A RAID log is the most pragmatic project artefact ever invented. Instead of maintaining a risk register, an issue log, an assumptions doc, and a dependencies tracker in four different places that no one keeps in sync, you maintain one table. Each row gets a type — Risk, Assumption, Issue, or Dependency — and the rest of the columns are shared. One source of truth, one weekly review, one export to your status report.
This free RAID log template is built to be useful in week one of a project and still useful in week thirty. Edit any cell in your browser. Add or remove rows in a click. Type-tag each entry. Filter by status mentally (or download to CSV and filter in Sheets). Persists in your browser so you don't lose work between sessions.
What each type means — and where teams usually mess this up:
- Risk — something that *might* happen and would have negative impact. If it's already happening, it's an Issue, not a Risk. This distinction matters because risks need mitigation; issues need resolution.
- Assumption — something you're treating as true to plan around, but haven't verified. "We assume the client will provide content by week 4." When an assumption breaks, it usually becomes an issue overnight. Mark assumptions and revisit them weekly.
- Issue — something that *is* happening right now and needs resolution. Owner + due date are non-negotiable. An issue without an owner is wishful thinking.
- Dependency — something outside your team's control that you need (or that needs you). External vendor delivery, another team's API, a legal sign-off. Track who you're waiting on and when you need it.
The seeded rows show one of each so you see the shape. Replace with your own.
The columns are intentionally minimal: Type, Title, Owner, Status, Due, Notes. We don't ask for probability and impact here (use the dedicated risk register template for that). We don't ask for category. We don't ask for fifteen RACI fields. A RAID log lives or dies by how often you actually update it — and the more fields you add, the less often anyone touches it.
Status moves through Open → In Progress → Blocked → Resolved → Closed. Anything in Blocked for more than a week needs to escalate. Anything Open without an owner gets one assigned in the weekly review or gets deleted — those are your only two options.
If you want this connected to actual tasks (where resolving an issue auto-creates work for the owner with the due date already set), that's what LemTask does. For a standalone RAID log you can keep in a tab and share via CSV, this is enough.