Back to the day job. No regrets

Seven months ago I walked out of full-time employment with a folder of half-formed ideas, an alarming amount of confidence, and the unshakeable belief that I was going to build something worth building. I had a plan. I had runway. I had momentum.

I did not have a marketing strategy, a co-founder, or any meaningful experience of how brutally indifferent the App Store is to a new arrival who doesn't already have an audience. Details.

A lot has happened since the last post. So let's get into it.

Clearing the Decks

Before the main event, a spot of housekeeping.

Chronode

Chronode is now free with no future updates planned. The core problem was always project detection. Getting it right reliably across every IDE and project-based application is a genuinely hard engineering problem, and one that would take significant time to solve properly.

Thirty downloads and two sales in six months told me everything I needed to know about whether that time was well spent. I also refunded both paying customers. One emailed back to say it was fine and I shouldn't bother. I refunded him anyway. It felt like the right thing to do.

I'm relieved more than anything. It was complicated, I didn't have the bandwidth to fix it properly, and the numbers didn't justify the effort. Letting it go was easy.

BillKit

BillKit has been retired from the App Store. I built it for my own use originally and the honest truth is that the likes of Sage were offering the same thing for free to acquire accounting customers. I never had a realistic shot at competing with that.

I still use BillKit myself, as it happens. A weekly habit at this point. But there's no version of the world where it was going to become a business.

Domain consolidation

This action was more pragmatic than strategic. I'd registered a separate domain for every app I built. Which sounds fine until you're sitting there with a portfolio and the domains alone are costing you a hundred quid a year just for the privilege of owning them. The traffic on those individual sites was low anyway.

App Store traffic comes from ASO, not from people typing your app's URL into Safari. So I ported everything over to subdomains of codel.dev. Simpler, cheaper, easier to manage. Web apps still get their own domains. But a macOS utility's website does not need its own postcode.

Kanodo

It's still on the App Store, still making the occasional sale, and still doing nothing in particular to deserve my attention. Which suits both of us fine for now.

Tushti: In Apple's Hands

If you read last month's post you'll know the full story behind Tushti. The short version for anyone who missed it: bare knuckle fighter, gratitude journalling, five days of building, twelve languages, a complete iOS app that somehow materialised while I was supposed to be focusing on something else entirely.

The update is that it's been submitted for review. Apple rejected it first time around for a missing terms link in the description. My reaction, honestly: here we go again. Always something. Fixed, resubmitted, and it's currently waiting in the queue. It'll get there.

Audibar V2: Done

Audibar V2 has been submitted for review.

As I write this, the screenshots across all twelve locales are done, the metadata is updated in every supported language, and the price is moving from £2.99 to £6.99. That last number makes me feel nothing except that it should probably have been higher from the start.

V2 is a substantial update. YouTube audio playback. Podcast support. Sleep timer. Playback speed control with pitch correction. Media key support. Control Centre integration. Finder integration for right-clicking audio files. Persistent panel position. The lot. It's a different product to the one that existed six weeks ago.

The YouTube angle in particular is something I haven't seen any other menu bar audio app do. You paste a URL and the audio plays in the app. Build a playlist of YouTube links alongside your local tracks. It's the kind of feature that either sounds gimmicky until someone tries it, or sounds obvious once they have. I'm betting on the second one.

Whether the price increase lands well is something I'll know more about next month. I'm not nervous about it. £6.99 for a utility you leave running all day is still very cheap. I've seen far lesser apps asking for £10 and up.

Building vs Marketing

I'll be brief on this because I've said it before and the situation hasn't materially changed.

I tried marketing properly. I did the directories, the social posts, the App Store listings, the whole process twice over for Audibar and Kanodo. The pattern was the same both times: a small spike, then nothing. Not because the apps are bad. Because that's just what happens when you have no existing audience and you're shouting into a crowded room.

I know now that I'm an excellent builder. I am genuinely not a marketing person. I don't enjoy it, I'm not good at it, and repeating the same process hoping for different results would be the definition of something unflattering.

Reddit, meanwhile, continues to shadow ban me into oblivion. I submitted an appeal months ago. The appeal process is 250 characters, roughly two sentences to justify your entire existence on the platform. I have heard nothing back. I remain a ghost wandering the shallow halls of Reddit, invisible to the world, commenting into the void. At this point I've made my peace with it.

The Phone Call

A few weeks ago I had a very short call with my old boss from the previous company I worked at. The person who originally hired me. Not his real name, but we'll call him 'Bob'. Bob left that company in January to take a CTO role at a new company, and he was, as it turns out, in the process of building an internal IT department from the ground up.

