Find out where AI actually fits in your firm before you buy anything
Most environmental firms are already using AI somewhere: a few people, a few tools, no shared standard. The question leadership is really asking isn’t “should we use AI?” but “where, first, and how do we know?” The Workflow Diagnostic answers that with evidence from your own work rather than a vendor’s demo.
What it is
A remote, session-based review of how your teams actually produce their work: the reports, permits, proposals, data pulls and QA loops that consume senior time. We sit with the people who do the work, trace the repeated friction, and separate the workflows where AI would move the needle from the ones where it would add risk for little gain. The output is a written set of findings, a ranked list of opportunities, and an executive readout your leadership can act on.
What you walk away with
- Clarity
Clarity on where time is lost
Which workflows, which steps, which roles. Usually a handful of patterns account for most of the recoverable hours.
- Ranked
A ranked opportunity list
Each opportunity scored on value, risk, effort and fit with your existing systems, so the first move is the right one.
- Defensible
A defensibility read
Where AI can be used inside client boundaries and professional-review requirements, and where it can’t. For firms doing Phase I ESAs this is the part that matters most; the standard we apply is written up in Defensible AI for Phase I ESAs.
- Decision
An executive readout
A leadership session that turns the findings into a decision: what to do first, what to skip, and what it would take.
Who it’s for
Environmental consulting firms, from a single-office practice to multi-office regional and national groups, where leadership is interested in AI but nobody owns it yet, or where tools have been bought and adoption has stalled. It fits before a training program, before a build, and before a platform decision. If you already know the one workflow you want fixed, you may not need it; say so on the call and we’ll tell you.
What it isn’t
It isn’t a tool recommendation exercise, a security audit, or a seminar. We don’t sell software in it, including our own. And it doesn’t require anyone to change how they work during the review.
Where it leads
Some firms stop at the readout with a clear plan they run themselves. Others go on to AI training and enablement for the teams the diagnostic identified, or to a targeted build for the one workflow that justifies it. Where the finding is that licensed tools sit unused, Copilot training is the third path. The diagnostic is designed so that any of them is grounded in evidence.
Examples of what a diagnostic surfaces
At an air-quality consultancy, the review found three workflows worth building around: trending continuous-monitoring data that analysts were assembling by hand, a first-pass safety data sheet review, and generating permit-compliance task lists from permit text. Each was a distinct, bounded piece of work with a clear reviewer, which is exactly the profile that pays back.
Trusted by environmental firms
Questions firms ask first
How long does an AI workflow diagnostic take?
A few weeks of scheduled remote sessions with the people who do the work, followed by synthesis and an executive readout. Scope is set on the first call.
Do our staff have to prepare anything?
No. The sessions work from how people already do their jobs. The only ask is time from the right people.
Is our client data at risk?
No client data is uploaded to any AI tool during the diagnostic. We review how work is done, not the work product itself, and we operate inside your existing confidentiality agreements.
What if the answer is “AI doesn’t help much here”?
Then that’s the finding, and it’s worth knowing before you spend on tools or training. The readout says so plainly.
Do you recommend specific tools?
Only where an opportunity genuinely needs one, and never our own software inside a diagnostic. Most findings are about workflow and review structure, not products.