A senior executive requests one small field, but the change touches three interfaces just before user acceptance testing. This scenario lab examines why the affected commitments matter more than the requester’s confidence that a change is free. Follow the request through an appr…
CURRENT EDITION: PMP 2026. Start with Season 2 for the updated certification preparation course. Previous seasons are archived and may not reflect current exam requirements.
The Certified PMP® Audio Course is your complete, on-the-go study companion for mastering the Project Management Professional (PMP®) certification exam. Designed for busy professionals, this Audio Course transforms every aspect of the Project Management Institute’s (PMI) exam content into structured, easy-to-follow lessons that blend clarity with real-world application. Each episode guides you through the principles of project leadership, process integration, stakeholder management, risk, quality, and agile methodologies—ensuring you understand not only the “what” but also the “why” behind each concept. Whether you’re commuting, exercising, or reviewing between meetings, this series helps you study smarter, stay motivated, and move steadily toward exam success.
The PMP® certification is the global gold standard for project managers, validating your ability to lead teams, deliver results, and manage complex projects under real-world constraints. It measures your proficiency across predictive, agile, and hybrid approaches—emphasizing leadership, strategic thinking, and adaptability. Earning the PMP® demonstrates mastery of the three key domains: People, Process, and Business Environment, and signals to employers that you can manage risk, align projects with organizational goals, and deliver value from start to finish.
Developed by BareMetalCyber.com, the Certified PMP® Audio Course provides structured instruction, practical insights, and exam-focused guidance to help you build the confidence, mindset, and discipline needed to pass the PMP® exam and excel as a certified project management professional.
Change is inevitable in projects, but unmanaged change erodes trust and confuses commitments. For an ordinary request affecting an approved baseline, capture the request, assess its effects and obtain the required decision before committing to changed work. Communicate the authorization and synchronize affected records in the sequence the applicable process requires. Routine backlog ordering and emergency action do not require an identical sequence or a change control board in every organization. The project manager protects coherent commitments while using the right path for the situation.
Welcome to Certified, the P M P prepcast. This is episode seventy six of season one. Today we are exploring project scenarios involving change control and baseline integrity.
The artifacts that protect baseline integrity are straightforward but critical. The change log captures every request and its status, serving as the authoritative record. Baselines for scope, schedule, and cost define the current plan of record and are only updated when formally approved. Backlog policies support reprioritization within the product and team authority already established. The affected commitment determines the route: a low-cost change may require another approval, while a substantial reorder may remain delegated. A change control board is one possible arrangement where the organization uses one. These artifacts prevent drift and disputes by ensuring everyone is looking at the same version of the truth.
Scenario one introduces a familiar pressure: a very senior executive requests a “small” field to be added to the system. They insist it is trivial, yet it touches three different interfaces. They also claim the change will have zero impact. The constraints are tight: user acceptance testing is five days away, and a vendor contract is in play. On the surface, this feels like something to just slip in, but P M I expects the project manager to resist this temptation. The presence of multiple interfaces, contracts, and imminent U A T means that even a small change could ripple significantly.
Compare the responses to the facts given. Verbal approval without the required analysis and authority would create an unsupported commitment, while immediate rejection would ignore possible value. Logging the request, assessing its effects and using the established decision route is stronger in this nonemergency case. The approval route may be delegated or may involve a change control board, depending on the affected commitments. These are original teaching options, not an official examination item or a rule that every request needs a committee.
For this nonemergency request, the important next step is an impact analysis before commitment. Examine the effects on scope, schedule, cost, quality, risk and contractual work. Show what additional testing is needed and which interfaces or approvals are affected. The authorized decision maker can then compare the options. If the request is approved, update the affected baselines and change log; if it is declined, record that decision too. A project manager may decide within delegated authority, while a request beyond that authority follows the established approval route.
The change record documents the request, analysis and decision status. The contract file helps the authorized owner identify whether supplier action requires a modification. Once a decision is made, communicate it and keep the affected planning records aligned before people rely on changed instructions. Record actual implementation and verification as they occur. This trail explains the decision and its execution, but a completed template is not proof that the decision was valid or the change works.
There are pitfalls to watch for in this scenario. Undocumented decisions, where changes happen but the log is not updated, create disputes later. Shadow changes, where team members implement requests informally, erode baselines. Baseline drift, where schedule, cost, and scope stop aligning, destroys trust in project data. The heuristic to remember is clear: policy first, personalities second. Even V I P requests must flow through the system. Respecting stakeholders means respecting governance, not bypassing it. P M I rewards project managers who resist pressure and preserve integrity.
