What drives me
I think the numbers are trying to tell you something.
Most of what I do for a living comes down to one belief: a set of books that has been reconciled properly will tell you the truth about a business, and a set that has not will tell you whatever you already wanted to hear. Seventeen years in, I have not found an exception.
Where this started
I spent my career inside group-practice healthcare finance — billing operations first, then revenue cycle, then controllership. Multi-entity consolidations, a close calendar that had to land, and a month-end that did not care how I felt about it. Seventeen years of that teaches you something specific: almost nothing in a set of books is wrong because somebody was careless. It is wrong because the process had a gap, and the gap had nobody's name on it.
The moment that shaped how I work was realising that a claim can be underpaid and look paid from every seat in the organisation. Nobody was asleep. The process worked exactly as written. The money left anyway, a few dollars at a time, in a bucket labelled "adjustments" that nobody was ever asked to explain. I have been chasing that class of problem ever since — the loss that is invisible precisely because the process is functioning.
Why I write software
Because at some point I was reconciling multi-entity cash by hand in spreadsheets with tens of thousands of rows, and it became obvious that the matching logic in my head was a program I had not written down yet. So I wrote it down. That became TrueBalance, a three-pass matching engine — exact, subset-sum, then fuzzy — that does the part of a reconciliation that is mechanical, and leaves me the part that requires judgement.
I co-founded JMWINC Developers on the same principle, and most of what we build comes from the same place: I hit a problem doing the actual work, got tired of working around it, and built the thing. That is a different starting point from most finance software, which is designed by people who have read about the problem.
When I say a reconciliation should be reproducible by someone else from the worksheet, it is not a stylistic preference. It is because I have had to build the tooling that re-performs it, and that only works if the method is explicit.
What I am actually like to work with
I will tell you what I found
Including when the answer is unflattering, and including when the problem turns out to be something I built. The value of this work is entirely in its credibility; the first time I shade a number to be easier to hear, I have nothing to sell.
I finish things
I would rather stay with a problem until it is closed than carry it into next week on a list. Most reconciliation debt exists because somebody meant to come back to it.
I document so you are not dependent on me
The method gets written down, the schedules get templated, and your team can run them. If the work only functions while I am in the building, I have built you a liability, not a system.
I will say when I am not the answer
If you need a bookkeeper, an audit, or a CPA, I will tell you on the first call rather than three months in. I am not a CPA and JMW-Inc is not a CPA firm.
Still learning the formal part
I am working through a BS in Finance and Technology at the University of Phoenix alongside the day job — not because seventeen years of practice needed validating, but because I want the theory underneath the instincts. There is a difference between knowing an allocation looks wrong and being able to articulate exactly why, and I would like to be reliably on the second side of it.
I also wrote a book, The Numbers Don't Lie, largely because the argument on this page kept turning out to be longer than a conversation.
What I will not do
- Put a client's figures, or a former employer's, on a marketing page — which is why you will not find a case study with numbers anywhere on this site
- Imply a client roster I do not have. JMW-Inc is building its book of business, and I would rather say that than invent testimonials
- Present an estimate as a reconciliation, or a plug as an explanation
- Take an engagement I do not think I will improve, because the first bad outcome costs more than the fee was worth
If any of that sounds like the person you need, let's talk.
Thirty minutes. Tell me how your close runs today and what you do not trust in your numbers.