AtlasFree guide
What is a RAID log?
RAID stands for Risks, Assumptions, Issues and Dependencies. A RAID log is the single place a project keeps all four, so nothing quietly turns into a crisis.
6 minute read · no sign-up needed
What each letter means
The four categories look similar on paper and behave very differently in a real project. Getting them in the right column is most of the skill.
- Risk — something that has not happened yet but would hurt the project if it did. It has a likelihood and an impact, and it needs an owner and a mitigation.
- Assumption — something you are treating as true because you cannot confirm it yet. Assumptions are dangerous precisely because they feel like facts.
- Issue — something that has already happened and is affecting the project now. Issues need action and a date, not analysis.
- Dependency — something outside your control that your work waits on: another team, a supplier, a sign-off, a system.
The difference between a risk and an issue
A risk is in the future and conditional. An issue is in the present and real. "The supplier may miss the data migration date" is a risk. "The supplier missed the data migration date" is an issue.
Coordinators most often get marked down for logging issues as risks. It reads as though the project has not noticed the problem yet, which erodes a sponsor's confidence quickly.
How often to update it
A RAID log that is updated before a steering committee and ignored in between is theatre. Update it when something changes: after a stakeholder conversation, when a date moves, when an assumption is confirmed or broken.
Every entry needs an owner and a next date. An entry with neither is a note, not a control.
What good entries look like
Write entries a sponsor can act on without asking you a follow-up question. Name the thing, the effect, the owner and what happens next.
- Weak: "Staff training risk."
- Strong: "Ward staff may not complete training before go-live, which would slow triage in week one. Owner: Margaret Hughes. Mitigation: two extra evening sessions booked, attendance reviewed 14 March."
Worked example — entries from a hospital records rollout
| Type | Entry | Owner | Severity | Next step |
|---|---|---|---|---|
| Risk | Ward staff may not finish training before go-live, slowing triage in week one | Margaret Hughes | High | Two evening sessions booked; attendance reviewed 14 March |
| Assumption | The supplier will provide a test environment by the end of the planning phase | James Okoro | Medium | Confirm in writing at the next supplier call |
| Issue | Data migration test returned 4% record mismatch, above the 1% tolerance | James Okoro | High | Root-cause review this week; re-test before the stage gate |
| Dependency | Clinical sign-off on the new triage form must come from the clinical lead | Rachel Adeyemi | Medium | Slot requested in the clinical governance meeting |
Free RAID log template
Copy this into a spreadsheet or document. One row per entry, one entry per real thing — never a row that covers three problems at once.
ID
R1, A1, I1, D1 — so people can refer to an entry in a meeting
Type
Risk, Assumption, Issue or Dependency
Description
What it is and what it would do to the project
Owner
One named person, never a team
Likelihood / Severity
High, medium or low — be consistent
Mitigation or action
What is being done about it
Next review date
A real date, not "ongoing"
Status
Open, being managed, closed
Common questions
What does RAID stand for in project management?
Risks, Assumptions, Issues and Dependencies. A RAID log is a single register holding all four for one project.
Is a RAID log the same as a risk register?
No. A risk register covers risks only. A RAID log is wider: it also tracks the assumptions you are relying on, the problems already happening, and the things you are waiting on from other people.
Who owns the RAID log?
The project coordinator or project manager keeps it, but every individual entry has its own named owner. A log where everything is owned by the coordinator is a warning sign.
Practise keeping a RAID log on a live project
In the Atlas simulation you run a hospital records rollout. Risks arrive by email, assumptions break, and your steering committee asks about entries you logged. You keep the RAID log yourself and get feedback on it.
More free guides