BCT(B) cheat sheet

Requirement types: business (why), stakeholder (who needs), solution (what), transition (how we get there).

A requirement is verifiable only if a test or inspection can objectively confirm it; avoid words like fast, easy, user-friendly.

Elicitation techniques map to context: interviews for depth, workshops for consensus, observation for tacit work, surveys for breadth, document analysis for legacy rules.

MoSCoW ranks scope: Must (non-negotiable), Should (important workaround exists), Could (nice to have), Won't (explicitly excluded this release).

Traceability links requirement to source, design element, test case, and change request; a requirement with no downstream test is unverified.

INVEST for user stories: Independent, Negotiable, Valuable, Estimable, Small, Testable.

Requirement quality checks: unambiguous, complete, consistent, feasible, necessary, traceable, and uniquely identified.

A business rule is a constraint on the business, not a feature; it persists even if the software is replaced.

Gap analysis compares current state capability against desired state and lists what must change to close the difference.

BPMN core elements: events (circles), activities (rounded rectangles), gateways (diamonds), sequence flows (solid arrows), message flows (dashed arrows across pools).

Exclusive gateway (XOR) takes exactly one path; parallel gateway (AND) takes all paths simultaneously; inclusive gateway (OR) takes one or more.

A pool represents an organization or participant; lanes within a pool partition roles or systems.

Little's Law: average items in system = arrival rate x average time in system (L = lambda x W).

Cycle time = total elapsed time per unit; throughput = units completed per period; capacity = maximum sustainable throughput.

Process mining uses event logs (case ID, activity, timestamp) to derive actual flow, bottlenecks, and rework loops.

Value-added analysis classifies steps as value-adding, business-value-adding, or non-value-adding (waste); only the first two survive ideal design.

Common waste categories in process work: waiting, overproduction, rework, handoffs, approvals, and defects.

To-be design starts from the customer outcome and removes steps that do not change the output or satisfy a control requirement.

Measure scales determine statistics: nominal (mode only), ordinal (median), interval (mean, differences), ratio (ratios and percentages).

Mean is sensitive to outliers; median is robust; report both plus count and spread when distributions are skewed.

Correlation coefficient ranges -1 to +1; correlation does not establish causation, only association.

Sampling error shrinks with the square root of sample size; quadrupling n halves the margin of error.

Data quality dimensions: completeness, accuracy, consistency, timeliness, uniqueness, validity.

A KPI needs an owner, a formula, a data source, a target, and a review cadence; without a formula it is not measurable.

Descriptive answers what happened, diagnostic why, predictive what will happen, prescriptive what to do.

Visualization rule: bar for comparison, line for trend over time, scatter for relationship between two measures, pie only for parts of one whole with few slices.

Dashboard design: one primary question per view, consistent scales, and no chart that implies a causal claim the data cannot support.

Option evaluation matrix scores each candidate against weighted criteria; weights sum to 1.0 and scores use a fixed scale.

Weighted score = sum over criteria of (weight x score); document the scoring rationale so results are auditable.

Build vs buy criteria: total cost of ownership, time to value, customization need, vendor viability, integration effort, lock-in risk.

Cost-benefit inputs include one-time implementation cost, recurring run cost, and quantified benefit; compute payback period = initial cost / annual net benefit.

Non-functional requirements (performance, availability, security, scalability, usability) constrain the design and must be measurable.

Design alternatives should be compared on fit to requirements first, then cost; a cheap option that misses a Must requirement is not viable.

A proof of concept validates technical feasibility; a pilot validates operational fit with real users on limited scope.

Risk register entry: description, likelihood, impact, exposure, owner, mitigation, and trigger.

Solution acceptance criteria are defined before build and mirror the requirement's verification method.

Power/interest grid: high power-high interest = manage closely; high power-low interest = keep satisfied; low power-high interest = keep informed; low power-low interest = monitor.

RACI assigns exactly one Accountable per activity; Responsible does the work; Consulted is two-way; Informed is one-way.

Workshop opening sequence: purpose, scope, ground rules, roles, agenda, then content.

Facilitator stays neutral on content and owns process; a separate scribe captures decisions and actions.

Parking lot records off-scope items with an owner so discussion stays on agenda without losing input.

Conflict modes: collaborate, compromise, accommodate, compete, avoid; collaborate suits high-stakes issues needing commitment.

Communication plan fields: audience, message, channel, frequency, sender, and feedback mechanism.

Closed questions confirm facts; open questions expand options; probing questions test assumptions.

Meeting minutes record decisions, actions (owner and due date), and open issues, not a transcript.

Triple constraint: scope, schedule, cost; a change to one affects the others, with quality and risk as balancing factors.

Critical path is the longest duration path through the network; it sets the minimum project duration and has zero float.

Float (slack) = late start - early start; activities on the critical path have zero float.

Earned value: CV = EV - AC; SV = EV - PV; CPI = EV / AC; SPI = EV / PV; CPI below 1 means over budget.

Estimate at completion (EAC) = BAC / CPI when current cost performance continues.

Change control steps: request, log, impact assessment (scope, schedule, cost, risk), decision, update baseline, communicate.

Kotter's eight steps begin with urgency and coalition and end with anchoring change in culture.

ADKAR sequence: Awareness, Desire, Knowledge, Ability, Reinforcement; a gap at any element blocks adoption.

Resistance to change is data about impact; address it with involvement, clear rationale, and visible sponsorship.

Core duties: competence, integrity, confidentiality, conflict-of-interest disclosure, and objectivity in recommendations.

Confidentiality persists after the engagement ends; share information only with authorized parties or as legally required.

Disclose any personal, financial, or family interest that could bias a recommendation before the decision is made.

Do not accept gifts or benefits that could reasonably be seen to influence professional judgment.

Present findings honestly: state assumptions, data limitations, and uncertainty; never suppress results that contradict a preferred option.

Respect intellectual property: cite sources, do not reuse licensed question banks or proprietary frameworks without permission.

Escalation rule: when instructed to act unethically, document the request, refuse, and raise it through the appropriate channel.

Only claim certifications, skills, and experience you hold; decline work outside your competence and refer it onward.

Personal data handling follows minimum necessary collection, purpose limitation, and access control.