Matt Bearman’s Blog

Reducing Screen time by 86%

2025-10-14T00:00:00+01:00

“How long is left on this?” I asked, referring to the TV show my wife was currently watching, in which I didn’t have much interest.

Sophie thumbed the upper most quadrant on the remote control’s directional doughnut, triggering the screen to display a progress bar overlay, culminating in the text “-8:17”

“About eight minutes, why?” She replied.

Previously this exchange would have happened differently. And by that I mean previously this exchange wouldn’t have happened at all. I wouldn’t have even asked how long was left, I’d have just reached for my phone, hitting up the usual suspects — Reddit, Ebay, Hacker News. If I was feeling particularly self loathing that day I may have also stopped in on Linkedin for a good thought leading.

Today was different though. My phone was way over the other side of the room. On a shelf. With an uncharacteristically high battery percentage for this time of day. And on that shelf it would stay.

“I was just wondering if I had time for a quick game of Mario Kart” I explained, before concluding that I did in fact have time. I slid the Switch from its dock under the TV, and fired up a 150cc Grand Prix on some of the new tracks we’d recently purchased. I came second on the first two races — I was a little out of practice at this speed, due to mostly playing at 50cc with my kids — but I managed to grab victory on the last two races, and took first in the Grand Prix over all.

I hadn’t just walked in on Sophie watching TV though, I’d already spent the last half an hour or so on my laptop trying to fix the SD card my youngest child’s digital camera had somehow managed to corrupt. Alas my efforts were in vain, that card was so fucked I couldn’t fsck it. It’s a shame as she’d taken some pretty good photos, as well as hundreds of the toilet for some reason, but the point is I tried. This was something that had been on my “to fix” pile for too long, not having my phone in easy reach was all it took for me to finally get round to it.

There is an obvious flaw in those two examples — I didn’t actually reduce screen time, I swapped one screen for another. The Switch and laptop still have screens. Reducing screen time isn’t necessarily my end game though. It’s more about eliminating the subconscious reflex to grab my phone as soon as I feel the slightest twang of boredom. I want to do something productive, and yes I consider playing video games productive. Instead of mindlessly staring at my phone’s screen when my mind was idle, I actively found some things to do.

Drastically reducing the time I spend on my phone is something that happened very gradually all at once. It’s been on my mind for years.

The catalyst for my recent breakthrough happened about a three weeks ago. My wife and I were walking around a nearby city, and two separate people nearly walked into us, because they were locked in fierce staring contests with their phones. They didn’t even notice the near collision, and just carried on living their lives. Of course people would almost walk in to each other in the pre-phone days as well, but at least they would have noticed and apologised for their inattentiveness. And then I would have apologised for existing, because I’m English and that’s just what we do.

This new world scares me, I’m scared for my kids. I know I sound old, and maybe I am, but this can’t be healthy. I don’t want my kids to grow up and become more zombies walking around oblivious to the world around them. And that has to start at home. If they don’t see us mindlessly mesmerised by the handheld slab of misery, maybe they’ll be less likely to end up that way.

I’ve tried before to reduce my phone use time. Back in 2020 I complied a list of dumb phones that might be able to replace the smart phone I had at the time, but never pulled the trigger. I cut most social media out of my life, hence the lack of Facebook, Instagram, TikTok etc on that earlier list of doomscroll destinations. I deleted the Reddit app, forcing myself to brave the toxic swamp of a mobile website if I wanna get my fill of memes, cat pictures, and meme cat pictures. I tried setting screen time limits for certain apps, but they’re just too easy to bypass.

Then one morning I hit upon a radical new technique. I’d leave my phone on the bedside table. How often do I really need my phone throughout the day anyway? I wouldn’t miss any calls or messages as they’d still come through to my watch.

The crazy part was it worked. Simply not having my phone in my pocket was all it took for me to stop habitually consuming the cerebral equivalent of a Big Mac. I turned my own laziness against me, I’m not gonna climb a whole fucking flight of stairs just to filter through all the AI garbage on Hacker News for the occasional nugget of interest.

I still feel that impulse to reach for my pocket in moments of boredom. When my fingers finding a vacant pocket it’s enough of a pattern interrupt for my conscious brain to kick in, remembering why my phone isn’t there. And so I find something more productive to do instead. As time has gone on I feel that impulse less frequently. And when I do want to reach for my phone, I remember it won’t be there much quicker, usually before my hand has even started moving.

