RAID log best practice: Why your risk management is failing

By Michael Gold. Published

RAID log best practice for agencies and consultancies. How to drive accountability and replace heavy risk process with real visibility of delivery.

Why is your RAID log failing to manage project risk?

A RAID log fails when categories are confused, ownership is unclear and updates happen only before meetings. Separate risks, assumptions, issues and dependencies. Give each item a named owner and practical actions. Keep information visible and current, connect mitigations to delivery tasks and run a focused weekly review.

The core mechanics of a high-performing RAID log

Every delivery manager has inherited a RAID log that was little more than a graveyard for good intentions. It sits in a dusty spreadsheet, updated thirty minutes before a client steering committee, then immediately forgotten. This is a failure of execution, not of the tool.

To make a RAID log work, you must first enforce absolute clarity on what each letter means. In many professional services firms, these definitions blur:

* Risks are potential future events that will negatively impact your timeline, budget, or quality. They require mitigation plans before they occur. * Assumptions are factors we believe to be true for the project to succeed, but which lack concrete proof. Every assumption is a hidden risk waiting to be validated. * Issues are current, active problems that are actively impacting the project. They do not need mitigation plans, they need resolution plans. * Dependencies are external events, deliverables, or decisions that your team requires to move forward.

When delivery managers treat these four categories as interchangeable, the log loses its utility. An issue masquerading as a risk leads to passive monitoring instead of active escalation. An unvalidated assumption leads to sudden, avoidable crises.

The visibility trap: Why more process is not the answer

When projects begin to slip, the instinctive reaction of many COOs and delivery directors is to introduce more governance. They schedule weekly risk reviews, mandate new approval gates, and demand daily status updates.

This reaction misdiagnoses the root cause. Across the agencies and consultancies we embed into, the first finding is almost never a process gap. It is a visibility gap with process wrapped around it. Status meetings exist because nobody trusts the status, approval gates exist because nobody trusts the estimate. Fix the visibility and a surprising amount of process becomes unnecessary (Gold Strategic Delivery engagements).

The RAID log should be the primary vehicle for this visibility. When your RAID log is accurate, accessible, and updated in real time, it acts as a single source of truth. You do not need a status meeting to discuss a critical risk if the mitigation plan, owner, and target resolution date are already clearly documented and visible to everyone. The goal of RAID log best practice is to eliminate the admin that stops delivery people from doing high-value work, not to add another layer of bureaucratic reporting.

Practical rules for active risk ownership

A RAID log is only as valuable as the accountability behind it. If every risk is owned by the delivery manager, the log is failing. To establish genuine ownership, you must apply three non-negotiable rules.

First, assign single-point accountability. A risk cannot be owned by "the dev team" or "the client". It must be owned by a specific individual who has the authority to execute the mitigation plan. If a client dependency is slipping, the owner should be the client stakeholder responsible for that deliverable, not your internal project coordinator.

Second, define actionable mitigations. A mitigation plan is not a statement of hope. Writing "ensure client delivers feedback on time" is not a mitigation. A real mitigation plan defines the specific actions you will take to reduce the likelihood or impact of the risk. For example, scheduling a daily fifteen-minute review slot with the client sponsor to unblock pending design decisions.

Third, integrate the RAID log into your existing delivery cycle. If you run a managed PMO for agencies, you know that risk management cannot be a separate activity. It must be woven into your weekly standups, sprint planning, and client retrospectives.

The trade-offs of different tooling options

There is no single perfect tool for managing a RAID log, but there are clear trade-offs depending on your delivery environment.

Spreadsheets are highly flexible and require zero training, making them excellent for client-facing communication. However, they lack automated tracking, version control, and integration with daily delivery tasks. They quickly become outdated if not rigorously policed.

Project management platforms like Jira, Asana, or Monday.com allow you to link risks directly to tasks and epics. This keeps risk management close to where the work is actually happening. The trade-off is that these tools can become overly complex for external clients or senior executives who require a high-level summary rather than a granular task list.

For most scale-ups and agencies, a hybrid approach works best. Maintain a highly visible, simplified version of the RAID log in a shared document or client portal for executive steering, whilst linking the underlying mitigations directly to developer tasks in your primary delivery tool.

How to run a 15-minute weekly RAID review

To keep your RAID log alive, you do not need hours of administrative meetings. You need a highly structured, fifteen-minute weekly review. This is not a session to read through every line of the document, but a targeted intervention focused on exception handling.

Spend the first five minutes on new risks and issues. Review anything added in the last week, assign owners immediately, and agree on initial mitigation steps.

Spend the next five minutes on critical path dependencies. Focus specifically on dependencies with a target date in the next two weeks. If a dependency is slipping, escalate it immediately.

Spend the final five minutes on stale items. Identify any risks or assumptions that have not been updated in the last fortnight. If they are no longer relevant, close them. If they are still active, challenge the owner on why progress has stalled.

This tight cadence prevents the log from becoming bloated and ensures that your delivery team remains focused on the highest-priority blockers.

An effective RAID log is not a tool for avoiding all project challenges. Projects are complex, and unexpected issues will always arise. Instead, the RAID log is a mechanism for professionalised delivery. It signals to your clients, your team, and your leadership that you are in control of the variables you can influence. By focusing on visibility over process, you remove the administrative friction that bogs down project delivery and clear the path for high-value work.

Common questions

Who should own a risk in a RAID log?

Assign each risk to one named person with the authority to carry out its mitigation plan. Do not assign it to a whole team. For a slipping client dependency, ownership should sit with the client stakeholder responsible for that deliverable.

How should you run a weekly RAID review?

Use fifteen minutes for exceptions rather than reading every entry. Spend five minutes on new risks and issues, five on critical dependencies due within two weeks, and five on stale items. Close irrelevant entries and challenge owners where active items have stalled.