We mean seams.
Not seamless.
Every business is a set of handoffs: a lead becomes a client, an order becomes revenue, hours become margin, an invoice becomes cash. Those handoffs are the seams. They are where a company creates value and where it quietly loses it. We find them, measure them, and decide, on purpose, which ones to automate and which ones to keep a person in.
Why the books are the last place a problem shows up
When an owner says "my numbers are wrong," they are usually right, and the bookkeeper is usually not the reason. The numbers are wrong because something upstream is wrong: nobody owns the handoff between spending money and recording it, revenue is booked by channel when decisions are made by product, inventory lives in the building but not on the balance sheet, or the marketing system and the billing system have never been introduced. The ledger just reports the damage.
Most accounting firms work from the ledger down. They take the transactions as given, categorize them, and produce a report. That is fine when the systems feeding the ledger are sound. In most owner-led companies they are not, and no amount of careful categorizing fixes a number that arrived broken.
We work from the source up. Orders, shipments, payouts, timesheets, CRM records, bank feeds. We rebuild what the business actually did, compare it to what the books say, and the gap between the two is the list of things to fix.
Where the name comes from
In 3D modeling, seams are the lines that let a flat image wrap around a solid object. Without them there is no shape. A company is the same: subsystems for selling, delivering, billing, paying people and paying taxes, joined at seams where one process hands off to the next. The seams are the structure.
Software vendors promise to make those handoffs seamless, and sometimes a single connection genuinely is. But an organization as a whole never runs seamlessly, and pretending it does creates a black box: information goes in, work product comes out, and nobody can explain the middle. Large companies have controls to keep that box open. Owner-led companies usually don't, and those are the companies we work with.
Seamless isn't the goal. A process that does no harm to the health of the whole system is the goal. Sometimes that process has a person in it.
So the question we ask about every handoff isn't "can this be automated?" It's "what does it cost if this is wrong, and who would notice?" We use automation and AI, quite a bit. We also think the ability to walk over and ask a coworker a question is a feature, not a workflow failure. When we say something is seamless, we'll mean it. Mostly we mean seams.
How an engagement runs
- 01
Intro call 15 minutes
You tell us what is broken, or what you suspect is. We tell you honestly whether we are the right people to fix it, and what we would look at first if we are.
- 02
Diagnostic 8–13 weeks
We rebuild the numbers from source data rather than from your reports: orders, shipments, payouts, CRM and phone records, the full ledger. Then we trace how money moves from first contact to cash and talk to the people who touch it along the way. You get a written findings report where every figure states its source, a ranked fix list, and a model you keep.
- 03
Build 4–12 weeks
We fix the seams in priority order: a chart of accounts designed around your decisions, written controls, revenue and cost models, bill-pay and approval workflows, and the connections between systems that never talked to each other. Some of this is software we build. Most of it is defining who owns each handoff, what finished looks like, and what happens when two numbers disagree.
- 04
Run Ongoing
The Fractional CFO engagement. This is not bookkeeping with a better title. It is the forward view of the business and the discipline that keeps it accurate: cash-flow projections, rolling forecasts and budgets, monthly tax planning, and a standing meeting where the owner approves spending with current numbers in front of them. How often we are in the books, weekly or monthly, depends on the business and the season. Each month we close the loop: what changed, what the numbers say now, and what to do about it, so pricing, hiring, capacity and the tax position get decided rather than discovered. Accountant of record for the whole function.
What that looked like for one client
A direct-to-consumer food manufacturer came to us with four years of books showing 58% margins and a 2025 loss. The prior accountants hadn't just missed the problem; they had built it. Inventory was accounted for incorrectly, the books were assembled after the fact, projections were run on top of them, nothing was reviewed, and the owner was getting no guidance on any of it. Nobody had ever looked at an order.
Diagnostic. Thirteen weeks. 450,000 orders, 56,000 shipments, product costing for nine lines, every ledger account. COGS had been split equally across products, inventory on the balance sheet was a third of what sat in the building, $111K of processing fees were missing and $1.4M of discounts were booked as revenue. Actual margin: about 33%. Actual result: profitable.
Build. A Financial Controls Framework the company now cites by section number. A chart of accounts with revenue by product type and COGS by category. Nine inventory accounts from raw materials to finished goods, relieved weekly at standard cost and trued up to monthly counts, with written thresholds for when a variance means recount. A revenue model that books each Shopify payout as one balanced entry matching one bank deposit. Centralized bill pay with a weekly approval call.
Run. Weekly gross margin by product. Rolling cash projections through peak season with the credit line planned instead of panicked. A daily check-in when the orders surge. A packaging supplier switch worth ~$51K a year and a shipping model that turned a loss into a steady profit, both found because the numbers were finally true enough to act on.
Principles
Friction on purpose.
Some handoffs should stay a little manual, because a person looking at a thing before it goes out is cheaper than a black box that is wrong. We automate aggressively where it is safe and keep humans where judgment matters.
Numbers you can act on.
A P&L three weeks after month-end is a historical document. Our reporting is built around the two or three decisions you actually have to make this month, and it shows up in time to make them.
Tax is planned, not discovered.
If the operations are right, the books are right, and if the books are right, the tax position is known in October, not April. Planning is part of every engagement for that reason.
Only clients we can fully support.
We cap the roster. Being extremely busy does not create meaningful work for anyone. If we take you on, you get the whole practice, not a slice of it.
Want to see your own seams?
The Diagnostic is the fastest way to find out where the money is leaking and what it would take to stop it. It starts with a 15-minute call.