrum adoption, remove barriers between stakeholders and teams).
-
Which four Scrum Master anti-patterns quietly undo the role?
Running the Daily Scrum as a status report, owning the board instead of the team, acting as the Developers' project manager, and stretching one person over too many teams.
-
Where does a Scrum Master send each impediment?
To the person who can actually remove it, which may be the team itself; the rest are escalated with evidence and options. The role holds no power to hire, assign work or appraise anyone, so it leads by making obstacles visible.
-
How does the Kanban Guide v2025.5 define Kanban, and what three practices carry out the strategy?
"A strategy for optimizing the flow of value through a process", balancing effective, efficient and predictable flow. The three practices are: define and visualise the workflow, actively manage items in the workflow, and improve the workflow — used together.
From the lesson Kanban: workflow, WIP and the three practices
-
What six elements make up a Definition of Workflow?
Work items (what counts as one), started and finished points, the states between them, how WIP is controlled (when to stop starting), explicit policies for moves, and a service level expectation (SLE) — a period with a probability of meeting it, from past cycle times.
From the lesson Kanban: workflow, WIP and the three practices
-
When is a To Do / Doing / Done board NOT a Kanban system?
Without WIP control, explicit policies, flow metrics, work item meaning and a probability-based SLE, three columns are only a board. The Guide calls a Kanban board the visualisation of a Definition of Workflow — without that, there is nothing for the columns to show.
From the lesson Kanban: workflow, WIP and the three practices
-
What makes a Kanban board a pull system, not a push system?
An explicit WIP control, with the policy that a new item starts only when the count is below it. The Guide: start work "only when there is a clear signal that there is capacity to do so." In push, work starts because it arrived or a plan said so.
From the lesson Kanban: workflow, WIP and the three practices
-
How do the Kanban Guide and Kanban University's Kanban Method differ?
The Guide is minimal: three practices, six DoW elements, four flow metrics. The Method is a management method laid over existing work, with change-management principles ("start with what you do now") and SIX general practices: visualise; limit WIP; manage flow; make the policies explicit; implement feedback loops; improve collaboratively.
From the lesson Kanban: workflow, WIP and the three practices
-
What are the four flow metrics a Kanban system must collect?
WIP (items started but not finished, counted at a moment), throughput (items finished per unit of time — exact count, not estimates), work item age (how long an open item has been in progress) and cycle time (elapsed time from start to finish, finished items only).
From the lesson Flow metrics: WIP, throughput, age, cycle time
-
How do work item age and cycle time differ, and why does age matter?
Both measure elapsed time from the start of an item. Cycle time belongs to finished items and describes the past (median 3 days, 85th percentile 15 here). Age belongs to open items and describes the present — the oldest open ticket here was 41 days. Age is the number a team can still act on today.
From the lesson Flow metrics: WIP, throughput, age, cycle time
-
Why is throughput an exact item count, not a total of estimates?
The Kanban Guide calls throughput "the exact count of work items" per period. Anyone can recount items on the board, while a total of points or hours cannot be checked at all, so say "the team finished 17 items this week", not "the team delivered 46 points".
From the lesson Flow metrics: WIP, throughput, age, cycle time
-
What does a cycle-time scatterplot show that a median hides?
Every finished item as a dot at the day it finished and the days it took, so the whole spread is visible. Here, dots stayed in a band of 1-9 days for seven weeks and climbed afterwards; the median of 3 days hid 20 items that took more than 15.
From the lesson Flow metrics: WIP, throughput, age, cycle time
-
How does a WIP limit improve cycle time without adding capacity?
Little's law: for a stable workflow, WIP = throughput × cycle time. Fixing throughput by team size means the only way to shorten cycle time is to carry fewer items. A WIP limit turns a push queue into a pull system; the waiting moves to a visible queue the team can reorder.
-
What did a WIP limit of 8 do to the helpdesk's 90 days?