I feel like a weight has been lifted from me. Literally. I’d forgotten what it was like to not constantly have something in my pocket. After just a couple of days without my phone on me at all times I found myself hating the feel of it in my pocket on the rare occasions it was there. The physical weight of it, and also the mental weight of knowing it’s there, ready to ensure I never feel bored again.

I want to feel bored. Bored is good. Bored is when you actively find something interesting to do. Be it trying to fix something, playing a video game, or even writing a few hundred words about reducing screen time.

I’m honestly kind of embarrassed — all the ways in which I’ve tried to reduce the time I spend on my phone, and it never occurred to me to simply not have it on me. It doesn’t even need to be at the other end of the house, just the other side of the room is enough to keep my lazy ass from doom scrolling.

According to my phone my screen time for today is 24 minutes, down 86% on the average from before I stopped keeping my phone on me at all times. Although I doubt I’ll keep the usage time quite that low, as I do like to use my phone for things like YouTube and podcasts as background noise while I do other activities, which I haven’t done recently. Although I’m not actually sure how screen time is calculated, ie: does listening to a podcast with the screen locked count? I suppose I’ll find out soon.

Battery life is an additional bonus. Before my phone would usually be sub-30% by the end of the day, often in power save mode, and sometimes would have already needed a mid-day charge. Now it’s finishing with the day with more than 80%. I don’t actually need to charge the damn thing every night any more.

This is particularly good for me. I’m currently rocking an iPhone 13 mini, and I’m dreading the day it dies and I can’t have a phone this small any more. Hopefully this change will help keep the battery going for longer. And that will allow Apple enough time to come to their fucking senses and make another small phone. There are literally dozens of us begging to pay not very much money for one.

— Matt

The Joy of Aeroplane Mode Programming

2025-04-30T00:00:00+01:00

Recently I got back to the basics of programming - just me, a computer, and a book - it was fantastic. No internet. No distractions.

It’s the same way I learned to program as a kid in the early nineties, before my family had internet access. I tried (and failed) to write a text adventure game for the Atari ST. All I had was the system, its built in BASIC interpreter, and the programming book it came with.

I found my way back to this simpler method of programming due to a recent retro computing acquisition - a Psion Series 3a. As a geeky kid in the nineties I would lust after these while flicking through the Argos catalog, however my pocket money would never quite stretch to obtaining one. So imagine my surprise and delight when I discovered a local charity shop had one for sale, for just £30. It was sold as “non-working” but after cleaning the battery contacts it came to life and has been working fine ever since.

Since the very beginning, all Psion models have shipped with their own built-in programming language, complete with onboard code editor and compiler. That language is OPL - literally just “Organiser Programming Language”, and it’s weird. I’ll save documenting all the language’s oddities for another post, but suffice to say it wasn’t a language I could trial-and-error my way to comprehension.

Originally I had no intention of learning OPL, but then a thought bubbled up that I could’t shake - the Psion Series 3a would be great for a Wordle clone. A quick search didn’t reveal an existing version, so I decided to have a stab at it myself. I found a copy of the official Programming Manual on eBay and set to work.

My Psion coding setup

I didn’t originally plan to do this without turning to the internet for help, in “Aeroplane Mode” if you will, it just kind of happened. The combination of the Psion’s portability, and its lack of internet connection, meant when I had a spare few minutes I could open up my 3a and its accompanying programming manual, and write some code.

I was instantly transported back to those early days of learning to program in Atari ST basic - finding the relevant information in a real live book, and trying it out on the machine. Rinse and repeat, until you have something that works.

It’s a very quick way to get into a flow state. Book, code, book, code, book, code. No distractions.

It kind blew my mind how simple it could be, and how I’d managed to loose touch with that simplicity after close to 30 years. I haven’t bought or used a real programming book since university back in the mid 2000s, when I bought the recommended books on C++ and graphics programming. Every language I’ve learned since then - PHP, JavaScript, Ruby, Go - I’ve learned through online material only, and I continue to turn to the internet when I get stuck.

Actually that’s not totally true, I do own a physical copy of “Why’s (poignant) guide to Ruby”, but that was pretty outdated by the time I got it, I just like the writing style.

The internet is a great resource, and allowed me to learn all those languages for free, ultimately giving me my career, but it’s not without its downsides.

