Planning: work, schedule, responsibilities and risks

1Breaking down the work (WBS)

The project breakdown structure (WBS, work breakdown structure) starts from the deliverables and breaks them down into ever smaller work packages, down to tasks that can be estimated (duration, cost, person responsible).

Example: “20 water points rehabilitated” → technical studies; procurement of materials; works; management committees trained; water quality control.

A useful rule: what is not in the WBS is not in the project.

2Gantt chart and critical path

The Gantt chart places each task on a timeline, with its dependencies (task B can only start once A is finished).

The critical path is the longest chain of dependent tasks: it gives the minimum duration of the project. A task on the critical path has no slack: any delay to it delays the end of the project. Other tasks have some slack (“float”).

In the field, international procurement, administrative authorisations and the rainy season are often on the critical path: plan them first.

3Who does what: the RACI matrix

For each task or decision, the RACI matrix shows who is:

• R (Responsible): does the work;
• A (Accountable): answers for it and approves it; only one person per task;
• C (Consulted): gives an opinion beforehand;
• I (Informed): is informed afterwards.

It avoids the two classic pitfalls: “everyone thought someone else was dealing with it” and “too many bosses for the same decision”.

4Risk register

A risk is an uncertain event which, if it occurs, affects the project. It is recorded with its probability and its impact (for example from 1 to 5); probability × impact gives its priority.

There are four possible responses: avoid (change the plan), reduce (lower the probability or impact), transfer (insurance, contract) or accept (with a contingency plan). Each risk has an owner and a review date.

The project plan also includes the procurement plan (what, when, which procedure), the budget, the monitoring and evaluation plan and the communication plan.

5Building a usable WBS and estimating the work

The WBS from the core course breaks down the deliverables. For it to really serve as a management tool, it must be linked to the logframe, the budget and the monitoring plan, and each work package must be able to be estimated, assigned and verified.

Step-by-step method

1. Start from the logframe results: each result becomes a level 1 branch. Add two cross-cutting branches that are too often forgotten: project management (kick-off, reports, committees, audit, closure) and accountability and monitoring and evaluation (baseline and endline surveys, post-distribution monitoring, complaints mechanism).
2. Break down to the work package: a work package is a set of tasks entrusted to a single person responsible, with a verifiable deliverable. The so-called "8/80" rule of thumb recommends work packages requiring between 8 and 80 hours of work; in humanitarian settings people tend to think in work packages of one to four weeks, which fit between two monthly monitoring meetings.
3. Number each element (1, 1.1, 1.1.1). This code serves as a common key between the work plan, the procurement plan and, where possible, the budget's cost-accounting codes.
4. Write the WBS dictionary: for each work package, one line specifying the deliverable, the acceptance criteria, the person responsible, the resources, the estimated duration, the estimated cost and the dependencies.
5. Validate with the support services: logistics checks procurement and transport lead times, finance checks consistency with budget lines, HR checks the recruitment required.

WBS dictionary extract

Estimating durations and costs

• Analogous estimating: based on a comparable project. Quick, but must be adjusted for the context (prices, access, season).
• Parametric estimating: unit cost or duration multiplied by quantity. For example, a drilling team completes a 60-metre borehole in 4 to 6 days in sedimentary terrain; a team of 6 artisans builds 4 to 5 family latrines a week. Reference costs published by clusters are invaluable.
• Three-point estimating (PERT): for uncertain tasks, estimate an optimistic (O), most likely (M) and pessimistic (P) duration; the expected duration is (O + 4M + P) / 6. Example: customs clearance of imported pumps, O = 2 weeks, M = 4, P = 10; expected duration = (2 + 16 + 10) / 6 = 4.7 weeks. The gap between M and the expected duration measures the risk: it justifies a contingency.
• Estimating by those who do the work: technicians, logisticians and community mobilisers estimate better than the project manager. Ask them for a range, then consolidate.