Throughput held at 11 items a week, the 85th-percentile cycle time fell from 15 days to 7, and 8 tickets were still in progress at the end, not 30. Nobody worked faster; the team simply stopped starting work it had no capacity to finish.
-
What does Little's law say about a stable workflow?
Average work in progress equals average throughput times average cycle time. When throughput is fixed by the size of the team, the only way to shorten cycle time is to carry fewer items at once — so a WIP limit shortens cycle time.
-
On which three occasions is a team's WIP limit really tested?
On the day the limit is reached, when an item is blocked and when a request is called urgent. Improving the workflow then means testing each change of limit against the flow metrics.
-
Where did the waiting go when the helpdesk replayed its 90 days under a WIP limit?
Into the queue. Counted from arrival, the 85th percentile did not fall and 20 tickets never started; they waited where they can still be ordered, split or declined.
-
What five planning horizons do agile teams estimate at, and what unit fits each?
Portfolio (one to two years, T-shirt sizes or similar past projects, board), roadmap (quarters, T-shirts in Sprints, PO + a few Developers), release (weeks to months, story points or counts), Sprint (two weeks, the Developers' own forecast), day (tasks of a day or less). Each horizon asks its own question.
-
Why do T-shirt sizes need a written scale, and how are they added?
Without a scale, one person's M is another's L and sizes can't be compared or added. Convert each size to its range, then add the ranges: S+M+S on a 2-3, 3-5, 2-3 scale makes 7 to 11 Sprints. A letter has no arithmetic; only the range behind it does.
-
When does planning poker beat affinity estimation, and when the reverse?
Planning poker sizes one item at a time (about a minute each), so 40 items take most of an hour. Affinity estimation puts all cards on one wall in silence (15 minutes for 40) and spends discussion only on the cards that keep moving. Use affinity for the first large pass, poker for disputed and new items.
-
When can counting items replace story-point estimates?
Once slicing has made items similar in size. The marketplace's six Sprints showed each finished item carried 2.40 to 3.20 points, with a correlation of 0.88 between points and item counts. A narrow band says counts will do; a wide one says slice the biggest items first.
-
When should you NOT estimate an item, and what to do instead?
When nobody can size it (slice it or research first — Grenning's "infinity card"), when it's a whole new initiative (use reference-class forecasting from similar past projects), and for tasks inside a selected item (if no decision depends on a number, don't produce one).
-
What does a burn-up chart show that a burn-down hides?
Scope change. A burn-down plots only work remaining, so added scope looks like lost progress (the marketplace's line rose at Sprint 3 although the team finished 12 points). A burn-up draws done work and total scope as TWO lines; the gap between them is the work left.
-
What two questions does a release plan answer, and in what form?
With scope fixed, when will it be done? With a date fixed, how much will be done by then? Both answers are ranges, not single dates — first release plans in Extreme Programming are "necessarily imprecise" and are revised regularly. The 2020 Scrum Guide defines no release plan.
-
Which three methods replace the single-mean release forecast?
The slowest and fastest past Sprints give the earliest and latest finish. Monte Carlo sampling builds thousands of futures from past Sprints picked at random, repeats allowed, and counts how often the release is done by each Sprint. A confidence level states the result in a form a stakeholder can weigh, such as 85 % likely by a given Sprint.
-
Who owns the release forecast in Scrum, and why is it not a promise?
The 2020 Scrum Guide holds the Product Owner accountable for managing the Product Backlog and its order; it defines no release plan. The plan is best kept as the Product Owner's forecast, redone after every Sprint, because only what has already happened can inform the next decision (empiricism).
-
What are Tuckman's five team stages, and what triggers a slip back?
Forming, storming, norming, performing and adjourning (Tuckman 1965; Tuckman & Jensen 1977). They describe behaviour, not a timetable. A new member, reorganisation or outage can send a performing team back to storming for a week; read the signals rather than count the weeks.
From the lesson Team development, motivation and trust
-
What three needs grow lasting intrinsic motivation (Ryan & Deci 2000)?
