Writing for people who live in the schedule
Schedule quality, forensic delay analysis and quantitative risk — explained by practitioners, with the charts to prove it. No fluff, no vendor-speak: the same standards and methods we built into the product, written up properly.
The DCMA 14-point assessment, properly explained
It has become the de facto quality standard for CPM schedules far beyond US defence contracting — and it is routinely misunderstood. Every metric, every threshold, where the 44-day limits actually come from, and which failures matter most: the reference guide we wish someone had handed us years ago.
The anatomy of schedule quality: what the DCMA check reveals across a portfolio
Bad schedules don't fail at random — they fail in a recognisable fingerprint, and those structural failures quietly predict which projects will slip months later. An analytical field model: the failure fingerprint, the link between structure and slippage, the decay curve, and what a top-decile schedule actually looks like.
How to open an XER file without Primavera P6 (free, in your browser)
Five honest routes compared — desktop viewers, trials, MS Project import, Excel and browser tools — plus the confidentiality question nobody asks before uploading a live commercial programme to a stranger's server.
Critical path vs longest path in P6: why total float lies to you
P6 has two definitions of "critical" and on any constrained or multi-calendar programme they disagree. Which setting to use, when float can't be trusted, and why every delay report should say which one it means.
How to read a P6 schedule when you're not a planner
The seven things to find before reading any bars, the five questions that instantly raise the meeting's IQ, and the "90% complete" confession hiding in the remaining-duration column.
NEC4 clause 31: getting your programme accepted (and why most aren't)
The Accepted Programme is the keystone of the contract — and the four rejection reasons are mostly quality failures in disguise. Clause 31.2 as a self-audit checklist, float ownership, and the deemed-acceptance trap.
The schedule narrative: the report everyone requests and nobody enjoys writing
A seven-section skeleton with a copy-paste template, the basis document vs the narrative, and why the report that admits the slip reads as competence — while the one that hides it becomes evidence of concealment.
The schedule basis document: what to include, and why it wins disputes
The companion document that records how the schedule was built — assumptions, durations, logic, constraints and exclusions. AACE 38R-06 walked through section by section, and why a contemporaneous basis is the strongest answer to "this baseline was never realistic".
The time-scaled logic diagram: seeing the driving path, not just the bars
A Gantt scatters the critical path down hundreds of rows. A time-scaled logic diagram — P6 Visualizer's signature view — condenses the driving chain onto one band so you can read it left to right. What it is, how to read one, and when it beats a Gantt.
Out-of-sequence progress: the setting that quietly moves your completion date
Retained logic, progress override and actual dates give three different completion dates from the same file. What each really does to a progressed network, and the four logic repairs that fix it properly.
Reviewing a monthly schedule update in thirty minutes
A working routine built on AACE RP 53R-06: the eleven things that can change between two revisions, the order to check them in, and the float-erosion signal that moves before the date does.
Schedule quality: what it is, and why it decides disputes
A high-quality schedule can still be wrong — but a low-quality one can't even be wrong honestly. Why structural integrity, not optimism, is what tribunals and boards end up trusting.
The five scheduling sins the DCMA check catches most
Open ends, lag abuse, constraint pinning, six-month activities and the unstatused baseline — what each one looks like in P6, why good planners commit them anyway, and how to fix them.
Leads, lags and logic density: the hidden mechanics of your network
The four relationship types and when each is legitimate, why leads break the maths, the lag-calendar trap almost nobody knows about, and what your relationships-per-activity ratio is really telling you.
Float: the most misunderstood number in project controls
Total vs free float, what negative float actually says about your committed dates, and the oldest fight in construction: who owns it. With the forward-and-backward-pass refresher you'll actually remember.
Baselines: the discipline that decides whether you can prove anything
The baseline is the contract's memory. When to re-baseline (rarely, formally) versus how it actually happens (quietly, to hide slip) — and why contemporaneous updates are gold when the dispute arrives.
Earned schedule: the fix for EVM's broken clock
SPI converges to 1.0 on every project, however late it finishes. How earned schedule restates performance in units of time — ES, SV(t), SPI(t) and the TSPI test that calls out fictional recovery plans.
The GAO ten best practices, and the three things the DCMA 14 cannot see
GAO-16-89G's four characteristics and ten best practices, mapped check-by-check against the DCMA 14 — and the three questions no structural test on a file can answer.
Resource levelling and schedule quality: what happens to your critical path
Loading, levelling and smoothing are three different states. Why levelling delays activities without adding logic — and why total float quietly stops meaning what you think afterwards.
Forensic delay analysis: choosing your method before someone chooses it for you
The AACE 29R-03 family tree and the SCL Protocol's six methods, mapped onto each other in plain language — and the records test that quietly eliminates most of your options.
As-planned vs as-built: the oldest delay method, done properly
Tribunals like it because it's transparent; opponents attack it because it can't isolate causation. When MIP 3.1 is the right call, how to build a defensible as-built, and why total-time claims keep failing.
Time impact analysis vs windows analysis: which one does your dispute need?
Two names everyone reaches for, frequently interchangeably and wrongly. One is prospective, one retrospective — and using the wrong one hands your opponent their cross-examination script.
Concurrent delay: the argument every big claim ends up having
The employer's best defence and the contractor's biggest fear. What true concurrency actually requires, why it's rarer than claimed, and how your choice of analysis method changes the answer.
EOT under FIDIC: your programme is a condition of entitlement
Sub-clause 8.3 says what the programme must contain and when. Sub-clause 20.2 sets the 28-day notice bar. And 20.2.7 lets the Engineer reduce an extension of time because the programme did not comply.
QSRA explained: what Monte Carlo actually does to your schedule
Why your deterministic finish date usually sits at P10–P30, what merge bias does at every path convergence, and how to read P50, P80, criticality and tornado charts like they matter — because they do.
Building a risk-loaded schedule: three-point estimates that mean something
Ranges from evidence rather than optimism, the too-narrow-range disease, why correlation can't be skipped, and the six steps that make a QSRA output worth believing.
Stop reading about schedule quality. Measure yours.
Drop a P6 XER or MS Project file in your browser — full DCMA 14-point check, analytics and risk in seconds. Nothing is uploaded.