Status Reports: Structure, RAG Definitions, Template
A status report that stakeholders read opens with three things: one overall rating, the decision or help the project needs from the reader, and the date by which it is needed. Progress detail, risks and metrics follow for the few readers who want them. The rating only earns trust if its colours have written definitions, and the clearest published examples come from governments that rate their major projects in public: the US federal IT Dashboard and the UK’s Government Major Projects Portfolio. The US record shows what happens when judged ratings go stale, the UK’s written definitions show how to pin each colour down, and the rules below are built from both.
| Element | What it contains | Why it sits there |
|---|---|---|
| Rating | Green, amber or red, with a one-line reason | Readers who stop after one line still know the health |
| Ask | The decision or action needed, a named owner, a date | The report exists to trigger action |
| Movement | Rating this period against last period, and the forecast against the baseline | Direction matters more than the colour |
| Evidence | Delivered results, open risks, next-period plan | Lets a sceptical reader check the rating |
| Links | Board, schedule, risk register | Detail stays one click away, off the page |
How the US federal IT Dashboard rates projects
The federal IT Dashboard has published agency ratings of major IT investments since June 2009 and, by GAO’s 2026 count, covers nearly 600 of them at 26 agencies. GAO’s reviews of it in 2016 and 2026 describe three separate ratings, and the split is worth copying.
- Cost and schedule ratings are calculated. The Dashboard takes agency-submitted planned and actual figures, works out the variance, and assigns a colour: a variance under 10 percent is green, 10 percent to under 30 percent is yellow, and 30 percent or more is red. GAO gave the same 10 and 30 percent bands in 2016 and again in 2026.
- The CIO rating is a judgement. The agency’s chief information officer assesses each investment against OMB’s evaluation factors and assigns a score from 1 (high risk) to 5 (low risk).
The six factors OMB suggested for that judgement double as a checklist for any status rating:
| OMB factor | Example test from the Dashboard instructions |
|---|---|
| Risk management | Risk log is current and complete; risks are clearly prioritised; mitigation plans exist |
| Requirements management | Objectives are clear and scope is controlled |
| Contractor oversight | Reports such as earned value and risk logs are received and acted on |
| Historical performance | No significant deviations from planned cost and schedule |
| Human capital | Qualified team and low turnover |
| Other | Anything else the CIO considers important to forecasting success |
The five scores map onto three colour families. In GAO’s October 2012 report on the Dashboard and its 2016 review, 5 (low risk) and 4 (moderately low) showed as green, 3 (medium) as yellow, and 2 (moderately high) and 1 (high) as red. By GAO’s 2026 description the Dashboard shades them apart: green, light green, yellow, light red and red. Either way, a five-point input on a red-yellow-green display keeps the nuance for the person rating while keeping the display simple for the reader.
What went wrong: ratings that showed less risk than reality
GAO compared CIO ratings with its own assessments of 95 investments in 2016. The CIO rating matched GAO’s 22 times, GAO’s assessment showed more risk 60 times, and less risk 13 times. GAO identified three contributing issues:
- Stale ratings. Forty of the 95 CIO ratings were not updated in the month GAO reviewed. A colour that nobody revisited kept describing last month’s project.
- Slow rating processes. Three agencies’ processes took longer than a month, so ratings rested on older data.
- Ratings that ignored active risks. Seven agencies’ processes did not focus on live threats such as funding cuts or staffing changes, and some only checked whether risk-management procedures had been followed. Procedural compliance is not the same as low risk.
The pattern repeated a decade later. In a report released on 7 October 2026, GAO-27-108416 compared ratings for 53 investments at 12 agencies: GAO’s assessment showed more risk 24 times, matched the CIO rating 27 times, and showed less risk twice. Twenty-one of the 53 ratings were not updated on time under the agencies’ own processes, and GAO made 17 recommendations to nine agencies. The full report notes that OMB’s Circular A-11 of August 2025 requires CIO risk evaluations at least once per quarter, and that in April 2026 OMB announced plans to sunset the Dashboard and replace it with a streamlined system, with no timeframe given.
The practical lessons for a team report are direct. Rate on a fixed cadence, date every rating, and base it on what could go wrong next, not on whether the team followed its process.