Autonomy (the person decides something that matters within clear limits), competence (the work is hard enough to need skill and clear enough to show progress) and relatedness (the person belongs to the team). A bribe or threat works only while in view and leaves all three unmet.
From the lesson Team development, motivation and trust
-
What is team psychological safety, and what is it NOT?
Amy Edmondson's 1999 definition: a shared belief that an interpersonal risk (asking, admitting an error, disagreeing) will not be punished. It is not comfort, not agreement and not low standards. Safe teams can still have hard conversations and ship on time.
From the lesson Team development, motivation and trust
-
How does a working agreement differ from a team charter?
A working agreement is the team's one-page record of HOW it works (core hours, meeting slots, decision rule, how a Review ends). A team charter wraps that in a wider purpose, customer and the decisions that sit outside the team. Neither replaces the Definition of Done, which is about the product.
From the lesson Team development, motivation and trust
-
What three practices keep a remote and hybrid team from quietly breaking?
Rotate the unsociable meeting slot across the sites (make the pain shared), write decisions down for the hours nobody is awake, and keep the work visible on a shared board. Fairness is a rotation, a written rule and a visible board, not a kind word in the Review.
From the lesson Team development, motivation and trust
-
What four leadership styles does situational leadership name, and what chooses between them?
Direct (lead decides and gives steps), coach (lead decides and explains why, asks for ideas), support (person decides, lead encourages and clears obstacles), delegate (person decides within agreed limits). Choose from the follower's READINESS for THIS TASK: how able and how willing they are.
From the lesson Leadership styles and influence without authority
-
Does readiness belong to a grade or to a task?
To a task. An expert at deployments can be a beginner at estimates, so the lead delegates the deployment plan she has run many times and directs her first cost estimate. Leading more tightly than readiness needs costs time: nine over-tight decisions lost 3.0 of 10.5 decision-days in the log.
From the lesson Leadership styles and influence without authority
-
What does Robert Greenleaf's servant leadership serve, and what doesn't it mean?
The growth of the people served. Greenleaf's test begins "Do those served grow as persons?" and asks whether, served, they grow freer and better able to serve others themselves. Serving is not obeying: the lead carries bad news upward and clears obstacles, yet holds the team to what it agreed.
From the lesson Leadership styles and influence without authority
-
What is influence without authority, and how does it win cooperation?
Help from people who do not report to you, such as a security team or a busy sponsor. It wins that cooperation through the other team's priorities, evidence and choices instead of pressure.
From the lesson Leadership styles and influence without authority
-
What five conflict modes does the Thomas-Kilmann MODE instrument describe?
Competing (press for your concern), collaborating (dig until one answer meets both), compromising (each gives up part for a quick middle), avoiding (step back for now) and accommodating (set your concern aside). Three questions choose one: stakes, relationship, time.
-
Which conflict mode fitted the tester and product owner's release dispute, and why?
Compromising. The change touches payments, so the outcome matters; the two will work together for years; and there were hours, not days. Collaborating needed more time and competing would cost trust, so the release ships tonight with the payment change switched off and its checks finish tomorrow.
-
What is principled negotiation, and how does it follow a conflict mode?
Settling a dispute on interests and fair standards instead of fixed demands. A conflict mode is a stance; negotiation is the conversation that follows it, and when more than two parties share the problem, facilitation replaces a string of one-to-one talks.
-
What is never negotiated or compromised?
Harassment and discrimination. Report them through your organisation's own route and use none of the conflict, negotiation or facilitation techniques on them.
-
Along which two dimensions do Thomas and Kilmann describe behaviour in a conflict?
Assertiveness, how hard you pursue your own concerns, and cooperativeness, how hard you try to meet the other person's. Their five conflict modes are built from those two.
-
How does the number of communication channels grow with group size?
n(n − 1) / 2 pairs. 10 people have 45 channels, 20 have 190, 22 have 231. A newcomer adds as many channels as there were people before, so the 23rd person adds 22. A plan does not add conversations; it chooses the few that must happen.
-
What seven columns does a communication plan carry?
