Build a RAID Log Your Team Will Actually Keep Current
Most teams own a RAID log. Far fewer keep one that reflects reality. The document is created during planning, looks reassuring in the kickoff deck, and then quietly goes stale while the project drifts. Entering 2021, with many teams still working remotely or hybrid and supply disruptions far from settled, a current RAID log is one of the cheapest forms of insurance a project manager can buy.
RAID stands for Risks, Assumptions, Issues, and Dependencies. The four categories matter because they catch different things. A risk has not happened yet; an issue already has. An assumption is something you have accepted as true without proof. A dependency is something outside your direct control that your plan relies on. Mixing them together is the first reason logs go unused: nobody can tell what they are looking at.
Set up the four columns properly
Keep the structure light enough that an update takes under a minute, but rich enough to drive a decision. For each entry, capture who raised it, the date, a plain-language description, an owner, and a status. For risks specifically, add likelihood, impact, and the response you have chosen.
Risks. Things that might happen and would hurt. Record likelihood, impact, and a response: avoid, reduce, transfer, or accept. A risk with no named response is just a worry.
Assumptions. Beliefs you are planning around — a permit will arrive by March, a vendor can still ship by sea. Note how you will confirm each one, because assumptions that fail quietly become your worst issues.
Issues. Problems happening now and needing action. Track the owner and a target resolution date, not just a description.
Dependencies. Hand-offs you rely on from another team, supplier, or approval body. Note the direction and the date you need it by.
Make updating it a habit
A log goes stale because reviewing it has no fixed home. Give it one. Put a five-minute RAID review at the top of your weekly status meeting and walk only the entries that changed or are overdue. When a remote team cannot rely on hallway conversations to surface problems, this standing slot becomes the place concerns actually get raised.
Assign every entry a single named owner — never a team or a role.
Close entries loudly: move resolved issues to a 'closed' section rather than deleting them, so the history stays auditable.
Re-test assumptions on a schedule; an assumption you set in October may be false by January.
Keep one source of truth and link to it everywhere, so nobody works from a stale copy.
The discipline matters more than the tool. A shared spreadsheet that gets reviewed every week beats a polished risk platform nobody opens. The goal is a living record that lets you see trouble early — while you still have cheap options to respond.
If you want help standing up project governance that your team will actually maintain, XNM's program & project delivery advisory can help you build it and embed the habit.