It started with a spreadsheet. Of course it did.
Aidan is a co-owner of Asset Construction, a civil and structural outfit out of Auckland who do everything from a driveway repair at your place through to working on a $350m government facility. He’s the kind of operator who actually knows his numbers - which, if you’ve spent any time around trades businesses, you’ll know is rarer than it should be.
And the way he knew them was a stack of spreadsheets, cobbled together over years, held up with formulas only he fully understood, sitting alongside a set of internal processes that lived partly in a shared drive and mostly in his head.
They worked. That’s the important bit. They weren’t a mess - they were a genuinely good system built from what they had on hand. It took him and his team hours every week to feed, it couldn’t be handed to anyone else, and one wrong drag of a cell could quietly change a number that big decisions got made on.
I looked at that stack and thought this was going to be a lot of work, but also what better opportunity to deeply learn about business - and it was exactly the thing I needed at that time.
The problem underneath the spreadsheet
Here’s the thing I’ve come to believe after a year and a half inside this industry’s numbers:
Most trades businesses aren’t unprofitable. They just can’t see where the money goes - so every decision is a guess.
They’re busy. The phone rings, the work comes in, the crew is flat out, invoices go out and money comes in. And at the end of the year the accountant tells them how they did, six months after they could have done anything about it.
Ask a hard question in the middle of all that and it gets uncomfortable fast. Which of those four jobs actually made money? Not turned over - made money, after the hours and the plant and the bills and the overhead. Is that digger earning its keep, or are you paying finance on a machine that sits on a yard four days a week? Have you got the cash to cover payroll, GST and provisional tax in the same fortnight? Are you actually pricing the next job off what the last one cost you, or off what you charged last time plus a bit?
Most operators can’t answer those quickly, and it’s not because they’re not sharp. It’s because the answers are scattered across Xero, a timesheet book, three spreadsheets and someone’s memory, and assembling them takes a day they don’t have.
You can’t fix what you can’t see. And by the time it’s visible in the annual accounts, the job’s finished, the price is set and the money’s gone.
What we built
FNA Manager sits on top of Xero and pulls it all into one place. Sign in with Xero and your contacts, invoices, bills and staff come across; you set your hourly rates for people and machines, tag your first job, and it starts telling you the truth.
Job costing that shows real margin against the quote, with hours, equipment, bills and invoices allocated as they happen. Cash flow that forecasts the bank balance rather than reporting it. Estimating that builds a price list over time, so the next quote is informed by what the last job actually cost. Equipment tracking - what each machine earns, what it costs to run, finance and lease included, with reminders for WoF, CoF, servicing and testing. Staff hours and true hourly cost, with timesheets that push through to Xero Payroll. GST, PAYE and provisional tax worked out from the actual books, with a tax pot that tells you whether you’ve got it covered. Retentions on progress claims tracked to their caps, and bond schedules with reminders.
And a nightly run over every machine that says, plainly: keep it, watch it, replace it, or sell it and hire instead.
It won’t quote the job, chase the payment or swing the hammer. That’s the operator’s. What it does is make sure they’re never the last to know.
Fifteen months of mornings and nights
I’ve been building this since May 2025. On my own, from nothing. I of course invited Paul to join me, but he decided to dedicate a year to working on personal growth, which has been, quite frankly, transformational for him. As much as I would have loved to have him on board from the beginning, he 100% made the right choice. He’s never been better, and he ended up helping me anyway, which is awesome.
Almost every morning before the day started. Almost every night after it finished. Any spare hour I could find at the weekend. Alongside client work, which pays the bills and does not politely step aside because you’ve got a product on the go.
If this sounds romantic, it isn’t - it’s just long lonely hours of flipping between “why am I doing this?” and “I can’t stop now”.
It’s fine for the first couple of months. You’re building the exciting parts and every session ends with something visibly better than when you sat down. Then you hit the middle, and the middle is long. The middle is where you’re implementing retention caps on progress claims for the fourth time because you got the edge case wrong twice and the client’s real data proved it. The middle is sitting down at nine at night, tired, on a feature you’ve been chewing on for a fortnight, knowing tonight’s session probably won’t finish it either.
There’s no launch buzz in the middle. There’s no client cheering you on. There’s just whether you sit down or you don’t.
What got me through it wasn’t motivation, because motivation doesn’t survive fifteen months. It was habit, and the fact that it was almost every morning rather than most mornings. Once you let it become “when I’ve got time”, you’ve already stopped - because you will never have time. Small, boring, relentless consistency, in the dark, for well over a year. Don’t get me wrong - some mornings I got up and couldn’t do a thing on it, just exhausted. I didn’t beat myself up over it; I knew that was part of the cycle too. I rested, built up energy again and carried on.