Audience, information, purpose, frequency, channel, owner and feedback route. The plan gives each audience its information, the reason for sending it, a rhythm, a channel, one owner and a way to answer.
-
Where does an honest RAG (red-amber-green) colour come from?
From numeric limits written before the first report. At the tech fest the unconfirmed venue booking sat in a chat thread the dean was never in, under a green status with no rule behind it, and in week 7 the dean still had not heard.
-
How did four teams and a lead circle cut the tech fest's 231 channels?
Teams of 6, 6, 5 and 5 have 15, 15, 10 and 10 channels inside them, and the four leads add 6 more, so 56 channels carry the work instead of 231. That is 175 fewer, a fall of 76 %.
-
What two halves make the engagement-and-expectations loop?
Engaging stakeholders (deciding, for every person or group affected, the level needed and the actions to move them there) and managing expectations (deciding in writing what the work will produce, kept aligned as scope and dates change). Each name carries current and desired engagement; the gap gets an action with an owner, date and sign of success.
-
What is an expectation gap, and what catches it?
It opens when the sponsor's criteria and the team's running work drift apart. A Sprint Review held as a working session catches it.
-
What replaces a flat refusal to a late request?
Three answers, each priced in scope, days and cost. They turn the refusal into a choice and send the decision to the right level.
-
What is knowledge transfer, and how is it checked?
Finding what only one person can do, capturing it and checking it with a test the receiver can pass. PMI's 2026 PMP outline lists Help ensure knowledge transfer among its People tasks.
-
Why would two of the e-bus depot's six stakeholders never move?
Their planned actions had no owner, no date or no sign of success, so nobody was due to do anything for them. An action that closes an engagement gap needs all three, and a gap of 2 or more levels needs sustained engagement rather than one meeting.
-
What does a quality management plan give each important requirement?
A standard, a metric, a target, acceptance criteria that turn all three into a pass or fail for one release, a check method, and an owner who keeps the check running after the project has moved on.
From the lesson Quality planning, assurance and control
-
How do quality assurance and quality control differ?
Quality control inspects a product against its criteria and accepts or rejects it (defects in one output). Quality assurance looks at the PROCESS that makes outputs and improves it (causes of defects across every output). A project needs both; testers alone cannot do assurance.
From the lesson Quality planning, assurance and control
-
What four groups make up the cost of quality (BS 6143-2 prevention-appraisal-failure)?
Prevention (stops defects being made: training, criteria), appraisal (finds defects: testing, audits), internal failure (defects found before release: fix, re-test), external failure (defects citizens find: helpdesk, emergency fixes). A failure-share figure (here 69.4 %) is the case for spending more on prevention.
From the lesson Quality planning, assurance and control
-
What does a Pareto chart show, and how do you read it?
Bars sorted from largest category to smallest, with a cumulative line. Look for where the line bends: in the portal three of ten categories held 158 of 200 reports (79 %). When categories cost very different amounts, draw a second Pareto by cost, not count.
From the lesson Quality planning, assurance and control
-
How do control limits differ from specification limits?
Control limits come from the PROCESS (NIST individuals chart: 3 standard deviations from the mean, estimated from the average moving range divided by 1.128). Specification limits come from the CUSTOMER or the requirement. A process can be in control and still break its specification.
From the lesson Quality planning, assurance and control
-
When do you use a cause-and-effect diagram vs. a PDSA cycle?
A fishbone finds the cause of a SIGNAL (a point outside control limits — a special cause). PDSA (plan-do-study-act, AHRQ) is for routine variation of the process: short cycles with the expected result written in Plan so Study has something to compare. Larger projects use DMAIC (ISO 13053-1).
From the lesson Quality planning, assurance and control
-
What time horizon does make or buy need?
The life of the purchase, not the first invoice. The school's build costs 762 thousand more at start and 120 thousand less each year; the two options only draw level after 6.3 years, while the devices are rated for five and the first contract runs for three. Buying costs 402 thousand less across three years here.
