{{ summaryTitle }}
{{ summaryValue }}

{{ summaryLine }}

Paste five columns. Separate multiple roles with semicolons.
Name the workstream or recurring process this matrix governs.
Choose the operating context used to review every activity row.
{{ activeImport.help }}
Describe whether assignments name stable functions, individual DRIs, or external parties.
The default requires one Accountable and at least one Responsible role.
Tailor review guidance for a working team, sponsor, evidence packet, or onboarding brief.
{{ fileStatus || 'Files stay in this browser and must be 256 KB or smaller.' }}
ActivityResponsibleAccountableConsultedInformedStatusCopy
{{ row.activity }}{{ roleText(row.responsible) }}{{ roleText(row.accountable) }}{{ roleText(row.consulted) }}{{ roleText(row.informed) }}{{ row.status }}
ActivityStatusAccountable checkResponsible checkCommunicationNext actionCopy
{{ row.activity }}{{ row.status }}{{ row.accountable_check }}{{ row.responsible_check }}{{ row.communication_check }}{{ row.next_action }}
RoleRACITotalSignalCopy
{{ row.role }}{{ row.responsible }}{{ row.accountable }}{{ row.consulted }}{{ row.informed }}{{ row.total }}{{ row.signal }}

The chart renderer is unavailable. The same values remain in the role load ledger.

Risk Register Generator Build a risk register from pasted rows or CSV with inherent and residual scores, owner-gap priorities, treatment cues, and review dates.
Checklist and Log Generator Turn one-item-per-line notes into Markdown checklists or logs while keeping source order and duplicates and adding optional context when needed.
Acceptance Criteria Generator Draft testable acceptance criteria from behaviors and edge cases, then check Gherkin or QA handoff wording for missing coverage and vague terms.
User Story Generator Draft a user story with a clear role, goal and benefit, then turn acceptance checks into structured criteria with readiness or splitting warnings.

Work often stalls at the point where “the team” is expected to act. A RACI matrix replaces that vague handoff with an activity-by-activity agreement about who does the work, who owns the decision, who gives input, and who needs the outcome communicated.

RACI stands for Responsible, Accountable, Consulted, and Informed. Responsible roles carry out the activity. The Accountable role owns the final decision or outcome. Consulted roles provide two-way input before the decision or work is complete. Informed roles receive updates without becoming another approval gate.

CodeQuestionCommon failure
RWho performs the work?No doer is named, or too many roles assume someone else will act.
AWho is answerable for the outcome?No owner exists, or several owners can veto one another.
CWhose input is needed before completion?Consultation grows into an informal approval committee.
IWho needs the decision or result communicated?Recipients are pulled into meetings instead of receiving a clear update.

The row is the important unit. One role may be Accountable for one activity and merely Consulted on another. A matrix that assigns broad departments everywhere can look complete while hiding the person or stable function that can actually make the decision. Conversely, naming individuals can become stale when people change roles, so the vocabulary should match how long the matrix must remain useful.

RACI is a conversation aid, not an automatic organization chart. It can expose missing ownership, overloaded decision makers, and excessive consultation, but it cannot decide whether the underlying process is sensible. The people named in the matrix still need to agree with the assignments, timing, authority, and escalation path.

How to Use This Tool:

Choose the review policy first so the same assignments are judged against the intended operating context.

  1. Name the Project or process, choose a Matrix profile, and set the role vocabulary, accountability strictness, and stakeholder audience. Together these choices define how rows and role loads are reviewed.
  2. Select the Import/template mode. Standard rows use Activity, Responsible, Accountable, Consulted, Informed. Role-code rows repeat an activity for separate role/code assignments. Handoff rows put the accountable owner before delivery roles.
  3. Paste or import the activity rows. Separate multiple roles with semicolons; tab, pipe, and CSV rows are accepted, and recognizable headers are skipped.
  4. Read Assignment checks row by row, then use Role load ledger to find concentrated decision, execution, or consultation work. Share the matrix only after the named roles confirm it.

