Weekly Project Status Reports
Last updated: September 14, 2026
Most project managers spend Monday morning rebuilding the same report. They open last week's plan, work out what actually moved, check who is overloaded, decide whether the project is still green, and write it all into an email before the sponsor asks. The work is repetitive, but it is not trivial, because the judgement in it is the part the sponsor is paying for.
Project Paced produces that report from your live project data. The facts are calculated, not written. The narrative is drafted for you, and you can edit any of it before it goes out.
What a weekly status report needs to contain
A report that a sponsor will actually read does four things. Anything beyond this is usually padding.
A status, and the reason for it. A RAG rating on its own invites an argument. A RAG rating with the specific evidence behind it ends one. State whether the project is red, amber or green, whether that is better or worse than last week, and which phase the project is currently in.
What changed since the last report. Not a restatement of the plan. The delta: what completed, what started, which dates moved and by how much, what was added or removed from scope, and how the team allocation changed.
What is at risk. Milestones that are overdue, milestones due shortly with no progress recorded, work that has stalled, people who are allocated beyond capacity, and any single person carrying too large a share of the project.
What decisions are needed. The specific calls the sponsor has to make, with the risk attached to each. A status report without an ask is a newsletter.
Deciding RAG status consistently
RAG ratings lose their meaning when each report applies a different standard. Two projects rated amber should be in comparable trouble. The usual failure is optimism early in delivery followed by a sudden jump from green to red near the deadline.
A defensible rating is derived from the data first, then adjusted for judgement:
- Count overdue milestones and the size of any date movements against the previous baseline
- Check whether work due soon has actually started
- Check whether anyone is allocated beyond their capacity during the coming period
- Only then apply what you know that the data does not show, such as a supplier warning or a resourcing decision awaiting sign-off
In Project Paced, the first three are calculated automatically from your milestones and allocations. The fourth comes from the project notes you write, which can escalate a rating but cannot quietly downgrade one below what the data supports.
How Project Paced generates the report
Each week the platform stores a frozen snapshot of every active project, capturing milestones and resource allocations as they stood. When you generate a status report, it compares your current project state against the most recent snapshot and calculates the difference.
That difference is computed by a deterministic engine, not by a language model. Every date, percentage, count and status in the report comes from your data. The AI is given the resulting evidence and asked to write the narrative around it, which means it summarises and escalates but does not produce numbers of its own. If AI generation is unavailable, the report still produces plain evidence bullets rather than failing.
The result covers the four sections above: status with rating, trend and current phase; what changed; what is at risk; and decisions needed.
Editing before it goes out
The generated report is a draft, not a submission. You can edit the headline, the rationale and every bullet in place, and edits are saved without consuming any AI allowance. The rating, the current phase and the underlying evidence stay data-driven and are not editable, so the report cannot be quietly rewritten to say something the project data does not support.
Regenerating replaces your edits with a fresh draft.
Sharing the report
| Format | Use |
|---|---|
| Branded HTML report sent to up to five recipients, with replies routed back to you and an optional personal note | |
| Slack or Teams | Formatted text copied to your clipboard, ready to paste into a channel |
| Reports (PDF) | The weekly narrative followed by an executive summary and team roster appendix |
Past weeks remain available, so a report can be re-sent or exported later without regenerating it.
Availability
Weekly status reports are included with Pro and Enterprise plans. Generating a report does not count against your AI timeline generation allowance. Reports can be regenerated after a short cooldown period per project.
Reports can also be read from Claude, ChatGPT and other connected assistants through our MCP server, including the rating, trend and recent history, which is useful for portfolio reviews across several projects at once.
New projects produce a current-state report in their first week and begin reporting week-on-week changes once a baseline snapshot exists.
Frequently asked questions
How often are snapshots captured? Once a week. Snapshots are retained on a rolling 52-week basis, so you keep a year of project history for trend analysis.
Can the AI invent a date or a percentage? No. All quantitative facts are calculated from your project data before the AI is involved. The AI writes the narrative around that evidence.
Do project notes affect the rating? They can raise it. A noted blocker or material risk can escalate a rating, but notes cannot lower a rating below the level the underlying data supports.
Does generating a report use my AI generation allowance? No. Status report generation is separate from timeline generation limits.
Can I send the report to my sponsor directly? Yes. Reports can be emailed to up to five recipients at a time, with replies routed to your own address.