How the UK Government Major Projects Portfolio rates projects
The UK government gives each project on its Government Major Projects Portfolio (GMPP) a Delivery Confidence Assessment (DCA) and publishes the ratings once a year, apart from a few exempted on grounds such as national security. The Infrastructure and Projects Authority (IPA) oversaw the portfolio until 1 April 2025, when it merged with the National Infrastructure Commission to form the National Infrastructure and Service Transformation Authority (NISTA), which now publishes the annual report. The definitions below come from the IPA’s Annual Report on Major Projects 2023-24, which covers 227 projects and defines the DCA as an assessment of the likelihood of a project delivering its objectives to time and cost. Its Annex A gives the wording of each rating, and it is the best published model of a RAG definition:
| Rating | IPA definition (condensed) | Reading |
|---|---|---|
| Green | Successful delivery on time, budget and quality appears highly likely; no major outstanding issues threaten delivery | Nothing is expected to need senior action |
| Amber | Delivery appears feasible but significant issues already exist and need management attention; resolvable if addressed promptly | A real problem exists but looks fixable if handled promptly |
| Red | Delivery appears unachievable; major issues with definition, schedule, budget, quality or benefits that do not appear manageable | The project may need re-scoping or a viability review |
Three details stand out. First, each definition describes the likelihood of success and the state of the issues, not the project’s percentage complete. Second, Amber requires that issues “already exist”, so the colour marks real problems, not a vague unease. Third, the 2023-24 scheme also has a Reset marker for a significant change to a project’s baseline that involves a business case refresh or change. A moved baseline shows up in the ratings as an event in its own right, not absorbed silently into the plan.
The IPA used a five-tier scale with Amber/Green and Amber/Red until June 2021, when it moved to the three tiers above. The report says DCAs “are not a comprehensive reflection of project performance” and that a changing DCA can show risks are being identified and mitigated. A rating that moves from green to amber because a risk was found early is the system working. That framing matters when explaining an amber to an executive who reads it as failure.
NISTA’s Major Projects Annual Report 2025-26, published on 13 July 2026, keeps the same DCA definition and the same three colours, and adds that a red rating “does not mean a project will fail”. At its end-March 2026 snapshot the portfolio held 189 projects: 29 green, 109 amber, 34 red and 17 exempt. Project owners update their rating each quarter, and where NISTA has supported or independently reviewed a project in the previous six months, NISTA’s own rating is published instead. From 1 April 2026 the portfolio was narrowed to about 80 of the most complex projects, with the rest managed by departments.
Write RAG definitions your team can apply
The government scales share a structure that fits a team report: a written test per colour, a mix of calculated and judged inputs, and a defined update cycle. A proposal modelled on them, with thresholds borrowed from the Dashboard’s 10 and 30 percent bands and meant to be adjusted for the project:
| Green | Amber | Red | |
|---|---|---|---|
| Schedule forecast vs baseline | Under 10% later | 10% to under 30% later | 30% or more later |
| Scope and benefits | All committed scope on track | A named item at risk, with a mitigation under way | Committed scope cannot be delivered as agreed |
| Open issues | None needing sponsor action | Issues exist; owner and fix date known | Issues without a workable fix |
| Reader action | None | Awareness; help may be requested | A decision is needed now |
Four rules keep the table honest:
- Calculate what can be calculated. The Dashboard does not let anyone choose the colour for cost and schedule; it derives them from planned and actual numbers. Do the same for schedule, using the forecast against the approved baseline.
- Make judged ratings rest on listed factors. Use a list such as OMB’s six factors, and name the factor that drives the colour.
- Take the worst of the inputs. If the schedule is green but a committed deliverable cannot ship, the overall rating is red.
- Date every rating and state its age. In both 2016 and 2026, GAO named ratings that had not been updated as one cause of the gap between CIO ratings and its own assessments.
For the issue and risk entries, a risk register gives each item an owner and a score so the report can cite it without restating it.
Template: a one-page status report
Use a fixed order so readers learn where to look. The bottom line comes first because the people who most need it read the least.
| Section | Content | Limit |
|---|---|---|
| Header | Project, period covered, report date, author | One line |
| Overall rating | Colour, previous colour, one-sentence reason | One sentence |
| Ask | Decision or action, owner, date, consequence of missing it | Up to three items |
| Schedule | Baseline date, current forecast, variance percentage | One table row |
| Progress | Results delivered, with a denominator (“9 of 12”) | Up to five lines |
| Risks and issues | Item, owner, mitigation, date | Top three |
| Next period | Committed results, not activities | Up to five lines |
| Links | Board, schedule, register | Links only |
The schedule row works only if the baseline stays fixed. GAO’s schedule assessment guide describes a schedule as a tool for measuring performance against an approved plan, which is why a forecast gets reported next to the baseline and never over it. The critical path method explains how to compute the forecast in the first place, and milestone planning covers choosing the checkpoints the report counts.
