Many businesses produce reports that are full but uninformative. The monthly management pack lists jobs completed, budget movements, milestones, safety statistics and project updates. Someone spends days assembling it. The meeting works through it page by page. And at the end, the owner is still unsure whether things are healthy, what has changed and what needs deciding.
The problem is rarely a shortage of data. It is that reports are designed around what can be reported rather than around what needs to be decided. Three further problems compound it: the way information is shown can hide the relationships that matter most; each layer of summarising removes the odd, early details that are the best warnings; and the most important fact is often buried where nobody reads it.
This article explains how to design reports around decisions, why forecasts matter more than status colours, how to set thresholds for escalation, why showing relationships between projects changes decisions, how to keep early warnings from being summarised away, and how to write decision papers that put the evidence where readers actually look.
A report is a decision tool
A report earns its place only if it improves understanding, prompts a decision or changes action. For each significant item, it should answer five questions:
- What has changed?
- Why has it changed?
- What outcome is affected?
- What happens if nothing is done?
- What decision or support is needed, and by when?
This moves reporting from describing activity to enabling intervention. A useful test for every page or metric is: which decision could this change? If nobody can answer, it probably belongs in a lower-level working document, not in front of the owner.
Measures also shape behaviour. If people are judged on staying green, they will defend green. If percentage complete matters more than verified results, reported progress will run ahead of real progress. For each measure, ask what behaviour it encourages and whether it can be improved without improving the outcome.
Forecasts matter more than status colours
Most reports focus on current status against the plan. But decisions are about the future. A project can be green today because problems have not yet reached a reported milestone, or red because of a past problem that has since been fixed.
Ask for forecasts instead: expected completion date, final cost, quality and benefits, with the assumptions behind them and how confident the team is. A date backed by a finished design, stable output and available people is very different from the same date that depends on an outstanding approval, a supplier catching up and overtime. The report should make that difference visible.
Leading indicators help here. Cost variance tells you what has already happened. Unanswered decisions, unresolved dependencies, rising defect rates, staff shortages and slipping supplier dates tell you what is likely to happen next.
Set thresholds for escalation
If everything is escalated, the owner is overwhelmed. If nothing is, problems arrive too late. Agree in advance which matters come up, for example anything that:
- exceeds an agreed cost, time or scope tolerance;
- threatens the expected benefit;
- crosses between teams or requires scarce shared people;
- changes a hard-to-reverse commitment;
- creates a safety, customer or reputational risk.
Escalation is not a way of passing responsibility upward. It is a way of getting authority, trade-offs or resources that are not available at the current level. Written escalation rules, stating who decides what and up to what limit, make this easier for everyone.
How information is shown changes the decision
The same information can lead to different decisions depending on how it is presented. In a 2013 experiment, Catherine Killen tested how different ways of showing dependencies between projects affected decision-making. Participants who saw dependencies as a network map, showing which projects relied on which, tended to make better portfolio decisions than those who saw the same information as a list or a table. The study was a controlled exercise with a simplified scenario, so it does not prove the same effect in every business, but its lesson is practical: presentation is part of the decision, not decoration.
Dependencies are hard to see in lists. A list may show that project A depends on B and B depends on C, without making it obvious that a decision about C will affect A. Many problems spread this way: a delayed system holds up a new product, which delays a customer transition, which keeps an old process running, which ties up people needed elsewhere. Each step is manageable. The chain is the real issue. A simple drawing of how the main initiatives connect can reveal what twenty pages of status reports conceal. The mapping dependencies across your projects article explains how to build one.
Killen’s study also found that having enough time was strongly associated with better decisions and greater confidence. A sophisticated report cannot compensate for a meeting that leaves ten minutes for the most important decision. Give significant, hard-to-reverse choices their own time.
Summaries remove early warnings
Every report is a summary of something more detailed, and summarising is necessary. But it is not neutral. It keeps what recurs, adds up and fits an existing category. It discards what is unusual, isolated and hard to describe. That is exactly what an early warning looks like.
The people receiving goods may notice that a supplier’s packaging has changed, paperwork is arriving later and the usual contact has changed twice in three months. None of this fits a supplier scorecard. By the time the problem shows up as late deliveries or quality failures, it has become measurable, and the cheap time to act has passed.
After serious problems, investigations often find that someone close to the work knew. Usually they did raise it with someone. The information simply did not survive the summaries on its way up.
Three practical safeguards help:
- A second, direct channel for unusual observations, kept deliberately small, with a clear idea of what belongs in it and a guaranteed acknowledgement. Its value depends on staying small; once it becomes another report, it gets summarised too.
- Regular direct contact between the owner and the people doing the work, not arranged visits where everyone has prepared what to say.
- A return path: people who raise something should hear what happened. When nothing visibly follows, people stop raising things, and the channel looks quiet while having stopped working.
The what your projects know that your plan doesn’t article looks at a related problem: findings that reach decision-makers but have nowhere to go.
Put the key facts where readers look
Decision papers, such as proposals for new equipment, business cases or major quotes, are rarely read from start to finish. Busy readers go to the summary, then the part that concerns them, then the numbers, then the risks they will have to answer for. A decisive fact on page twenty-two, behind background and method, is effectively absent from the decision.
