Three engagements, one sequence.

Send us three to five high-stakes communications: the customer letter, the policy announcement, the benefits notice, the form, the onboarding screen, the chatbot transcript. We audit them against published plain-language and accessibility standards, against one another, and against the commitments established by your leaders and company values. Then we trace each failure upstream to the policy, the template, or the decision that produced it.

You get:

1. Findings ranked by severity, paired with the relevant standard
2. The likely operational cost of leaving each failure unchanged
3. Rewrites of the highest-severity passages
4. A working session with the people who own these communications
5. The operational measures to monitor afterward, so you can tell whether the revisions worked

The audit is the entry point for every organization we work with. It is scoped so a company can buy it without committing to anything else, and it tells both of us whether a larger engagement is warranted.

Request a Voice Audit

For organizations where the audit finds a systemic problem: different teams producing different versions of the company's voice, with no shared standard and no one accountable for the result. We build the standards, ownership, and testing that carry leadership's intent into every communication.

What we build:

1. The Voice Governance Standard: operating rules, guidance for recurring situations, shared terminology, decision rights, and escalation paths
2. An ownership map: who owns the voice, who approves its use, and who intervenes when it fails
3. A review and escalation workflow that fits how your teams already work
4. Templates for the communication types that recur, built from the governed versions
5. A sampling and testing protocol, so the standard is checked against what actually goes out
6. A pilot on one live workflow, so the system is proven before it is extended

We build it with the people who will own it, and we train them to run it, so the voice belongs to the organization.

For organizations using AI to draft, adapt, or deliver consequential communications. We define and test what your AI systems are permitted to say. Your team or vendors build and run the systems.

What we build:

1. The permission boundary: what the system may answer, what it must send to a person, and what it says when it does not know
2. Test cases and acceptance criteria for each kind of answer the system is allowed to give
3. A sampling protocol for live output, so failures are found by you before they are found by a customer
4. An incident review and correction process, with the record that lets you show what the system was allowed to say and how you checked it

This engagement can follow an audit, or stand on its own when the AI system is the one doing the communicating.