Worked example: a help-center launch
The following is a sample report for a fictional project, with invented figures, using the template and the rules above. The project began on Monday 2 November with a baseline launch of Friday 11 December (30 working days) and a forecast that has moved to Friday 18 December.
Help-center launch: weekly status, week 3
| Field | Entry |
|---|---|
| Overall rating | Amber (last week: Green). Three articles lack product screenshots and the preview environment is not yet approved for capture. |
| Ask | Product lead: confirm by Thursday whether the preview environment may be used for screenshots. If not confirmed, the launch owner chooses between moving the review and releasing the nine finished articles first. |
| Schedule | Baseline launch 11 Dec; forecast 18 Dec; 5 working days later, a variance of 17% of the 30-day plan. Amber under the 10-30% rule. |
| Progress | 9 of 12 articles passed technical review. 96 of 120 planned hours used (80%) for 75% of the articles, so effort is running ahead of output. |
| Risks | 1. Screenshot access (owner: product lead; mitigation: confirm environment; date: Thursday). 2. Reviewer availability in launch week (owner: editor; mitigation: book backup reviewer; date: next Monday). |
| Next period | Capture screenshots for three articles; complete the final review; release the checklist for launch day. |
| Links | Article checklist; access request; milestone plan |
The percentage is defined: 5 working days of slip against a 30-working-day baseline. The amber follows from the table of definitions and not from a feeling, and the ask names a person, a day and the consequence of silence. The baseline stays at 11 December in the report; the forecast sits beside it. If the access decision arrives late and the variance reaches 30 percent, the same rule produces a red without any argument about tone.
One detail is easy to miss. The effort figures show the project is 80 percent through its hours at 75 percent of its articles. A report that gave only the article count would look slightly better than the plan does. When two honest measures point in different directions, show both with their denominators.
Match the report to its reader
The same facts go to different readers in different shapes.
| Reader | Needs | Keep out |
|---|---|---|
| Sponsor or executive | Rating, ask, forecast against baseline, business impact | Task lists, internal jargon |
| Technical lead | Dependencies, technical risks, test and deployment status | Budget narrative |
| Client | Deliverables, dates, budget position | Sprint-level figures; say “the login module is in testing”, not “velocity was 32 points” |
Cadence follows the decisions the reader must make. A common pattern is weekly for a sprint-based team, every two weeks for a multi-month project, monthly for an operations stream, and daily for a project in recovery. The US federal minimum for CIO ratings was monthly from June 2015, per GAO, and is quarterly under OMB’s August 2025 circular, the same interval at which UK project owners update their delivery confidence ratings. Whatever cadence is chosen, keep it fixed. For distributed teams, the written report replaces a status meeting, and the team communication guide covers where the report sits among other async channels.
Check the report before sending it
- The colour agrees with the text. An unresolved launch dependency should not sit under a green rating.
- Every percentage has a denominator and the same unit each period. Ten tasks of twenty is not half the effort when the remaining ten are larger.
- Every request has an owner and a date. Replace “support needed” with the action, the named role and the deadline.
- The rating was updated this period. GAO found 40 of 95 CIO ratings not updated in the month it reviewed in 2016, and 21 of 53 not updated on time in 2026.
- Amber was reported early. The UK reports note that a changing rating can reflect risks being identified and mitigated. A project that goes green to red in one report gave its readers no chance to help.
- Details are linked, not pasted. The first page carries the decision and its evidence; the board and register carry the rest.
Sources
- GAO-16-494, IT Dashboard: Agencies Need to Fully Consider Risks When Rating Their Major Investments (June 2016)
- GAO-13-98, Information Technology Dashboard: Opportunities Exist to Improve Transparency and Oversight of Investment Risk at Select Agencies (October 2012)
- GAO-27-108416, IT Dashboard: Selected Agencies’ Investment Ratings Fail to Fully Consider Risks (7 October 2026)
- GAO-27-108416 full report (HTML)
- Infrastructure and Projects Authority, Annual Report on Major Projects 2023-24
- NISTA, Major Projects Annual Report 2025-26 (13 July 2026)
- GAO-16-89G, GAO Schedule Assessment Guide (December 2015)