In an adaptive setting, first identify who can order work and which commitments constrain the choice. In Scrum, the Product Owner is accountable for Product Backlog ordering, while Developers plan their work and discuss changes to selected work with the Product Owner. The Sprint Goal and quality must remain protected. A request within those boundaries may be handled through normal backlog decisions. A request affecting an external contract, funding authorization or other controlled commitment follows the relevant governance path as well. The deciding factor is the commitment affected, not whether someone calls the request small.
Let us extend the executive's request into a fictional equipment-rental project. The requested field records a customer's preferred delivery entrance. A screen change looks simple, but the same value passes through the reservation service, the dispatch interface and a supplier's handheld application. The approved release contains only the main street address. Before promising the new field, ask what business problem the executive is trying to solve. The answer is repeated delivery attempts at large buildings, not a desire to add one more box to a screen.
That answer creates more than one possible response. One option adds the new structured field throughout the chain. Another improves the existing delivery-instructions process for a limited pilot. A third defers the change until the next release while gathering failed-delivery evidence. The team should compare those alternatives against the actual need. Repeating the executive's preferred solution as the requirement would hide the opportunity to achieve a useful result with less disruption. An alternative is only credible if dispatchers can actually use it and the relevant owners accept its limitations.
The integration lead estimates two working days for the reservation change, three for the dispatch interface and two for the handheld change. Those three activities can run in parallel because separate qualified people are available. End-to-end testing then requires two days, followed by one day of acceptance review. On that simplified dependency model, the earliest complete path is six working days, not ten. The controlling branch is the three-day interface activity, followed by two days of testing and one of review. Adding all three parallel estimates would overstate the duration; ignoring the testing and review would understate it.
There are only five working days before the planned acceptance event. Under the stated assumptions, the full change misses that event by one day even without an adverse surprise. The estimate should also identify confidence and resource assumptions rather than disguising them as a promise. If the same specialist must perform both interface and handheld work, those activities are no longer parallel and the calculation changes. The project manager therefore verifies the dependency and availability assumptions with the people doing the work before presenting the number as a decision input.
The cost comparison needs the same care. Assume the three development tasks total seven person-days at eight hundred dollars per person-day. Testing requires two additional person-days at six hundred dollars, and acceptance review one at seven hundred dollars. The modeled incremental labor cost is seven thousand five hundred dollars. The supplier also quotes two thousand five hundred dollars for its authorized interface work, making ten thousand dollars in total. These are fictional estimates, with the supplier work defined as separate from the internal labor to avoid double counting. A six-day calendar duration and ten person-days of internal effort are compatible because some people work concurrently.
Now apply the local authority rule stipulated for this case. The project manager can approve a change costing up to twelve thousand dollars only when it does not alter an external milestone or a supplier agreement. The request falls under the monetary limit but fails the other conditions. It therefore requires the appropriate milestone and contract decisions. Treating the budget threshold as a universal permission would be a mistake. Delegation often contains several conditions, and satisfying one does not erase the rest.
The decision brief should separate the known need, estimated impact and decision requested. It might explain that delivery failures are observed, that the proposed complete change is estimated to require six working days, and that the options include moving acceptance or choosing a permitted narrower response. It should identify the benefit evidence still missing, such as how often the entrance information would have prevented a failed delivery. The sponsor can then assess value rather than being asked to approve a vague technical task. The contract owner separately determines how any supplier commitment can be changed.
Suppose the authorized decision is to defer the structured field and approve the limited instructions pilot. The project team must make that exact scope visible. A rejected implementation option is not the same as a rejected business need. Record which alternative was chosen, what the pilot is permitted to do, what evidence it will collect and when the product owner will reconsider the full field. Otherwise, the same request may return a week later as an allegedly forgotten commitment. Traceability protects the learning as well as the baseline.
Suppose instead that the full field is approved with a new acceptance date. Approval must reach the implementers before they treat the change as authorized work. Synchronize the relevant schedule, scope and supplier instructions in the sequence the organization's process requires, rather than deliberately leaving the old plan in circulation until implementation is finished. The controlled record should distinguish the approval date, planned effective date and actual implementation date. Those dates can differ. Recording all three accurately is more useful than pretending that a single signature instantly changed the system.
Configuration identification makes the decision testable. The acceptance team needs to know which release candidate contains the field and which interface specification it expects. A test report from an earlier candidate does not automatically prove the changed candidate works. Conversely, a targeted change does not automatically invalidate every unrelated test. The technical and quality owners assess the affected evidence and select the necessary verification. The project manager ensures that the plan funds and schedules that work without inventing technical acceptance authority.
