07 · Project and strategy M&E
The project management–M&E interface
Project management and M&E measure the same project with different questions: PM asks whether the work is on scope, on schedule and within budget; M&E asks whether the work is producing the change it promised. A programme can be perfectly on schedule and achieving nothing — and unless the two functions share artefacts, data and review moments, only one of them will notice.
Last updated · Reviewed against 4 cited sources
Two professions, one project
Every funded project is watched by two measurement traditions that grew up separately and still, in most organisations, barely speak. Project management descends from engineering and cost control: its questions are whether the scope is being delivered, on schedule, within budget, to specification. Monitoring and evaluation descends from public policy and programme science: its questions are whether the delivered work is producing outputs, whether the outputs are turning into outcomes, and whether the whole undertaking remains worth doing [3, 4].
The two traditions have separate professional bodies, separate vocabularies, and — critically — separate data. The Project Management Institute maintains the PMBOK Guide as the delivery profession’s standard of record; its seventh edition reorganised the discipline around principles and performance domains, measurement among them, signalling that even the delivery profession now treats measurement as a first-class concern rather than a reporting afterthought [1]. On the other side, the results-based management literature has spent two decades arguing that implementation monitoring — tracking activities and spending — is necessary but radically insufficient, because it cannot say whether anything changed [3].
The honest statement of the interface is this: a project can be on scope, on schedule and on budget, and be failing. Delivery metrics measure the machine; results metrics measure what the machine is for. Neither stream is a proxy for the other, and a governance arrangement that reads only one of them is flying with half its instruments.
The hand-off map: who owns which data
The cleanest way to organise the interface is to be explicit about which function owns which artefact, and where each hands data to the other.
| Artefact | Owner | What it answers | Hand-off |
|---|---|---|---|
| Work breakdown structure | Project manager | What work exists | M&E maps output indicators onto WBS elements |
| Schedule and milestones | Project manager | When work completes | M&E times data collection off milestone dates |
| Budget and cost reports | Project manager | What delivery costs | M&E joins burn rate to results in reviews |
| Indicator plan and reference sheets | M&E officer | What change is measured, and how | PM sees which deliverables carry measurement duties |
| Monitoring data and analysis | M&E officer | Whether outputs and outcomes are emerging | PM reads results at stage gates, not only at evaluations |
| Risk register | Shared | What could derail delivery or results | Both functions feed and read it |
The most common failure is duplication with divergence: the project plan says a training package is 90% complete while the M&E system, counting verified attendance records, says 60% of the target has been reached. Both numbers may be defensible on their own definitions. A project that has not decided, in writing, how percent-complete relates to indicator achievement will discover the discrepancy in front of a donor.
The WBS as monitoring skeleton
The work breakdown structure — the hierarchical decomposition of the project’s full scope into deliverable-oriented elements — is the delivery profession’s foundational artefact, and it is the best free gift project management offers M&E. A well-built WBS enumerates everything the project will produce. That enumeration is precisely the population of things output monitoring must cover.
The conversion discipline is straightforward:
- Every terminal WBS element should map to at least one output indicator, or to an explicit decision not to measure it. Unmapped elements are work the results system cannot see; unmapped indicators are measurements of work nobody has planned.
- Every milestone should be a verifiable event, not a date. “Curriculum approved by the ministry, approval letter on file” is a milestone; “curriculum work complete” is a mood. Milestone verification criteria belong in writing before the milestone, for the same reason indicator definitions belong in reference sheets before data collection.
- Milestone dates drive the measurement calendar. Output data collection is scheduled off delivery dates; outcome data collection is scheduled off delivery dates plus the lag — the point below.
Handled this way, the WBS becomes the shared spine: the project manager reads it as a delivery contract, the M&E officer reads it as a sampling frame, and both are reading the same document.
Percent-complete is not an outcome
The delivery statistic most likely to impersonate a result is percent-complete. It is a legitimate and useful number — for what it measures, which is the proportion of planned work performed. It carries no information about effect. Eighty per cent of boreholes drilled says nothing about water consumption; eighty per cent of training sessions held says nothing about practice change. The results-based management literature exists in large part because governments and agencies spent decades reporting implementation percentages as achievements and were eventually asked, by legislatures and publics, what all the completed activity had accomplished [3].
The practical rule: percent-complete may appear on a results dashboard only in a clearly labelled delivery lane. The moment a slide aggregates “82% complete” with outcome indicators into a single traffic light, the programme has quietly asserted that doing the work and achieving the change are the same thing — the exact proposition M&E exists to test [3, 4].