When looking online for the solution to a problem, it’s rarely as simple as just finding the answer. There’s often opinions to wade through - five different people on Stack Overflow giving five different answers, each convinced theirs is the only way. Then there’s the rabbit holes that suck you in, and links to other pages begging to be followed. Sometimes my searching will lead me to an interesting blog post on the subject, “Cool, I wonder what other posts they have?” Suddenly half an hour’s passed and I’m reading a very convincing argument in favour of Hungarian Notation.

Throw modern AI coding assistants in to the mix, with their lag and hallucinations, and it’s a wonder anyone can get anything done these days.

None of that can happen with a book. You find the answer, you go back to coding. Book, code, book, code.

I don’t know if this method of programming has a name, or if it even needs one, but I’m gonna go ahead and name it “Aeroplane Mode Programming”. Originally I wanted to call it “Offline Programming”, but that’s already the term for programming robots in a simulated environment. My second choice was “Unplugged Programming”, but that’s apparently programming without a computer at all.

Anyway, the point is if you don’t have a book for your language of choice - get one. And not an e-book, get a real, dead tree, analogue book. E-books are great for linear fiction, but nothing beats being able to quickly flick through pages when it comes to random access reference materials.

It doesn’t even matter if you’re not sure what the “best” book for your language of choice is. Any book that allows you to achieve and maintain a flow state is better than wading through the conflicting opinions of multiple experts online.

It definitely worked well for me. I was able to learn OPL and write a Wordle clone in a couple days, using just a book and an early 90s PDA with a 640 x 160 pixel monochrome display.

Now if you’ll excuse me, I’m off to buy a book on Go.

– Matt

Senior Engineers: Simplify and add lightness

2024-06-19T00:00:00+01:00

A former engineering manager once told me something that has stuck with me through my career more than anything else:

“The job of a senior engineer is to simplify”

That one simple sentence has done more for my growth as a software engineer than anything else.

He went on to explain how in general junior engineers add complications, mid level engineers tend to maintain the status quo, and so senior and above must reduce complexity.

I love the term “simplify“, it’s universal

By simplifying code it becomes easier for other engineers to understand
By writing good documentation we can simplify the process of new engineers getting up to speed with a project
By simplifying processes we can give others more time to focus on the real work
By writing good commit messages and PR descriptions we can simplify code reviews
By pairing we can simplify a task
By mentoring more junior engineers we can simplify their levelling up process
We can even simplify others lives directly, eg: by taking something off their plate if they seem snowed under

You may at this point be thinking this is nothing new, I’ve just given a standard senior engineer job description, and while that is kind of true, it’s more about the framing - it’s not just what we do, but why we do it.

This nugget of wisdom was bestowed upon me during my first few months as a senior engineer, and if I’m honest with myself, I don’t think I fully deserved the title at the time. Sure I had the coding chops, but it was my first “proper” job after spending eight years working alone on my one-person SaaS business. I had some huge knowledge gaps when it came to soft skills.

I suspect I only got the job as senior engineer by virtue of my age and years of experience.

Having my job description distilled into that single short sentence really helped the role click for me, and allowed me to grow in to engineer I am today.

Hey look at that, by creating a simple job description, my manager simplified my career growth. Simplicity really is the gift that keeps on giving.

I hope that by passing this knowledge on I can help simplify the lives of some other newly senior engineers that are struggling with the title.

And maybe even some that aren’t so new to being a senior.

I’ve certainly worked with some seniors who seemed hell bent on making things more complicated. The cynic in me wonders if it’s a deliberate ploy - by introducing complex code and obtuse practices, they become the gatekeeper, making themselves appear all-knowing and indispensable.

However, the realist in me knows those engineers probably aren’t scheming their way to job security, they’re just unlucky. I feel very fortunate to have had the ephemeral senior title explained to me so succinctly. Especially as I was so new to it.

Simplicity should be a goal no matter what we do in life. Colin Chapman, the founder of Lotus Cars said it best:

“Simplify, and add lightness”

Note: While this quotation is generally attributed to Colin Chapman, there is some debate about the exact wording, and whether he was even the first to say it

I love this phrasing. “Simplify” comes first, that’s the primary goal. And “add lightness” has a beautiful juxtaposition to it - to make something lighter you generally take things away. The concept of adding lightness almost sounds wrong, but still somehow makes sense.