Interpreting Results:

Fix ownership or Audit exception means at least one blocking rule failed. Review policy means the row is structurally usable but contains a choice worth confirming, such as shared accountability, too many Consulted roles, an R/A overlap, or a broad accountable role. Ready means the selected policy found neither type of issue.

Role-load signals are prompts for discussion, not workload measurements. They count assignments, not effort, duration, authority, or availability. A person accountable for one high-risk decision may carry more real load than a role responsible for several routine activities.

Parser warnings also deserve review. A short row can leave assignments blank, a duplicate activity can create competing versions, and an incomplete role-code row may be ignored. Compare the normalized assignment matrix with the original source before treating the counts as complete.

Technical Details:

RACI evaluation combines row-level ownership rules with matrix-wide role-load thresholds. Import modes normalize different column layouts into the same R, A, C, and I arrays, then case-insensitive duplicate removal keeps each role once per code and activity.

Transformation Core

Import modeSource columnsTransformation
Standard rowsActivity, Responsible, Accountable, Consulted, InformedOne source row becomes one activity.
Role-code rowsActivity, Role, RACI codeRepeated activity names are grouped; aliases such as owner, approver, reviewer, and audience map to A, C, or I.
Handoff rowsActivity, Accountable owner, Delivery roles, Reviewers, AudienceOwner maps to A, delivery to R, reviewers to C, and audience to I.

Formula Core

Decision and execution load thresholds grow with the number of activities while respecting profile minimums and vocabulary or strictness adjustments.

D=max(1,Dmin+Δvocabulary+Δstrictness,⌈nrdecision⌉) E=max(1,Emin+Δvocabulary,⌈nrexecution⌉)

n is activity count, D is the Accountable-assignment threshold, and E is the Responsible-assignment threshold. The selected profile supplies minimums and ratios:

ProfileConsulted limitDecision thresholdExecution thresholdExtra row rules
Project delivery3max(3, ceiling of 55% of activities)max(3, ceiling of 60%)None
Product launch4max(3, ceiling of 50%)max(3, ceiling of 55%)At least one Informed role
Operations runbook2max(2, ceiling of 45%)max(3, ceiling of 50%)None
Regulated change4max(2, ceiling of 45%)max(2, ceiling of 50%)At least one Consulted and Informed role; review R/A overlap

Named-people vocabulary lowers both thresholds by one; DRI-plus-team vocabulary lowers only the decision threshold by one. Both also flag broad accountable roles. Audit-strict mode lowers the decision threshold by one more.

Rule Core

ConditionDefault effectPolicy variation
No Accountable roleFixWorkshop draft treats it as Review.
More than one Accountable roleFixShared-warning and workshop modes treat it as Review.
No Responsible roleFixNo relaxation.
Consulted count above profile limitReviewSuggests trimming roles or splitting the activity.
R or A role also appears in C or IReviewAudit-strict mode makes the overlap a Fix.
Required C or I role missingReviewDepends on profile.

Role-load signals use an ordered rule. Accountable assignments at or above D produce Decision load. Otherwise Responsible assignments at or above E produce Execution load. Otherwise a role with at least two Consulted assignments and more C assignments than R plus A is Consult-heavy. An all-Informed role is an Observer; a role with one total assignment is Light touch; remaining roles are Balanced.

Source text is limited to 65,535 characters. Empty activity names and incomplete role-code rows are ignored with warnings; a result requires at least one parsed activity, and activities without assignments return ownership fixes.

Worked Examples:

Launch approval with one clear owner

“Approve launch plan” assigns Delivery as Responsible, Product Lead as Accountable, Security as Consulted, and Sponsor as Informed. Under the product-launch profile and single-accountable policy, the row is Ready because it has one A, at least one R, an informed audience, and no conflicting role overlap.

Change lead holds R and A

In a regulated-change profile, the same Change Lead is both Responsible and Accountable. Audit-strict mode records that R/A overlap for policy review while still distinguishing it from missing ownership. The reviewer can then decide whether execution and approval must be separated for that control.

References: