Process improvement built around how your team works.
Write down the standards, capture what your senior people know, and train the next ones — then bring in AI where it actually helps. Start with a scoped assessment; add the rest one phase at a time.
Book a 30-minute consultHow it worksFirms ready to improve how they work
Firms where the knowledge that matters is field judgment, review habits, and twenty years of "how we do it here" — exactly what turnover, growth, or a careless AI rollout tends to flatten. The first question is usually what's only in someone's head.
Owners who want the way the firm works written down, taught, and owned — including clear rules for AI — without hiring a full-time role to get there. The first question is usually where to start.
Start with a scoped assessment
Every engagement begins with a written assessment of your actual workflows. Standards, training, and AI work are added where they help, one phase at a time.
Process & workflow assessment
A working session with the people who own the workflows, then a written assessment covering:
- Selected workflows, and where their knowledge actually lives
- Undocumented steps, gaps, and data constraints
- Candidate improvements, each with an owner
- The review point that makes each change safe to try
- A prioritized next step
Internal standards & documentation
Standard operating procedures, QA/QC procedures, review checklists, and a standards manual your people will actually open — written from how the work is really done, and kept current as it changes.
CAD-level standardsLayer conventions and production checklists are produced under Deliverable Support.
Knowledge management
Capture what your senior people know before it walks out the door, and organize the project files, record research, and past work your team keeps re-finding — so the answer lives somewhere other than one person's memory.
Staff training & development
Onboarding paths and development plans that move people from technician to party chief to licensure, built on your own standards. Includes role-based sessions on using AI tools well: what they're good at, where they fail, and how to check the output.
AI implementation guidance
Which workflows are worth piloting, which tools fit, and the data rules and human review point that make each pilot safe to try. Build/buy-neutral, with a second opinion when a vendor pitch needs one.
AI committee & Center of Excellence
Charter, membership, decision rights, meeting cadence, and the policies a committee actually needs: acceptable use, data handling, vendor review, disclosure. Then ongoing facilitation so the committee makes decisions instead of collecting them. Useful once there is something to govern — not a prerequisite for starting.
Fit call first. Then the work, one phase at a time.
Fit call — 30 minutes, free
The consult you book on this site. We talk through where the work depends on one person's memory, what's worrying leadership, and whether an assessment makes sense. No deliverable, no pitch.
Working session & written assessment
A scoped, paid engagement: a working session with the people who own the workflows and the risk, then the written assessment above — what to capture, what to pilot, what to leave alone, and what to decide first. Build/buy-neutral, and written so it can be checked later.
The work itself, one phase at a time
Standards and documentation, training blocks as teams come online, committee facilitation on a set cadence, and a second opinion when a vendor pitch needs one. Each is scoped separately.
A map, not a memo
A workflow readiness map from a sample assessment
Four workflows, where their knowledge lives, who owns them, whether each is a pilot candidate, and the human review point that makes it safe to try. One becomes the first pilot; one stays off the table because it's a professional determination.
Illustrative sample for a fictional firm. Not client work.
Questions that come up on the fit call
Do we have to be using AI for this to be useful?
No. Most engagements start with standards, documentation, or training — the groundwork that makes any later change stick. AI comes in when there's a workflow it actually helps, and not before.
Do we need an AI committee to start?
No. Most firms start with the assessment and one small pilot. A committee is worth forming once there are decisions to make on a regular basis — I'll say so when that point arrives, and not before.
Do you work with the tools we already have?
Yes. The assessment starts from your existing systems and the tools people are already using. Recommendations are build/buy-neutral.
Who from our team needs to be involved?
For the assessment, the people who actually own the workflows and the risk — usually a project or survey manager, a senior technician, and whoever answers for data handling. A few hours each, not weeks.
How do we know it's working?
Each change — a new procedure, a training path, an AI pilot — is written up with a baseline, a review point, and what "better" would look like before it starts. It's evaluated against that, not against a vendor's claims.
Do you sell software?
No. Recommendations start with your needs and existing systems. If a custom tool turns out to be the right answer, it is scoped separately under App Development — never assumed.
Start with the conversation.
In 30 minutes we'll cover where the work depends on undocumented know-how, what's keeping leadership up at night, and whether an assessment is the right next step.
Book a consult