The ol’ impostor syndrome still rears its ugly head from time to time, but most days I feel I’m good at my job. And that’s because I’m not trying to be a senior engineer, I’m just trying to simplify.

– Matt

🥭

2023-01-25T00:00:00+00:00

I’m writing this on my phone as there is a cat on my lap. Her name is Mango, she’s 7 years old, and she spends most of her time on my lap these days.

Mango is dying, and I think she knows. She has renal lymphoma (cancer of the kidneys), and likely has 2 - 3 months left. Chemotherapy was an option, but not a very good one. The specialist said it might extend her life by a few months. We decided it wasn’t worth the additional stress it would put on Mango. So she’s taking steroids that will give her the best quality of life.

I’m not sure why I’m writing this, or if I’ll ever publish it. Partly I just want to gather my thoughts, and partly I want the world to know about Mango, and what a perfect companion she’s been.

She’ll be survived by her brother, Monkey. An equally awesome cat, I really fear for how he’ll deal with losing his sister.

We adopted them as 8 week old kittens, long before we had any human children. To this day my wife and I think of the cats as our children - siblings to our human kids.

Mango and Monkey when we first adopted them

Mango, like her brother, is incredibly affectionate. As many cats are, contrary to the stereotype. I adore them both, but have always had more of a connection with Mango, just as my wife has more of a connection with Monkey. They seemed to choose their favourite human, settling on our respective laps the second day we had them.

A favourite memory of mine is when I had my first covid vaccination it left me feeling pretty rough. I spent most of the afternoon dozing on the sofa. Mango curled up next to my stomach and stayed with me for hours.

Mango keeping me company while I sleep post vaccination

She and monkey could tell when my wife was pregnant, taking it in turns to sleep on or near her stomach.

Mango is a beautiful fawny ginger colour, unusual for a female. She has little tufts on her ears, and leopard spots on her belly, which she will gladly let me rub for hours on end. She also loves chin tickles and head scritches, closing her eyes and lifting her head for latter as if to try and get more. She purrs hard and often, it has a sort of croaky quality to it.

Mango enjoying head scritches (and distracting me from work)

She has character, and loves to eat, and yells at us incessantly while we prepare her food, especially if it’s wet food.

She likes to climb, and I can always tell when she’s bitten off more then she can chew, and is unsure of how to descend from the lofty height she’s attained, as she does what I call the “tonguey meow”. It has a specific sound to it, but more distinctive is the way she puts her tongue between her bottom canine teeth. It’s adorable. I’ve just realised I don’t think I have a photo or video of this, and now I probably never will.

She frequently sleeps in our bed with us. She can take up a huge amount of space for a small animal, but it’s worth it for the relaxing sound of her purring, as well as the additional heat in winter.

I always thought our cats would live long lives, seven years isn’t enough. I love her company, and wish I could have seven more years of it. My youngest human child probably won’t even remember Mango existed, but she’ll definitely hear a lot about Mango when she’s old enough to understand.

She’s always spent a lot of time on my lap, I regularly feel her kicking and twitching as she dreams. I’ve heard pets likely dream about their humans, I hope she has sweet dreams.

One of last pictures I have of her

I deliberately didn’t really proof-read or edit this post. With the exception of adding pictures and this little foot note, it is exactly as I wrote it - on my phone’s notes app with Mango sleeping on my lap.

Mango passed on Monday 7th November, 2022. I miss her everyday. Her brother Monkey predictably had a period of mourning. He still has days when he seems a bit lost, however he’s doing much better now.

Adopting an animal is a strange thing - we love them like children even though we know in all likelihood we’ll out live them. Yet I wouldn’t trade the time I had with Mango for anything.

Thoughts on working a four day week after six months

2022-10-05T00:00:00+01:00

It’s a great time to be a software engineer, demand is high, as is compensation. This presents a great opportunity for anyone who values free time over disposable income. People like me. For the last six months I’ve been working a four day work week, taking a 20% pay-cut and essentially becoming part-time.

The point of this post is to discuss how I came to be a part timer, and more importantly discuss the pros and the cons of doing so.

How it started

I’ve seen this question pop up a few times on sites like Hacker News, how can an you work a 4 day week as an employee?

Of course it’s easy if you’re a founder or freelancer, but how do you get management buy-in when you’re an employee?