Finally, watch for a shadow change after the decision. An engineer may add the field privately to be helpful, leaving it hidden behind a configuration switch. The immediate response is to establish what was changed, where it exists and whether any users or interfaces are exposed. Preserve the record and involve the responsible technical owner in a safe disposition. Do not automatically order a blind rollback, because removal can itself cause harm. The authorized outcome may be removal, isolation or a properly assessed change, but it must not become accepted merely because someone already spent effort on it.
Concurrent requests expose another weakness in isolated impact analysis. Suppose a second approved enhancement also needs the same specialist to verify the dispatch interface. Each change requires four hours of that specialist's testing time, but only six hours remain before the acceptance cutoff. Each looks feasible when examined alone; together they require eight hours and exceed available capacity by two. No single estimate was necessarily dishonest. The integration failure was treating the same capacity as available twice. Review the combined workload and ask the authorized owners to resolve the sequencing or milestone effect.
The technical interaction may matter more than the hours. One change may add the entrance field while another changes the structure of the dispatch message. Testing each on a separate development branch does not establish that the combined release works. Identify the intended combined configuration and the evidence needed for that configuration. Where there is a genuine conflict, present feasible sequencing alternatives rather than asking the tester to certify an untested combination. The choice might be to release one change first, extend verification or defer both, depending on their value and constraints.
Approval can also expire or become conditional when assumptions change. An authorization based on an available supplier slot does not automatically apply after that slot disappears. Check the decision's actual conditions and the project's delegation before assuming that a materially different implementation is covered. Notify the decision owner of the changed fact, assess its consequences and obtain any required new decision. Preserve the original approval rather than editing its rationale to fit the new circumstances. This lets a later reviewer distinguish a legitimate adaptation from work that quietly outgrew its permission.
Now consider a severe production defect with a release planned for tomorrow. In this fictional organization, an emergency change procedure delegates containment and recovery authority to named responders. The project manager activates that route and ensures the right technical and operating owners are involved. If immediate protection is needed, authorized containment proceeds while the team captures evidence and communicates. Testing, approval, rollback or another safe recovery method then follow the applicable emergency procedure. Urgency changes the route and timing; it does not create unlimited authority.
Patching without considering authority, safety or evidence is a weak response. Using the established emergency path is stronger: triage the defect, take authorized protective action, assess the proposed repair, obtain any required approval and preserve a recovery option. Record the decision and validate the outcome. Waiting for an unrelated sponsor meeting is unnecessary when the required authority has already been delegated. If that authority is missing or the incident meets a reporting threshold, prompt escalation belongs in the response. The appropriate choice follows the facts and the actual procedure.
The emergency policy identifies authority, and the recovery plan describes how the team will respond if the repair fails. The decision record shows who acted, when and on what evidence. After stabilization, the team completes required records and reviews the result. These records support a review, but documentation alone does not establish that the system is safe or that every obligation was met. Technical validation and the applicable approval conditions still matter. Emergency governance needs both effective action and an honest account of it.
The emergency scenario has a different clock. Imagine a fictional reservation service that is confirming the same equipment unit for two customers. The incident commander has delegated authority to disable new reservations for the affected equipment category while existing bookings remain accessible. That action is containment within a specific boundary. It is not permission to rewrite pricing, delete disputed bookings or alter every service. Narrow authority can support fast action precisely because responders understand what they may do without waiting for an ordinary review meeting.
The project manager helps maintain a clear picture of the incident while the operating team leads its response. Record when duplicate confirmations began, which release and configuration were active, how many bookings are potentially affected and what is known versus still being investigated. Preserve relevant records through the approved handling process. An early count of affected bookings is provisional if the team has not finished reconciliation. Reporting it as a final count can mislead customer support and the sponsor even when the number was honestly estimated.
A proposed rollback introduces a critical decision branch. In the first branch, the team has verified that the earlier application version can read the current data and that returning to it will not lose confirmed transactions. Rollback may then be a feasible recovery option within the incident procedure. In the second branch, the latest release permanently changed the reservation records and the old version cannot interpret them correctly. The same rollback command could make recovery harder. The appropriate technical owner must assess containment, forward repair or another safe recovery method rather than assuming that reversal is always available.
Testing under urgency should be selected for the relevant risk, not eliminated because the ordinary schedule no longer fits. A targeted check might demonstrate that a booking is created only once, that cancellation still releases the correct equipment and that previously confirmed reservations remain readable. These checks are examples for the fictional system, not a universal test suite. The release decision must also consider monitoring, support coverage and the response if the defect returns. A small patch can have a large consequence when it touches shared transaction logic.