-
What questions do an RFI, RFQ and RFP each ask the market?
Request for information (RFI) asks what is possible; request for quotation (RFQ) asks the price of something already specified; request for proposal (RFP) asks bidders how they would solve a stated problem. Which to send depends on what the buyer still does not know.
-
Who carries the cost risk under each of the three contract families?
Fixed price: the seller (overrun comes out of its profit, savings stay with it). Time and materials: the buyer, up to the cap it agreed. Cost-reimbursable: the buyer carries almost all of it, paying allowable costs plus a fee. The choice follows how well the work can be specified today.
-
Which five contract types from Part 16 of the U.S. Federal Acquisition Regulation does the lesson name?
Firm fixed price; fixed price with an incentive fee, a target cost and profit with savings and overruns shared up to a ceiling price; time and materials with a not-to-exceed cap; cost plus a fixed fee; and cost plus an incentive fee or an award fee. Agile work is bought by fixing the time and cost of each increment and leaving the scope inside it variable.
-
Why must evaluation criteria and weights be fixed BEFORE any bid is opened?
Otherwise the yardstick moves to suit the bids. Vendors are chosen against criteria fixed in advance, scoring quality and cost together or taking the cheapest technically qualified bid. Afterwards a vendor scorecard, service levels and service credits keep performance visible.
-
What five pieces of work make up project closure?
Acceptance against written criteria, transition readiness (trained support, working runbooks, reconciled data), administrative closure (end every contract, account and login), a final retrospective with owners for the changes, and a benefits register that keeps measuring benefits months after go-live.
-
What evidence shows transition readiness, rather than a promise?
Evidence the support team can see, such as trained support staff, runbooks that work, alerts that reach someone and migrated data that reconciles. Every closure piece needs evidence like this, not a feeling that it is done.
-
What does a benefits register do, and who owns it after closure?
It gives each benefit an owner, a baseline, a target and the months when it is measured, such as readings at 3, 6 and 12 months from go-live. The business owns realisation, not the project team, because benefits happen in daily use after the handover; each owner reports readings to the steering group.
-
What must a final retrospective produce to be worth holding?
Changes, each with an owner, to the checklists, templates and estimating data that the next project will use. It turns what happened on this project into practice the next one starts from.
-
Does a project stopped early go through closure?
Yes. A project stopped before its planned end is closed by the same five pieces of work, with the reason for stopping recorded.
-
How much of the school ERP budget went back to the sponsor at closure?
209 thousand of the 3,864 budget. The rollout spent 3,655 of its 3,680 cost baseline and never released its 184 of management reserve, and the unspent money goes back to the sponsor at closure.
-
How does project governance differ from project management?
Governance is the frame: who may decide what, on what evidence, how often, where a decision goes when it is too big. Management is the work inside those limits: planning, assigning, tracking and correcting. A quick test: who could stop the project? Always a governance body, never the lead.
From the lesson Project governance, PMOs and compliance
-
What are tolerances and management by exception?
A tolerance is the deviation from plan a level may absorb or decide without going up (such as 200 thousand and 5 days for the lead, 1,000 thousand and 15 days for the sponsor). Management by exception: a level reports up only when a forecast breaks its tolerance, and then at once, with an exception report.
From the lesson Project governance, PMOs and compliance
-
What four endings must a real stage gate allow?
Go, go with conditions, hold or stop. A gate that can only say go is a review. The committee decides on a pack of evidence at the end of each stage whether the project has earned the next stage; that protects the money the next stage would spend.
From the lesson Project governance, PMOs and compliance
-
What three kinds of PMO does the PMBOK Guide name, and how do they differ?
Supportive (templates, training and advice; requires little), controlling (also requires projects to follow its methods, templates and gate reviews), and directive (manages the projects itself; project managers are its staff). Choose by what the organisation needs and who should employ the PMs.
From the lesson Project governance, PMOs and compliance
-
How should a project treat a compliance requirement?