For me the change coincided with changing jobs around a year ago. I was very upfront with every company I interviewed with that I would like to work a four day week. To my surprise none of them seemed to have any issue with this. I ultimately joined Krystal, and while they were happy with my desire to work a four day week, they did require me to work a full five days for the first six month probationary period, which I was happy to do. And true to their word as soon as my probation was over, I dropped down to a four day week.

This actually fell under their flexible working policy, which only kicks in after probation, hence the delay. Had I stayed with my previous employer, I would have also been able to drop down to four days there as well, as they had a very similar flexible working policy. So if you are looking to make that change I’d recommend looking into your employers flexible working policy if there is one.

Failing that, the main thing to do is just ask. Software engineers are in high demand at the moment, so take advantage of that. Most managers will recognise that it’s better long term to keep you for 80% of the time, than to lose you completely to a more forward thinking company. If you do get a hard no, and the four day week is something you really want, then it’s time to dust off your CV, there are plenty of companies that will oblige you. Just be upfront about your requirements from the start.

Of course, all of this is coming from the perspective of a senior Ruby engineer in the UK. Your mileage may vary depending on skill set/seniority/location/etc.

How it’s going

Awesome. I’ll cover this in more detail in the pros section, but I can’t recommend it enough. There are some cons that I didn’t see coming, however the pros outweigh the cons by an order of magnitude.

If you are in the fortunate position of being able to live comfortably on 80% of your current salary then go for it.

Pros

You gain more than you lose

Changing to a four day week cost me 20% of my salary, but in return I get an extra 50% in time off (my two day weekend became a three day weekend). It almost feels like I’ve cheated the system somehow.

Another aspect to this perk is taking time off work. When I worked a five day week, if I were to take four days off work, I’d have at most 6 days off in a row. Eg: If I took Monday to Thursday off, I’d then not be working from Saturday to Thursday - six days.

Now if I take four days off I get ten days off in a row. Ten days! That’s a 66.6% increase, at the cost of just 20% salary. It’s a no brainer.

Of course a flip side is it didn’t just cost me 20% of my salary, it also cost me 20% of my holiday allowance, however I still come out on top. As a full time employee I had 31 days of paid holiday a year. Now I have 24.8 days paid holiday, except I don’t work any mondays. There are roughly 52 Mondays in a year, so really I now get 76.8 days holiday a year.

As the numbers show, you really do gain more than you lose.

More time for family and projects

This was my primary motivation for wanting a four day week. I have two young children, an awesome wife, a house to improve, and too many hobbies and projects. Now I get an extra 52 days a year for all these things.

Bad news buffer

Another advantage of a flexible working policy is I can change my working arrangement more than once - in my case at most once every six months.

This means if something unexpected and expensive happens - like maybe an corrupt cuntservative government causes runaway inflation while tanking the value of the pound with their fiscal incompetence (hypothetically, of course) - I can return to working five days a week, giving myself a 25% pay rise in the process.

It great to know I have this option, although I hope to never need it. In the event of a one-off unexpected large expense that clears out our savings, I could work five days a week until we’re back in a good financial position, and then drop back down to four again.

The weekend comes round fast

This is another case where the 20% difference punches above its weight. My work week is only one day shorter, however it seems the weekends rush up on me at lightning speed. Since changing to a four day week I’ve never found myself willing the weekend to hurry up and arrive. My wife has made a similar observation.

There is also a down-side to this, which I’ll cover in the next section.

Cons

Tuesday morning brain fog

You know that feeling when you take some time off work, and when you return you can’t remember what the hell you were doing and it takes some time to ramp up again? I feel that at the start of every week.

I assume this is more of a “me” problem, and it’s likely different for other people, however I’ve learned that three days is the threshold for this brain fog to set in. Maybe I was previously fine with a two day weekend because I was used to it, however I’m six months in and the brain fog persists, so I suspect it’s not something I’ll ever get over.

It doesn’t take long for the fog to clear, so it’s not a great loss, but it’s not a great way to start every week. I could also solve this by taking a single extra day off mid-week, and sticking to a two day weekend, but I’ve no interest in doing that.

Awkward conversations

This was something I never thought about - I find it a little awkward telling my co-workers I no longer work Mondays. I worry if they’ll think I’m lazy and/or overpaid. Admittedly that’s some projecting on my part - I am both lazy and overpaid. It’s not just me though, I’m 100% convinced all software engineers salaries are obscene (that’s a subject for another post). Of course I’m not going to turn down an obscene salary when it’s offered to me, especially for doing something I enjoy and have no moral issues with.

