Introduction
Weekly updates are common in fast-moving phases such as commissioning, recovery and the final months before a key milestone. Done well, they keep the team focused on what drives the date. Done poorly, they become a ritual that moves dates around without anyone acting on them.
A useful weekly update follows the same five steps every time, takes a predictable amount of effort and ends with decisions, not just a new finish date.
1. Set a firm progress cutoff
Agree a fixed cutoff, for example end of shift on Thursday, and use it as the data date. Everyone reports progress up to that moment and no later. Without a firm cutoff, some trades report Monday's status and others report Wednesday's, and the update describes a moment that never existed.
2. Validate progress before accepting it
Reported progress should be checked against evidence: daily reports, inspection records, delivery tickets and a walk of the areas that matter most. Focus validation effort on critical and near-critical activities. That is where an optimistic percentage does the most damage.
Ten minutes on site checking three critical activities is worth more than an hour checking every percentage in the office.
3. Calculate and compare
Before calculating, correct any out-of-sequence progress in the logic and add any new work. Then calculate and compare against the previous update, not just the baseline. What changed on the critical path? Which near-critical paths lost float? Which milestones moved, and by how much?
4. Write a narrative that explains
The narrative should answer three questions in plain language: what happened this week, why the forecast moved, and what is at risk next week. Keep it short. A weekly narrative that takes more than two minutes to read will not be read.
5. Assign actions
Every risk to a milestone should end in an action with an owner and a date. Track these actions in a register and review them at the start of the next update meeting. This is where the update turns into management.
Practical checklist
- Fixed cutoff agreed and used as the data date.
- Progress collected from every responsible party.
- Critical and near-critical progress validated against evidence.
- Out-of-sequence logic corrected, new work added.
- Comparison with last week's update, not only the baseline.
- Two-minute narrative: what happened, why it moved, what's next.
- Action register updated with owners and dates.
Common mistakes
- Accepting percentages without evidence, especially on critical work.
- Comparing only to the baseline, which hides this week's movement.
- Narratives that list activities rather than explaining variance.
- Updates with no actions, which turn the process into reporting for its own sake.
Conclusion
A weekly update is worth the effort only if it changes what people do the following week. A firm cutoff, validated progress, an honest comparison, a short narrative and clear actions make that happen.