Like scope. Each row of a compliance register has one owner, named evidence, a due date and acceptance criteria. On an agile team it becomes a backlog item whose criteria name the evidence, so it is planned, built and finished like any other item, not a launch-week scramble.
From the lesson Project governance, PMOs and compliance
-
What three roles does India's DPDP Act name, and what is the heaviest penalty?
Data fiduciary (decides why and how personal data is processed), data principal (the person the data is about; for a child, parent/guardian too), data processor (processes data on the fiduciary's behalf, under a valid contract). Heaviest penalty: up to ₹250 crore for failing reasonable security safeguards.
From the lesson Project governance, PMOs and compliance
-
What does each level — project, programme, portfolio — answer, and how is it judged?
Project: How do we deliver this output? Judged by scope, time, cost, quality. Programme: Which related work brings this benefit? Judged by benefits realised. Portfolio: Which work deserves our teams? Judged by value for the capacity used. PMI keeps PgMP and PfMP credentials for the two upper levels.
From the lesson Programmes, portfolios and OKRs
-
Why is a portfolio a capacity choice, not just a priority list?
Starting beyond capacity adds waiting, not work done. Six teams on six initiatives finished sooner than on ten and gave up 522 thousand rupees less benefit, with the first initiative live in month 1.0 instead of 1.7. Each initiative needs the same team-months either way.
From the lesson Programmes, portfolios and OKRs
-
What are OKRs, and what grade range is expected?
Objectives and Key Results, as Google's re:Work guide describes them: ambitious objectives, each with about three key results, each graded on a scale of 0.0 to 1.0, with an average of 0.6 to 0.7 expected. OKRs are public, and they are neither employee evaluations nor a shared to-do list.
From the lesson Programmes, portfolios and OKRs
-
What are the four Key Value Areas of Scrum.org's Evidence-Based Management?
Current value, unrealised value, time to market and ability to innovate. Evidence-Based Management measures the value an organisation delivers in these four areas and moves towards its goals in small, measured steps.
From the lesson Programmes, portfolios and OKRs
-
What three problems do scaling frameworks exist to solve?
Dependencies, integration and alignment. On the exam platform 14 of 18 items needed two to four teams (23 hand-offs), only 4 items were done within one Sprint, and the Fees team received 8 of its 10 items from another team, so others' plans set its priorities.
-
What distinguishes Nexus, LeSS, SAFe and Scrum@Scale?
Nexus, from Scrum.org, is for roughly three to nine Scrum Teams on one product and adds a Nexus Integration Team. LeSS runs many teams as one Scrum, with one Product Owner, one Product Backlog and one Sprint. SAFe groups teams into Agile Release Trains that meet for PI Planning, usually two days every eight to twelve weeks. In Scrum@Scale, teams join into Scrums of Scrums, supported by two leadership forums.
-
What can a group of teams do before adopting a scaling framework?
Reduce dependencies first, by regrouping the teams as feature teams that can finish an item alone, so less needs coordinating. The Spotify model, a 2012 description of one company's squads and tribes, is not a framework to install.
-
How does a component team differ from a feature team?
A component team owns one part of the system, as the platform's Forms, Fees, Identity, Centres and Results teams do, so most items need several teams and hand-offs between them. A feature team can finish an item alone, so less needs coordinating.
-
Why did renaming everything to Scrum barely move the insurance department's lead time?
The names changed and the decisions did not, because the same people still decided every change on the same monthly cycle. Events rose from 0 to 520 a month, releases stayed at 2 a quarter and the median lead time moved only from 94 days to 91.
-
What do Kotter's accelerators and Prosci's ADKAR describe?
Kotter's firm describes eight accelerators of organisational change, from creating a sense of urgency to instituting change; his 1995 article on why transformations fail drew on more than 100 companies. Prosci's ADKAR model treats change one person at a time, through five elements, namely awareness, desire, knowledge, ability and reinforcement.
-
What are cargo-cult Scrum and water-scrum-fall?
In cargo-cult Scrum the events are held while decisions stay where they were. In water-scrum-fall, up-front analysis and a late large release wrap the Sprints. Both stall adoptions; so do Sprints without a usable Increment, a backlog owned by a committee, and metrics turned into targets.
-
What is evolutionary change from the Kanban Method?
The Kanban Method starts with what you do now and pursues small agreed steps, each kept or dropped on evidence. Evidence-Based Management then judges the adoption by outcomes in four key value areas, not by the events held.
-
What are a guiding coalition and a short-term win?
A guiding coalition is a group whose members, between them, control what the change needs, and who lead it as a group. Short-term wins are improvements delivered early and made visible, so people see the change working.
-
What three categories of project tool exist, and when does each fit?
Work-tracking tools (items, states, owners; boards and issue trackers), schedulers (activities, dependencies, dates; Gantt charts), spreadsheets (anything the team defines; small one-off lists). A detailed critical-path schedule stays in a scheduler; a Kanban board with WIP limits stays on a board.
-
What does it mean to configure a board from the Definition of Workflow?
Each DoW element becomes something on the board: states → columns in flow order, WIP control → a number above the columns it covers, pull policy → a written rule or card order, start and finish points → where cycle time's clock starts and stops. A board that ignores the DoW is decoration.
-
How should free tiers and open-source tools be compared?
Against weighted criteria that include data-export format and access-control granularity, with a must-have rule for anything a tool must support. A tool that scores well but exports only as a screenshot is a lock-in risk; limits change, so re-check vendor pricing pages.
-
What four practices keep project data safe when choosing tools?
Access control (each person sees only what they need; no shared logins), data export (CSV opens anywhere — test before committing), offboarding (disable, remove, rotate credentials, confirm — written checklist), and vendor terms (data ownership, location, what happens if the vendor shuts down). All matter under India's DPDP Act.
-
What are three good uses of an AI assistant on a project, and one bad one?
Good uses: a first draft of a short artefact (charter paragraph, risk list, status note), a summary of a long thread, and a prompt list that opens a conversation (retro questions, test scenarios). Not for: taking a decision. The project manager still signs the charter and the risk owner still owns the response.
-
What are the four stages of the verify-then-decide loop for work with an AI assistant?
The prompt stage decides what leaves the room. The draft stage reads the whole output. The verify stage opens the source for every fact, and a claim without a source goes out. The decide stage records who owned the decision. The stages are not optional.
-
What three kinds of error slip past a quick read of an AI draft?
A number close enough to look right (eighty people instead of eighteen); a plausible name (a colleague who left two years ago); a fabricated citation (a document that does not exist or a 404 URL). Only reading each claim against the register, schedule, RACI or document it should have come from catches all three.
-
Does accountability move to the AI when a draft helped produce an artefact?
No. Human accountability stays with a named person: the project manager signs the charter, the risk owner owns the response, the Scrum Team agrees the Sprint Goal. The assistant's draft is one input to the decision, not the decision itself.
-
What two things does a metric need to earn its place on a dashboard?
One question for one named reader, and a decision the reader will take differently depending on what the number says. Without that pair, a number only occupies screen area; with it, the number is auditable.
-
What does Goodhart's law warn about, in a project context?
Once a yardstick becomes the thing people are rewarded for, it stops measuring what it was chosen to measure. Charles Goodhart made the point about British monetary policy in 1975, and Marilyn Strathern about audit in universities in 1997.
-
Which reader and decision does each of three warehouse metrics serve?
85th-percentile cycle time helps the delivery team decide which open items to escalate. Cost performance index helps the sponsor decide whether to release part of the reserve. Deployments a week help the engineering lead decide whether the release pipeline needs change.
-
How does a cycle-time number change once it is tied to performance ratings?
People move the start timestamp forward, split big items into pieces that each finish fast while the user gets the result no sooner, label slow work as blocked and quietly drop the oldest open items. The number improves and the system does not.
-
What four kinds of work do all those project titles cluster into?