It’s particularly awkward when a meeting or something is proposed for a Monday and I have to interject and remind everyone that I won’t be available. On the plus side this happens less frequently now most of my co-workers know and remember that I’m only in Tuesday to Friday.

The weekend comes round fast

Occasionally when I’m getting stuck into a nice juicy project, I’ll suddenly find it’s Friday afternoon but I’m still in the zone and not ready for a three day break. Of course that used to happen when I worked five days as well, but now not only is it more frequent, but I have to wait even longer before I can get stuck in again.

The future

As I’m sure you’ve guessed I have no intention of going back to a five day week unless I absolutely have to. If finances allowed it I’d take it a step further and drop down to three days. No matter how much I like my job and the people with whom I work, I can’t imagine ever working more than I need to. There’s just too many other things I want to do during my brief stint on this earth. I’ve always felt this way, however becoming a father has definitely intensified this feeling - I’ve got two awesome children now, I want to minimise my chances of missing time with them because I have to work.

I’m acutely aware of how privileged I am to be in this position, and I constantly remind myself that however unlikely, the computing industry could implode and my run of luck could end. So I’m damn sure going to make the most of this while I can. For some that may mean working long hours to save lots of money for the future, and there’s nothing wrong with that - I can certainly see the appeal of FIRE.

For me it’s all about time - I’d rather have the time now while I’m young(ish), able bodied, and have small children to watch grow up. If this whole industry goes to shit in ten years time and I have have to start working six days a week doing what ever work I can get to keep a roof over our heads, I’ll have no regrets.

– Matt

Selling my startup - one year on

2021-06-24T00:00:00+01:00

It’s been a little over a year since I sold Saber, my SaaS website feedback tool. I was Saber’s founder, creator, and only employee. It was hard to let my baby go. But was it the right decision?

Yes, yes it was.

Before selling I definitely had some fears and concerns:

What will they do to my baby?

This was my biggest concern about selling in general. Luckily this fear was mostly put to rest before I’d even signed anything. From speaking to Steve (Saber’s new owner) before agreeing to sell I was happy with his plans and confident he would stick to them.

I’m happy to report my confidence was well deserved. Here we are 14 months later and externally Saber has barely changed at all. Steve and his team have given the home page a much needed refresh and fixed some issues behind the scenes, however most of his effort has been rightfully focused in marketing.

I’ve heard many stories of small SaaS tools like Saber being bought, bled dry, and then shut down. I wouldn’t have sold to anyone who I thought might do that and I’m happy to report my initial impression of Steve was spot on. I don’t think I could have found anyone better to buy Saber.

Will I miss the freedom?

This one is little more tricky - I do miss the freedom over my time. However I’d already been working full time for FreeAgent for a few months before I sold Saber, so that ship had already sailed. From the ages of 22 to 34 I was almost entirely working for myself, setting my own hours with total freedom. Having lived that way for so long it’s a challenge to stick to someone else’s arbitrary schedule.

I’ve recently moved on from FreeAgent, I now work full time for Krystal. FreeAgent was a great place to work and I’d recommend it to anyone, however Krystal is more suited to me. I’ll be working four days a week at Krystal, which will give me some more free time.

There’s another aspect to freedom - financial freedom - and that’s been awesome. The combination of selling Saber and working for a company with a good salary has given my family a level of financial freedom we’d never previously known. We’ve been able to very quickly save a deposit to buy our first house and live comfortably on my salary alone.

There’s a balance to be struck and I’ve been feeling that side project itch lately, but for now being a full time engineer is what’s right for me. I’ve no doubt that at some point in the future I’ll be taking all the lessons I learned from Saber and applying them to a new project.

Will I miss working on Saber?

This wasn’t so much of a fear for me. Having spent 8 years working on Saber I was pretty bored of that one code base. I built it all from scratch and knew it inside out. Working for FreeAgent and then Krystal meant throwing myself into their unfamiliar codebases. I’ve learned so much more than I would have working on Saber.

After the sale I continued working on Saber as a freelancer for a few hours a month. That’s come to an end now and I can honestly say I don’t miss it.

