Needs the policy beside the work
Reviews valued a shared source of truth, clause search, version history and workflows that replace scattered email threads.
Accord
Independent B2B concept
A traceable path when contract policy and business pressure disagree.
I designed a desktop concept that helps reviewers compare a clause with company policy, prepare an exception and record the final decision.
Why this problem
Once a clause breaks policy, someone has to justify the exception, find the right approver and leave a usable record.
I narrowed the concept to clauses that break company policy. The reviewer needs the clause beside the playbook and any useful precedent. They also need the business rationale and the right approver.
Context and role
I started broad, then chose one contract decision.
As the solo product designer, I owned the desk research, problem framing, workflow, interface and prototype. The first idea covered AI and contracts in general. Research narrowed it to policy exceptions.
Commercial contract reviewersPublic reviews and product docsPractitioner interviews next
Research 01 · Public practitioner evidence
The recurring complaints were lost context and unclear ownership.
I read a sample of recent practitioner reviews across Ironclad, Juro and SpotDraft. Their review pages contained 664 published reviews at the time. The sample helped me find recurring workflow needs. It does not estimate prevalence.
Needs status and clear requirements
Visibility reduced repeated messages between legal and commercial teams. Structured intake made missing context visible earlier.
Needs evidence and a durable record
Approval routing, activity logs and clear ownership mattered because the decision had to remain traceable after signature.
301 Ironclad reviews182 Juro reviews181 SpotDraft reviewsPublic review sample
Research 02 · Product benchmark
I compared how three tools handle policy, approvals and audit history.
Across the three products, the common pieces were a playbook, a named approver and an audit trail. I built the prototype around that handoff.
Playbook-led review
- Flags non-standard terms against configured positions.
- Supports fallback language, exceptions and clause-level approvers.
Keep policy and the clause connected. Make the approver explicit.
Ironclad documentation ↗Recommended next action
- Reviews comments and redlines against a selected playbook.
- Keeps the playbook available beside the AI recommendation.
Expose the basis of the recommendation and keep acceptance human.
Juro product update, March 2026 ↗Conditional approvals
- Routes review or approval through explicit conditions.
- Resets decisions when relevant contract versions or values change.
Make status and re-review visible throughout the workflow.
SpotDraft documentation, January 2026 ↗Public product researchPractitioner sessions plannedPrototype built for those sessions
Problem framing
What I knew, what I assumed, and what still needs an interview.
Desk research showed that the clause and policy had to remain together. The workflow also needed a named owner and a durable record. The exact role split and handoffs still need practitioner interviews.
Seen in desk research
- Reviews use clauses, playbooks, owners and approval status.
- Exceptions need a rationale and decision record.
Design assumptions
- Policy and precedent are useful beside the proposed clause.
- A structured request may reduce clarification.
Still unknown
- How often exceptions create delays.
- Who owns each step across organisations.
Before high fidelity
I mapped the handoff before designing the interface.
The early frames checked the sequence and information order. They also exposed missing evidence and reviewer disagreement as product states.
Exploration
Three passes shaped the workflow.
I moved to high fidelity only after the workflow exposed the actors, information and failure states the interface needed to support.
Who hands what to whom?
A simple flow separated reviewer, approver and policy owner, then identified the decision record as the final output.
What must be visible together?
The clause sits beside policy and precedent. The reviewer adds the rationale and proposed fallback before requesting approval.
When should the system stop?
Weak evidence, missing context and reviewer disagreement route the case to stronger evidence or a named person.
Solution
Four decisions shape the exception workflow.
Decision 1
Start with the work that needs attention.
The queue shows deviation, risk, owner, due date and status before a contract is opened.
Tradeoff
Each row has one primary risk label. Metadata stays neutral and orange is reserved for the deviation that needs action.
Decision 2
Keep policy beside the clause.
Proposed wording and company guidance share one working view.
Tradeoff
The three-column desktop layout would need a dedicated focus mode on narrower screens.
Decision 3
Structure the exception before approval.
The request captures why the exception matters, the proposed fallback and who needs to decide. Missing context is visible early.
Tradeoff
Required fields make requests comparable. Free text leaves room for cases the form did not anticipate.
Decision 4
Make the decision and its record explicit.
The approver sees the relevant context before acting. The final decision is written to an audit timeline.
AI boundary
AI summarises sourced material and suggests a next step. A named person approves, and weak evidence pauses the recommendation.
Edge cases
Three conditions pause the recommendation.
Weak evidence, missing context and reviewer disagreement each require a clear next step.
Evidence gap
Weak sources pause the recommendation.
The reviewer is directed to stronger sources or the policy owner.
Feedback stays attached.
The approver returns the request with the missing context.
Both positions remain visible.
A named policy owner receives the disputed evidence before deciding.
Workflow outcome
The audit trail keeps the decision and its rationale together.
The final screen keeps the rationale, evidence and approver in one record.
Validation plan
Start with the last exception they handled.
The concept has not been tested with Legal Ops practitioners. I would run six sessions. Each person would reconstruct a recent exception, then complete the same handoff in Accord.
Prepare the exception
Commercial counsel or contract managers compare the clause, policy and supporting evidence.
Supply the context
Sales, procurement or operations partners explain the business need and follow the decision.
Decide and record
Legal leaders or policy owners approve, reject or return the exception.
Reconstruct the real handoff
Identify the trigger, missing information, waiting points, people involved and record left after the decision.
Complete one connected case
Find the exception, compare the evidence, prepare the request, decide and recover the final rationale.
Planned research6 practitioner sessionsNo usability result claimed yet
Next step
Put it in front of Legal Ops.
The first sessions would test the roles, required context and decision record before any impact claim.