A complete WBS includes management, monitoring, coordination and accountability time. In a 12-month project, management and monitoring commonly account for 15 to 25% of the team's total effort. Ignoring them amounts to planning a project that nobody manages.

6Calculating the critical path and building a realistic schedule

Identifying the critical path "by eye" works for six tasks. Beyond that, it has to be calculated. The method is simple and can be done in a spreadsheet; every project manager should be able to explain it in an interview.

The four-pass method

1. List the tasks, their durations and their predecessors (the most common relationship: finish-to-start, task B starts when A is finished).
2. Forward pass: for each task, the early start (ES) is the maximum of the early finishes of its predecessors; the early finish (EF) = ES + duration. The largest EF gives the project duration.
3. Backward pass: starting from the end, the late finish (LF) of a task is the minimum of the late starts of its successors; the late start (LS) = LF − duration.
4. Total float = LS − ES. Tasks with zero float form the critical path.

Worked example (weeks)

The project lasts 16 weeks. Two paths are critical: C → E → F → H and D → F → H. Practical consequence: the pump order must go out in the first week, even before the studies are finished, which requires a standard specification and agreement on the budget. The land agreements have only two weeks of float: a three-week delay would make them critical.

Typical critical tasks in the field

• International procurement: 8 to 16 weeks between expression of need and receipt, counting the tender, manufacturing, sea freight and customs clearance. Customs exemptions can add several weeks.
• Tenders: depending on the procedure, 2 to 8 weeks between publication and signature, more if the donor requires international advertising.
• Recruitment: 6 to 10 weeks for a qualified national position, 8 to 14 weeks for an international staff member, plus visas.
• Authorisations: memoranda of understanding with ministries, drilling permits, ethical approval for certain surveys.
• Seasons: rains (access, civil works), harvests (availability of labour and communities), holiday periods, school years.

Making the schedule robust

• Plan backwards from fixed constraints: contractual end date, rainy season, interim report date.
• Place a visible project buffer at the end of the critical path (for example 4 weeks over 12 months) rather than padding each task: a buffer hidden in each estimate gets consumed by everyone; a shared buffer is managed by the project manager.
• Compress if necessary by fast-tracking (starting drilling on sites where agreements are signed while the others are being negotiated) or by crashing, i.e. adding resources (a second drilling team); each option has a cost and a risk.
• Level resources: check that the same engineer is not scheduled on three sites in the same week.
• Set a baseline: the approved schedule is frozen; all updates are compared against it.

The first ten weeks of a twelve-month project are often the most decisive: major procurement, recruitment, agreements and the baseline survey all happen at the same time. A project manager who lets these weeks slip has no margin left for the construction season.

7Budget, procurement plan and resource plan

The work plan, the budget and the procurement plan describe the same project from three angles. They must be built together and remain consistent: an activity without a budget line cannot be paid for; a budget line without an activity will be questioned at audit; an unplanned purchase will arrive too late.

Building the budget from the WBS

1. Cost each work package: quantities × unit costs, separating staff, materials, transport, services, cash transfers and partner costs.
2. Add support costs directly linked to the project: share of support staff, offices, vehicles, communications, security, according to documented allocation keys accepted by the donor.
3. Add indirect costs at the rate authorised by the agreement.
4. Check the key ratios: cost per beneficiary, share of support costs, unit costs compared with cluster benchmarks.
5. Draw up the month-by-month disbursement plan: it is used to request advances from the donor, to plan cash flow and later to measure the burn rate.

Procurement plan

For each significant purchase, the plan specifies: the item and its specification, the quantity, the estimated value, the applicable procedure according to thresholds, the date by which the need must be expressed, the required delivery date and the person responsible. Thresholds and procedures vary by donor and internal policy. Typically, an NGO policy provides for a purchase on a single invoice below a small amount, a request for quotations from at least three suppliers above it, and a restricted or open tender above higher thresholds, with an evaluation committee and advertising. For US funding, 2 CFR 200 distinguishes micro-purchases, small purchases (simplified acquisition) and formal procedures according to thresholds set by the regulations; European donors define their own thresholds and origin rules in the partnership annexes. The stricter rule between the internal policy and the donor's applies.

Human resources plan

The plan specifies the positions, arrival and departure dates, the percentage of time charged to the project and the expected workload per month. It reveals the peaks: in a construction project, the workload of the engineer and supervisors peaks during the dry season. The budget must cover the legal end-of-contract obligations for national staff (severance pay, notice), which are often forgotten and paid from ineligible funds at the end of the project.

Monitoring and evaluation plan and communication plan

The monitoring and evaluation plan (or indicator tracking matrix) specifies for each indicator: definition, calculation method, source, frequency, person responsible, disaggregation and target per period. The communication plan indicates who receives what information, in what form and how often, building on the stakeholder engagement plan. Both plans have tasks and costs that must appear in the WBS.

8An operational RACI and advanced risk management

A RACI that is actually useful

The RACI becomes useful when it covers the decisions and deliverables that cause problems, not every task. Fifteen to twenty-five lines are enough. It is built in a meeting with the people concerned, then checked against four rules: only one A per line; at least one R; no line where everyone is C (paralysis); no one who is A on everything (bottleneck).

Notes: a contract award is never approved by the requester alone (segregation of duties); the beneficiary list is drawn up by a team separate from the one that distributes; sensitive complaints are handled outside the operational line.

Advanced risk management

The register from the core course becomes a management tool when it incorporates four additional elements.

• Categories: security, access, operational (procurement, logistics), fiduciary (fraud, corruption, diversion), protection and safeguarding (sexual exploitation and abuse, child protection), legal and compliance (sanctions, counter-terrorism), reputational, contextual (prices, currency, climate), partner.
• Risk appetite: the organisation defines which levels of risk it accepts by category. Appetite is generally very low for fraud and zero for sexual exploitation and abuse, higher for operational risks inherent in emergencies.
• Inherent and residual risk: the risk is scored before mitigation measures, then after. If the residual risk exceeds the appetite, it is escalated.
• Early warning indicators and trigger thresholds: what signals that the risk is materialising (cement prices up by more than 15%, incidents on the road), and what triggers the contingency plan.

Escalation thresholds (example on a scale of 1 to 25)

Sharing risks with partners and donors

Transferring all risks to the local partner through the contract is not an answer: the partner bears them without the means to manage them. Recent thinking in the sector, in particular the Risk Sharing Framework developed by a group of donors and humanitarian organisations, encourages open discussion of risks with the donor from the design stage, agreement on what is and is not accepted, and advance planning of how losses will be handled when they occur despite reasonable measures. The project manager documents the measures taken: this is what will make it possible to demonstrate due diligence if a loss occurs.

A risk register is only of value if it changes decisions. At every steering committee, ask three questions: which risks have changed level, which early warning indicators have been triggered, which mitigation actions are behind schedule?

9Full case study: the integrated plan for the Kalema project

Let us return to the Kalema WASH project scoped on day 1: 22,000 people, 12 months from February to January, EUR 1,205,000, partner AMD. The project manager has three weeks after signature to produce an integrated plan approved by the steering committee.

1. The level 2 WBS

2. The budget by heading

3. The baseline schedule (month 1 = February)

Reasoning: months 6 and 7 (July and August) are the rainy season; no drilling is scheduled, and site C works are completed beforehand. The remaining four boreholes are scheduled in months 9 and 10, leaving two months of buffer before the contractual end date. The interim report is prepared in month 6 for submission at the end of month 6.

4. The initial risk register

The steering committee approves the plan at the end of week 3. The baseline is frozen: the schedule, the monthly disbursement budget, the procurement plan and the register become the monitoring references for day 3.

