One Spring morning early last year, my wife began complaining to me about an application she had been using for her job. She’s a family law attorney and typically her work is far removed from my world of cyber security and software engineering. But since there was some overlap here, my ears perked up. She jokingly said I could probably write a better app than the one she’s using now. I’m up for a challenge, and I’d been looking for an excuse to brush up my coding skills since they’ve gotten a little rusty being on the management side of things.

Intrigued, I asked her what she was paying for it. She said it cost her $500 annually, $250 per seat for 2 seats, which wasn’t much, but enough to get me out of bed. If I could spend a couple hours putting something together and save her half a grand a year, that would certainly be worth it. I then asked her what problem the app was solving. The goal was to determine child support owed based on a large number of factors. The gist of how it worked was running a bunch of numbers through various formulas, similar to a tax form. Enter numbers into row 1, 2 and 3, add those up, stick them in 4, rinse and repeat that several more times.

I told her she didn’t need an app, because I could do that for her in a spreadsheet. I’m one of those people who loves spreadsheets. I use them quite often in my day job, not because I have to, but because they make my life a whole lot easier. So, an opportunity to save my wife $500 and show off my spreadsheet expertise was too good to pass up.

Needless to say, putting it together wasn’t quite as simple as throwing some sums into a spreadsheet. I also had to use a VLOOKUP once. I was able to put together a working spreadsheet in a little over an hour, and at first glance it looked good. I shared it with my wife, patted myself on the back for a job well done, and promptly forgot about it.

A week later, I got a call from my wife in the middle of the day, which is slightly unusual. ‘Your spreadsheet is wrong.” She’d been trying to use it to run the numbers for a case and they weren’t matching the app she was still using. This isn’t terribly surprising, since software (even spreadsheets) rarely works perfectly the first time. So, I loaded all that context back in my brain, figured out the problem (some minor issue with a formula), fixed it, and went back to work.

At the start, this was an interesting side project for me. But for my wife, this was an essential tool she used as part of her job. She relied on it to determine the amount her clients would pay or receive in child support. And when it comes to child support, every dollar can be fiercely contested, so even small errors matter. Furthermore, once an amount is agreed upon, changing it can be very expensive and time consuming. So, accuracy was paramount.

As engineers, we sometimes forget there is a customer on the other side of what we’re working on, relying on our work. So, we can put out code that checks the boxes for requirements, but have we put ourselves in the customer’s shoes? Do we really understand how and why they are using our product? What small things will delight them and what seemingly minor nuisances will drive them nuts. Without that empathy, I feel it’s easy to ship ‘good enough’ products that aren’t the easiest to use or have more defects than we’d like.

So, what started as just me entering formulas in a spreadsheet, turned into me actually becoming somewhat proficient in Virginia Child Support Guidelines. By understanding the problem I was solving, I was able to find issues before the customer, i.e. my wife, found them.

She did find several more minor issues, but fortunately each time the fixes became smaller and easier. After a month, she was ready to trust it entirely. And the timing was perfect. The original, hard to use app was up for renewal. She chose to not renew her subscription and started relying solely on my spreadsheet. Six months later, she continues to use it and hasn’t regretted her decision (or at least that’s what she tells me).

I could’ve stopped there, saved my wife some money and had a good story to tell other parents at kids’ birthday parties (Having two young children, we go to quite a few). However, my wife saw an opportunity. Why not turn this spreadsheet into a bonafide application that we could sell to other local attorneys? We could charge half the price of the existing app, and offer a better experience. It was hard to argue with her logic. So, what had started out as a way to scratch my coding itch was turning into something much larger.