Schedule health is auditable
M&E officers tend to treat the project schedule as the project manager’s private weather. This forfeits a real asset, because schedule quality is not a matter of opinion: it has checkable properties, and the U.S. Government Accountability Office has published an open, detailed statement of them. The GAO Schedule Assessment Guide’s best practices require, among other things, that the schedule capture all the work in the WBS, that activities be logically sequenced with real predecessor–successor links rather than fixed dates, that resources be assigned, that the critical path be valid and continuous, that total float be reasonable, and that a baseline schedule be maintained against which progress is measured [2].
Two of these matter directly to results measurement:
- A logic-linked schedule produces credible forecasts. If activities are tied by dependencies, a slip propagates visibly, and the M&E calendar can respond before the endline lands in the wrong season. A schedule made of manually typed dates absorbs slippage silently until it is unrecoverable.
- The critical path tells M&E where delivery risk concentrates. Indicators fed by critical-path deliverables are the ones whose data will arrive late when anything goes wrong; monitoring plans should flag them and hold fallback measurement windows.
Where a donor or supreme audit institution reviews the project, the schedule will be assessed against exactly these properties [2]. A programme whose own team has never checked them is choosing to hear the verdict from its auditor.
Results lag delivery — plan for the gap
Outputs appear when work completes. Outcomes appear when people and institutions respond to the outputs, which takes time — a season for agricultural practice, a school year for learning, longer for institutional change. This lag is the single most common cause of false null findings in project-cycle measurement: an outcome survey timed to the delivery milestone, because the milestone was the only date anyone had, measures a change that has not had time to occur [4].
The integrated calendar therefore needs three lanes: delivery milestones; output measurement (at or shortly after delivery); and outcome measurement (delivery plus a stated, justified lag). The lag is a substantive claim about how the intervention works — it belongs in the programme’s design logic, and the theory that generates it is framework territory this site deliberately leaves to its practice-focused sibling.
Governance rhythms: gates, reviews, and one dashboard
Delivery governance runs on stage gates and milestone reviews: periodic decision points where the project’s continuation, scope or budget is confirmed. Results governance runs on monitoring reviews, mid-term reviews and evaluations. The stage-gated tradition — PRINCE2 is the widely used named alternative to PMI’s framing, noted here by name only — and the results tradition both assume they are the project’s real decision-making forum, and in a badly wired project each makes decisions the other never hears about: the delivery board re-scopes an output whose indicator the M&E unit continues to report against; the mid-term review recommends a pivot the schedule cannot absorb.
Three fixes, all cheap:
- Put a results agenda item in every stage gate. Not the full dashboard — the two or three indicators most exposed by the decision at hand, presented by the M&E officer in person [3].
- Time-align the review calendars. A mid-term review that reports three weeks after the annual re-budgeting meeting has chosen decoration over influence. The evaluation-planning guidance is blunt that evaluative work must be timed to feed decisions, not follow them [4].
- Build one integrated dashboard with two labelled lanes — delivery (milestones, burn, schedule variance) and results (outputs, outcomes against target) — rather than two dashboards maintained in mutual ignorance. The design questions this raises are covered in the reporting cluster.
Where the interface goes deeper
Two neighbouring pages take the delivery-measurement thread further: earned value management formalises the schedule–cost–work triangle into a single analytic system, and is the right tool for works-heavy components — with sharp limits this cluster is candid about. The risk register is the other genuinely shared artefact, where delivery threats and results threats belong in one place. And the mid-term review is the institutional moment where the two professions are forced, usefully, into the same room.
Checklist for a working PM–M&E interface
- Every terminal WBS element maps to an output indicator or a written decision not to measure it.
- Milestones have written verification criteria; milestone dates drive the measurement calendar, with stated lags for outcome measurement.
- Percent-complete appears only in a labelled delivery lane, never aggregated with outcome indicators.
- The schedule is logic-linked with a valid critical path and maintained baseline — checked against the GAO best practices, not assumed.
- Stage gates hear results; results reviews are timed to feed budget and re-planning decisions.
- One risk register, fed and read by both functions.
Sources
- A Guide to the Project Management Body of Knowledge (PMBOK® Guide) — Seventh Edition and The Standard for Project Management — Project Management Institute, 2021.The profession's standard of record. Paywalled; cited here for its existence and structure — including its shift to principles and performance domains, among them measurement — not for paraphrased content.
- Schedule Assessment Guide: Best Practices for Project Schedules (GAO-16-89G) — U.S. Government Accountability Office, 2015.An open, auditable statement of what a healthy schedule looks like: logic-linked activities, a credible critical path, reasonable float, a maintained baseline.
- Ten Steps to a Results-Based Monitoring and Evaluation System — The World Bank, 2004.Kusek & Rist. The standard treatment of the results side of the ledger: what results-based monitoring adds to implementation monitoring.
- The Magenta Book: Central Government Guidance on Evaluation — HM Treasury, United Kingdom, 2020.Distinguishes process and impact questions and ties evaluation planning to the delivery cycle.