Once service stabilizes, compare the records with the actual sequence of events. An approval entered afterward must not be backdated to imply that it preceded an action. Where emergency policy authorized an earlier containment action, record that authority and the later review honestly. Where someone exceeded authority, preserve that fact and route it through the appropriate review. A tidy timeline is not more valuable than a truthful one. The follow-up can improve the procedure without rewriting what happened.
The learning question is narrower than asking who made the mistake. Did responders know their limits, could they find the current configuration, and did the recovery option match the data state? An improvement might add a compatibility check to the release preparation or clarify which role can disable the affected function. Assign an owner and a way to verify that the improvement works. A retrospective that merely says to communicate better leaves the same decision gap available for the next incident.
A third situation begins after an approved schedule change. The report still uses old dates, a supplier is working from an earlier notice, and the team is unsure whether the approval also authorized additional cost. Review the actual decision and its effects. A schedule change does not automatically require every baseline to change, and it does not itself authorize new spending. Correct records that fail to reflect the approved decision, then route any newly identified cost or contract impact through the required authority. Integration means consistent affected commitments, not moving every baseline together.
Ignoring the inconsistency would leave tomorrow's steering committee with unreliable information. Updating only the dates while concealing a newly discovered cost impact would also be incomplete. The useful response is to reconcile the approved change with the current records, assess unresolved effects, and communicate what is authorized and what still needs a decision. Involve the contract owner before changing supplier commitments. If an additional approval cannot be obtained before the meeting, report that limitation plainly instead of making the report appear fully aligned.
Integration requires consistency among the commitments actually affected. The plan, version history and supplier instructions should reflect the authorized decision without implying that every baseline must change. A current cost forecast can disclose a new impact even while additional funding approval is pending. Give the steering committee that distinction rather than making the report look aligned through unsupported adjustments. The useful record shows what was approved, what was implemented and what still requires a decision.
There is also a reporting trap after an approved change. In a fictional rollout, the original completion commitment was the end of week twelve. A properly authorized scope change moved that commitment to the end of week fourteen. The latest forecast now predicts completion in week fifteen because testing is running late. All three dates answer different questions: what was originally agreed, what is currently authorized and what the team now expects. Showing only the current baseline and calling week fifteen an approved date would conceal the new variance.
A forecast should reflect credible current evidence even before an additional change is approved. Updating that forecast is not the same as granting the team more time or money. The status report should say that the forecast is one week later than the current authorized commitment, explain the cause and identify the response or decision needed. Keep the original baseline history available so stakeholders can understand the effect of successive authorized changes. Neither freezing the forecast at an impossible date nor repeatedly resetting the baseline produces honest control.
A pending request also needs a separate identity. Suppose a supplier proposes a new component with a shorter lead time, but technical acceptance has not been completed. The planning team may model an alternative schedule using that component while keeping the authorized plan distinct. Label the alternative as conditional and record the assumptions that would have to become true. Do not send its dates to downstream teams as commitments. Scenario planning helps decision makers compare possibilities; it does not silently approve the scenario.
When records disagree, reconcile them against the actual decision and the intended configuration rather than declaring the newest timestamp authoritative. A recently edited spreadsheet may contain an unauthorized assumption, while an older signed decision may remain the controlling commitment. Ask who owns each record, which decision it represents and whether a later authorized change exists. Correct the discrepancy with a traceable explanation and notify the people relying on it. Baseline integrity is therefore not a contest to find the most recent file; it is the ability to connect current work to valid decisions.
A more reliable memory aid is to identify the decision and the authority before choosing the route. For an ordinary baseline change, impact analysis and approval normally precede implementation. For immediate danger, use the applicable protective authority without waiting for an unrelated administrative sequence. For ordinary backlog ordering, apply the product and team decision rights already established. In each case, keep the affected records accurate and communicate the outcome. The exam situation determines which step is needed now; a slogan cannot replace that judgment.
Backlog flexibility also has boundaries. Reordering work within an authorized product objective differs from changing a funding ceiling, a contract or a committed external date. A seemingly minor feature can cross an important privacy or interface boundary. Conversely, a substantial reorder may be within the Product Owner's existing authority. Identify the affected commitment and the appropriate owner. Keep the team plan, backlog and project records coherent without inventing a change control board requirement for every adaptive decision.
Baseline integrity lets stakeholders understand what was approved, what changed and why the current forecast differs. It supports adaptation because the team can discuss an honest starting point. For further study, use the change-control material in chapter thirty-eight of P M P for Busy People. As you leave this lab, practice identifying the affected commitment and the person who can authorize the next step. That question is more useful than assuming all changes need the same paperwork or that urgency permits an undocumented promise.
For more certification study resources, books, and podcasts, visit Bare Metal Cyber dot com. Thank you for listening to Certified, the P M P prepcast.