Write for the way people read:
- State each section’s finding in its first sentence, not its topic. “Market analysis” is a topic. “Two of our three growth assumptions depend on the same customer group” is a finding.
- Use the compression test. If a section cannot be summarised in one sentence with a consequence, either the thinking is unfinished or the section is trying to do too much.
- Put the most decision-relevant fact on the first page.
- Write the summary yourself, and ask someone who did not write the paper to summarise it independently. Where the two summaries differ, the paper is unclear or does not support a claim the author thinks it does.
Selection is fine; suppression is not
There is a difference between leaving out detail and leaving out something that would change the decision. A simple test: would this item, if known, change the decision or the conditions attached to it? If yes, it belongs in the summary, even if it weakens the case. If no, it can go in the body or an appendix. An omission discovered later does far more damage than the fact itself, because it casts doubt on everything else in the paper.
A practical structure for the front of a report
| Section | Content |
|---|---|
| Outcome | What the business or initiative is meant to achieve |
| Position | Current state of delivery, quality, risk and readiness |
| Movement | What has materially changed since last time |
| Forecast | Expected results and how confident we are |
| Exceptions | Matters outside agreed tolerances or needing cross-team action |
| Decisions | Options, recommendation, who decides and by when |
Detailed evidence stays available behind this, for anyone who wants to check it.
A worked example
This is an illustration. A manufacturing business with 60 staff produces a 38-page monthly management pack. The monthly meeting takes three hours and rarely makes more than one or two decisions. Twice in the past year, problems arrived as surprises: a key customer’s order volumes fell sharply, and a supplier’s quality collapsed. In both cases, people on the floor had noticed early signs months before.
The owner redesigns the pack:
- The front four pages follow the structure above: outcomes, position, movement, forecasts with confidence levels, exceptions and decisions required.
- Every remaining page is marked with the decision it supports. Pages without one are moved to a shared folder for anyone who wants them.
- A one-page map shows the five improvement projects, the people and equipment they share and which depends on which.
- A short “unusual observations” item is added, fed by shift leaders and the receiving team, with each observation acknowledged and followed up by name.
- Decision papers must state their main finding and recommendation on the first page, and the owner asks a second manager to summarise each significant paper before the meeting.
Within three months, the meeting takes about 90 minutes and typically makes four or five decisions. The dependency map reveals that two improvement projects need the same maintenance technician in the same fortnight, and one is resequenced. An unusual observation from the receiving team, that a supplier had started sending part shipments without notice, leads to a supplier review that heads off a quality problem before it reaches customers.
How this applies to a small Australian business
Small businesses often have no formal reporting at all, or have inherited packs that nobody questions. Practical steps:
- Design reports around decisions, and ask what decision each item could change.
- Ask for forecasts with confidence, not just status colours.
- Track a few leading indicators.
- Agree escalation thresholds.
- Draw the dependencies between your main initiatives.
- Create a small direct channel for unusual observations, and always close the loop.
- Put the finding first in every section and every decision paper.
- Give important decisions enough meeting time.
The bad news early article covers how a business teaches people whether it is safe to raise problems.
Signals worth watching
- Reports growing while decisions stay slow.
- Status colours changing without new evidence.
- More time spent explaining the past than forecasting the future.
- Bad news arriving informally before it appears in reports.
- Questions in meetings already answered somewhere in the pack.
- Decisions later reversed on facts that were in the original paper.
- Unusual observations recorded as “other” or not at all.
Common mistakes
- Equating completeness with usefulness.
- Treating presentation as cosmetic.
- Relying on traffic lights without forecasts or consequences.
- Summarising away early warnings.
- Burying the decisive fact behind background.
- Leaving out inconvenient facts that would change the decision.
Frequently asked questions
How long should a monthly management report be? Long enough to support the decisions it needs to, and no longer. For many small businesses, a few front pages plus supporting detail on request is enough.
Who should write the summary? The person accountable for the recommendation. Delegating the summary often means the author never confronts whether the paper actually makes its case.
Are dashboards useful? Yes, for focusing attention on a few key measures. But they direct attention to what they show and away from what they leave out, so pair them with forecasts, exceptions and a channel for unusual observations.
How do we stop the unusual observations channel filling up with complaints? Define what belongs in it: things that seem wrong, new or unexplained and could matter. Acknowledge every entry, and redirect routine complaints to the right place.
What if the owner prefers reading everything? That is fine, as long as the most important information is also on the first pages, so that a shorter reading would reach the same decision.
Questions to ask
- Which decision does each part of our report support?
- Do our reports forecast or only describe?
- What would we see if we drew the links between our main initiatives?
- Where are early warnings being summarised away?
- Do people who raise concerns hear what happened?
- Is the most important fact in our last decision paper on the first page?
Bringing it together
A report is valuable only if it changes understanding or action. Build reports around decisions, ask for forecasts and confidence rather than colours, set clear escalation thresholds and show the relationships between projects, not just their individual status. Protect early warnings from being summarised away with a small direct channel and a reliable return path, and write decision papers so the key facts sit where readers actually look. A short report that changes a decision is worth more than a complete one that merely survives the meeting.
Source: KEVOS notes, drawing on C. P. Killen, “Evaluation of project interdependency visualizations through decision scenario experimentation”, International Journal of Project Management (2013), and teaching material on executive and portfolio reporting. Examples and figures in this article are illustrations.