This morning I got a WhatsApp message from Bob. He'd been given the go-ahead to start recruitment and, having reached out the week before, was hoping the interest was still there on my end.

While I knew this was a possible opportunity, I hadn't thought much more of the original phone call a few weeks ago. My first reaction was surprised, intrigued, interested. I wasn't expecting anything.

My second reaction, roughly ten minutes later when I looked at Bob's LinkedIn post, was a moment of mild deflation when I saw Vue and React in the requirements. Not because I'm opposed to learning new things. I'd have bitten the bullet if I'd needed to, and of the two Vue was the more palatable option.

But I've been working with Livewire through versions 2, 3 and now 4 for several years. I know how to use it, how to optimise it, and how to make it do everything Vue or React can, just differently. I'd rather be a genuine specialist in a stack I believe in than a reluctant beginner in one I don't. The LinkedIn post gave me a moment's pause. Nothing more than that.

I got on a call with Bob an hour later. No offer on the table yet, there's a proper process to follow, but it was the kind of conversation that leaves you feeling quietly optimistic rather than quietly terrified.

Without going into specifics that aren't yet set in stone, it was the kind of conversation where the stack, the setup, and the working arrangement all pointed in the right direction. The sort of role I'd have been actively looking for in a few months anyway.

Bob already knows exactly what I can do. He has watched me build solutions from EU enterprise HRIS systems to Reporting Suites and Asset Auditing platforms. It sounds like I'm blowing my own trumpet, but honestly, I excel in building solutions and getting shit done, so why not?

Looking back at the Vue and React in the job listing. It's become apparent that it's standard language for posting publicly, and nothing to do with what he wanted from me specifically.

I'm going back into full-time permanent employment. Hopefully.

Why Full-Time, and Why Now

I'd been planning to look for part-time work around summer anyway. The original thinking was three days a week contracted somewhere, covering the bills while keeping the remaining days for my own products. It was a sensible plan.

Full-time changes that calculation. The products don't disappear. Audibar is in review, Tushti is in review, Prismo still exists and still has the best path to recurring revenue of anything I've built. But they become side projects properly rather than the main event.

Seven months of building has taught me a lot. I've shipped seven products: four macOS apps, one iOS app, two web SaaS applications, two retired web apps, and two landing pages that went nowhere. That's a lot of building. The commercial results have not matched the quality of what I've built. Not because I built the wrong things necessarily, but because I'm not a marketer and I don't have the appetite to become one.

The thing I'm genuinely great at is building. The thing this role requires is building. The friction that's been constant for seven months, finding customers, promoting apps, watching traffic graphs and wondering why nobody clicked, disappears entirely. I get to just make things work.

There's also the age question, which I'll say plainly because it's real. I'm 50 in seven weeks. The open market for developers skews younger and cheaper, and AI tools are accelerating that pressure. Going through recruitment in the current market, competing on buzzwords for a stack that isn't even mine, against candidates half my age at half the cost, that would have been brutal.

I'm genuinely excited. That counts for a lot.

What Eight Months Actually Taught Me

If I could go back to 31st of August 2025 and visit the old me standing at the door with his folder of ideas, here's what I'd tell him.

• Focus on ASO first, not websites. The websites matter less than you think for App Store traffic. Put that energy into the listing itself.
• Build smaller apps first. Kanodo and Chronode were complex, ambitious, and took real time. The thing that generated daily installs was a menu bar audio player. Start simple.
• iOS first. The App Store ecosystem is more accessible than I initially gave it credit for. I'd have started there.
• Don't plan a marketing sprint. If the app is good enough, some of it finds its own way. If it doesn't, a weekend of directory listings won't fix that.

But the headline would be: I don't regret it. Not any of it.

The independence was real. The building was real. The learning was real. And I genuinely might not have been in a position to recognise this opportunity as clearly, or to value it as much, without the months of seeing what the alternative actually looks like.

Seven apps in seven months. Two retired, one going free, one waiting, one shipped, two in review. Reddit banned. Directories mostly useless. Marketing mostly hard. Building consistently excellent.

It wasn't the plan. It was better than that. It was honest.

That's all for now. Audibar V2 and Tushti are both in Apple's hands. Prismo is sitting patiently in the queue. And somewhere there's a tarantula app with 2,159 tests and 100% coverage waiting for its moment.

We'll see how the rest of March pans out...