10Common mistakes, variations by context and donor expectations

The most frequent planning mistakes

• Planning without the support services: the project manager discovers in month 3 that the tender procedure takes 8 weeks. Remedy: a planning workshop with logistics, finance and HR from week 1.
• Ignoring seasonality: civil works scheduled in the middle of the rainy season, distributions scheduled during harvests. Remedy: a seasonal calendar built with communities and overlaid on the Gantt chart.
• A work plan without dependencies: a list of dates that does not show the effect of a delay. Remedy: enter the predecessors and recalculate the critical path.
• Underestimating start-up: very low spending in the first three months followed by a rushed catch-up, non-compliant emergency purchases at the end of the project. Remedy: a realistic disbursement plan and tracking of the burn rate from the first month.
• A risk register filled in for the donor and never reread. Remedy: review at the steering committee, with early warning indicators.
• A theoretical RACI: the matrix says one thing, practice another. Remedy: build it with the people concerned and review it after three months.

What donors and auditors check

• Consistency between the logframe, the work plan and the budget: every activity has resources, every cost is linked to an activity.
• Compliance with procurement procedures according to thresholds, traceability of evaluation committees, segregation of duties.
• The burn rate compared with the activity schedule: significant underspending at mid-term calls for explanations and often a revision.
• The quality of the risk analysis, including fiduciary and safeguarding risks, and the existence of real mitigation measures.
• Compliance with sanctions regimes and the screening of suppliers, partners and, depending on the context and applicable law, certain contractors.

Variations by context

In an unstable context, a good plan is not one that never changes: it is one that makes it possible to measure immediately the effect of a change on the end date, the budget and the risks.

11Interview questions and key references

Interview questions

1. "Explain the critical path to a team member who is not familiar with it." Expected answer: the longest chain of dependent tasks; it sets the minimum duration; its tasks have no float; a concrete example from the field (international procurement, rainy season); what you do to protect the critical path.
2. "The pump supplier announces a six-week delay. What is your approach?" Check whether the task is on the critical path; calculate the effect on the end date; options (back-up supplier, air freight, reorganising tasks, postponing sites to the next season) with their costs; decision at the right level; inform the donor if the end date or deliverables are at risk; update the register.
3. "How do you build a project budget?" From activities and quantities, verified unit costs, support costs allocated according to documented keys, indirect costs at the authorised rate, control ratios, disbursement plan, validation by finance, taking the donor's eligibility rules into account.
4. "Give three major risks for a cash transfer project and your response." Strong examples: inclusion or exclusion errors (verification, complaints), fraud or informal taxation by third parties (post-distribution monitoring, confidential channels), inflation or market disruption (price monitoring, trigger thresholds, switch to in-kind assistance), protection risks linked to payments.
5. "How do you prevent a RACI from remaining theoretical?" Participatory construction, limited to sensitive decisions, a single A, review after a few months, use when disputes over responsibilities arise.
6. "How do you share risks with a local partner?" Joint assessment, mitigation measures funded in the partner's budget, security and compliance costs covered, discussion with the donor, no wholesale transfer of risk through the contract.

Key references

• PM4NGOs, PMD Pro Guide (chapters on planning, scheduling, budgeting and risk).
• Project Management Institute, A Guide to the Project Management Body of Knowledge (PMBOK Guide), for the critical path method and estimating.
• Sphere Project, Sphere Handbook 2018, WASH chapter and standards on planning.
• Cash Learning Partnership (CALP), Programme Quality Toolbox and resources on market analysis.
• International Committee of the Red Cross, Rapid Assessment for Markets and market analysis tools.
• The Netherlands, International Committee of the Red Cross and InterAction, Risk Sharing Framework (2023).
• InterAction, Pathways to Risk Sharing.
• IASC, Grand Bargain commitments, in particular on localisation and harmonised, simplified reporting.
• United States Government, 2 CFR 200, sections on procurement standards and budget revisions.
• European Commission, humanitarian partnership annexes on procurement and cost eligibility.
• Logistics Cluster, Logistics Operational Guide (LOG), for procurement and transport lead times and procedures.

Key takeaway: a credible plan rests on explicit dependencies, realistic procurement lead times, a budget linked to activities, a RACI limited to the decisions that matter and a risk register that triggers action.