Coordinating a project someone else leads (coordinator, PMO analyst), leading one project end to end (project manager), running related projects together (programme manager) or a steady flow (delivery manager), and the two Scrum team-level accountabilities (Product Owner, Scrum Master).
From the lesson Project careers: coordinator to lead
-
How does PMI count project leadership experience?
In MONTHS, not projects, over the past 10 years. Overlapping months count once (two parallel projects for 6 months counts as 6, not 12). Only professional work counts — school projects, personal events and routine office tasks do not. A dated log is the honest input.
From the lesson Project careers: coordinator to lead
-
What five artefacts go into a project portfolio to show judgement?
The charter, the delivery plan, one page of the risk register, two status reports (the one that first called a slip, and the one that proposed a trade-off), and the retrospective summary. Together they show choices a prose resume cannot. Remove names, amounts and anything personal under India's DPDP Act.
From the lesson Project careers: coordinator to lead
-
What red-flag phrases should you look for in a project posting?
Phrases that give one person authority over a team supposed to self-manage ("manages the team", "assigns work to developers"), or responsibility over an outcome without the money or decisions to shape it ("owns delivery without a budget"). Ask the hiring manager which of a long requirements list matters most.
From the lesson Project careers: coordinator to lead
-
What does each of the six compared credentials ask for before the exam?
PMP (PMI, 36 months leadership + 35 taught hours), CAPM (PMI, 23 taught hours), PMI-ACP (PMI, 2 years agile + 21 agile hours), PSM I (Scrum.org, no requirement, course recommended), CSM (Scrum Alliance, 16-hour course), PRINCE2 Foundation (PeopleCert, no requirement).
From the lesson PMP, CAPM, PMI-ACP, PSM, CSM, PRINCE2 compared
-
What are the PMP's entry rule, exam format and renewal cycle?
With a degree, 36 months leading projects (60 with secondary school only) and 35 hours of training. The exam has 180 questions in 240 minutes, the fee is paid once the application is accepted, and the credential is renewed with 60 PDUs every 3 years.
From the lesson PMP, CAPM, PMI-ACP, PSM, CSM, PRINCE2 compared
-
How should you choose which project credential (if any) to pursue?
Start from your own record: months of leading projects, years of agile work, ability to sit a course. Those three facts rule out half the table. Then read ten postings for the role you want; if none name a credential, hours and fees may build a better record of work (a led project, a written retrospective).
From the lesson PMP, CAPM, PMI-ACP, PSM, CSM, PRINCE2 compared
-
Why always check the current fee on the owner's own page?
Owners revise fees and formats, and PMI's credential pages say the exam fee is paid once an application is accepted, so read the current figure before you book. A payment card issued in India is billed in rupees at its bank's rate on the day, and the bank may add charges.
From the lesson PMP, CAPM, PMI-ACP, PSM, CSM, PRINCE2 compared
-
What five rounds do PM and Scrum Master interviews draw on, usually in order?
Screening (your last project in two minutes), scenario questions (a situation and "what next"), framework check (definitions against the 2020 Scrum Guide or PMP outline), a walk-through of papers from a project you ran, and behavioural questions about what you did on a named occasion. Confirm the order with the recruiter.
-
What does a framework-check round compare your answers against?
Their sources. For Scrum that is the 2020 Scrum Guide; for predictive and hybrid work it is the domains of the published examination content outline for the PMP. Every Scrum claim you make should be one the interviewer can check in a public source.
-
How do project manager and Scrum Master interviews weigh the same rounds differently?
A project manager is pressed harder on plans, money and stakeholders. Scrum Master interviews push on the 2020 Scrum Guide and on how a team is coached rather than directed.
-
What does a screening round usually cover?
Twenty to thirty minutes on your last project in two minutes, the team size, your own part in it and the tools you worked in.
-
Why rehearse without lists of questions reported from named employers?
Such lists are unreliable, they go stale, and they steer you towards reciting an answer instead of thinking in front of a stranger. Rehearse the shape of each round instead, with Scrum claims you can check in a public source yourself.