That’s the whole trick. I wish it were more exciting, but it isn’t - you just have to do the time.
What I’ve learned
The other products were the rehearsal
I’ve written before about Static Contact, Client Invoices and the Mangawhai Directory, and I’ve said in those posts that our own products are how I keep my hands as good as my opinions.
I don’t think I fully understood what I was practising for until I was in the middle of this one.
Static Contact taught me to say no to features and to price honestly. Client Invoices taught me how to live inside someone else’s platform - Xero specifically, which turned out to matter enormously. The directory taught me that the data is the mountain and the code is the foothills. Every one of those lessons got spent on FNA Manager (and fed back into the other products too), at scale, on a build that would have flattened me if I’d attempted it first.
If you’re sitting on an idea that feels too big: go and build three small things properly first. Not as a delay - as training. You’ll build the big one in half the time and it’ll be twice as good. I’m always amazed that people think they can do this first try, looking forever for that one-million-dollar idea. Like skipping the years of training, playing up the leagues and winning the NBA.
And the frame works forwards as well as backwards. If this one doesn’t work out - if the beta stays small and it never becomes what I think it can be - then it was just another practice run for the next one. That’s fine too.
A domain expert in your company is a cheat code
Aidan shaped this product out of his own spreadsheets and processes, and he uses it every day in a real business with real money on the line.
I cannot overstate what that’s worth. I’ve spent a lot of my career trying to extract domain knowledge from people through workshops and interviews and discovery documents, and it works, sort of, slowly. This was different. Every feature got tested against “would you actually use this on Monday?” by someone whose own margin depended on the answer. When I built something that was clever but wrong, I found out in days rather than after launch. He also already had processes, staff, equipment and reports in place for me to compare against - what an incredible benefit that was! I could check my work against something already worked out and real, rather than discovering it and then figuring out if I was right or not, or if it was even useful.
There’s a version of this project where I build what I think trades businesses need. It’s a much worse product and I’d have shipped it a year ago. Having a partner with skin in the game who says “no, that’s not how it works on site” has been the single biggest factor in this being any good.
Money software is a different sport
When Static Contact drops a form submission, someone misses an enquiry. That’s bad. When this thing gets a number wrong, someone prices a job off it, or under-reserves for provisional tax, or keeps a machine they should have sold.
That changes how you work. It slows you down in ways that feel frustrating and are completely correct. Reconciliation, rounding, GST edge cases, credit notes, part-payments, jobs that span financial years, the difference between what a machine cost and what it’s worth. You test more. You trust yourself way, way less. You build the boring safety rails before the interesting feature.
I’m a more careful engineer than I was in May 2025, and it’s this project that did it.
Building alone is quiet in both directions
No one to hand the hard decision to. No one to talk the architecture through with at eleven at night. No one to notice you’ve been going the wrong way for three days.
But also: no meetings, no negotiating, no compromise. Fifteen months of holding an entire system in your head means you know every corner of it, and that pays off constantly - I can find and fix things in minutes that would take a new pair of hands a week.
I honeslty don’t know if I’d recommend solo as a default, but it’s an experience! For this one, at this stage, it was right and I’m glad I did it. I now have a resilience that I didn’t know I had, and you only gain it by doing this kind of work.
It’s working
The bit I’m proudest of isn’t the software.
It’s that Asset Construction is running a better and more profitable business because of it. Real decisions - about pricing, about plant, about cash - are being made on real numbers instead of a feel and a guess. That’s the whole point. That’s the entire reason to build anything.
Twenty-odd years in this trade and I’ve worked on some genuinely large, genuinely complex things. This is the one I’m most proud of, and it’s not even close.
Still in beta
FNA Manager is in beta and still growing. It’s live, it’s in daily use in a real business, and there are things on the list I haven’t built yet. I’d rather say that plainly than pretend otherwise - if you come on now, you’re coming on early, and your feedback will genuinely shape where it goes.
It’s $289 a month plus GST for the whole business, however many of your team need access, month to month. First two months free, no card to find out whether it fits.
If you run a trades or construction business in New Zealand, you’re on Xero, and you’ve got a nagging feeling you don’t really know which of your jobs make money - go and have a look at fnamanager.com.
And if you’ve got a spreadsheet you built yourself, that only you understand, that quietly runs your whole business - I’d genuinely love to hear about it. That’s usually where the good ones start.