I do miss some of the non-coding aspects. At FreeAgent I was a back-end engineer only. At first I thought that was great, but over time I found I missed being a full-stack guy and learning many disciplines. It’s one of the reasons I decided to change jobs. With Krystal I’m still primarily a Rails engineer, but it’s much more of a full-stack role.

In summary

Overall selling Saber has been a positive experience and something I’m very glad I did. It was never my plan to build a business up to sell, but I have no regrets.

This will be my last post on the subject of Saber. I know the majority of my readers originally found this blog through the side project to profitable startup series, but it’s time to move on. I don’t want to be that guy who constantly writes about the business he used to run. There’s new stuff on the horizon and I hope you’ll enjoy following along with my future projects as much as my time running Saber.

– Matt

How WhatsApp convinces you to give it your contacts

2021-01-08T00:00:00+00:00

The Internet is currently ablaze with the news that WhatsApp will soon be requiring more user data to be shared with Facebook. This reminded me of another nefarious thing WhatsApp does that I haven’t seen anyone else talking about:

WhatsApp uses dark patterns to coerce you into giving it access to your contacts

While WhatsApp will work without access to your contacts, the user experience is deliberately degraded. Without access to your contacts, WhatsApp prominently displays each participant’s phone number above their message, with their display name shown next to it, almost as an afterthought:

The name is very low contrast compared to the background, where as the phone number is listed first, in larger text, and a high contrast colour. Your eye is immediately drawn to the phone number, making it hard to see at a glance who’s talking. The display name will also be truncated if it’s too long for the available space.

Even worse is the list of chats, which only displays the phone number for non-group chats, no names in sight:

You do at least get the user’s profile picture here, but not everyone adds a profile picture, and even if they do, they may choose one that is not easily recognisable.

There’s no good reason for WhatsApp to treat privacy conscious users this way. The name field is required when setting up a WhatsApp account, so every user has a name that could be displayed here instead of just the phone number.

While writing this, I decided to install Signal, to see how it handles labeling chats without phonebook access. I wasn’t surprised to learn Signal choses the user friendly and logical option - displaying the participant’s name, with the phone number below it in smaller text:

Signal also helps distinguish between chats with users who haven’t added a profile image, by generating a uniquely coloured circle containing their initial. This isn’t unique to Signal, or even a new idea, it’s a tried and tested pattern. Compared with WhatsApp’s “generic bust”, it presents further evidence that WhatsApp is made deliberately hard to use if you don’t give it access to your contacts:

When I first installed WhatsApp I didn’t give it access to my contacts, as I tend to err on the side of minimal permissions, and I couldn’t see why it would need access. However, it didn’t take long for me to become frustrated with this user experience. I was constantly struggling to find conversations, or identify who was talking in group chats.

I asked a friend with more WhatsApp experience if there was a way to make it show names instead of phone numbers. Their response was “Oh, that’s easy. You just need to give it access to your contacts.” This wasn’t the solution I was hoping for, and I resisted for as long as possible. Yet eventually the dark pattern wore me down, and I granted WhatsApp access to my contacts.

I joined WhatsApp in late 2014, at which point they’d already been acquired by Facebook. I find myself wondering if this was a preexisting pattern, or something that happened after the acquisition, possibly at the behest of Facebook.

Using display names as the primary identifier would be a vastly improved experience for those who chose not to give WhatsApp access to their contacts. Instead WhatsApp provides this unnecessarily worse experience, in an attempt to gain access to user’s contacts. And it works.

I’d love to be able to say that I’ll be replacing WhatsApp with an alternative like Signal, but WhatsApp’s network effect is a strong one, and it’s currently how I keep in touch with my family.

– Matt

Full-time web developer vs being a full-time dad

2020-10-31T00:00:00+00:00

For the last 13 years, I’ve been some form of professional web developer, and for the last 21 months, I’ve been a dad. We’re lucky enough that my salary means Sophie (my wife) is able to be a stay at home mum. We both love this arrangement, as we don’t need to sort out child care, and as I work remotely from home, our daughter gets to spend a lot of time with us.

Note: We decided before she was born that we didn’t want our daughter to have an online presence until she was old enough to decide for herself. This is why I won’t be referring to her by name. My apologies if that makes this a little awkward to read.

In August this year, Sophie had to be rushed into hospital for life-saving surgery, which thankfully was successful. She did however have a long recovery period, so for two weeks after the surgery, I didn’t work.

I spent that two weeks being a full time dad, carer, and house husband.

Under normal circumstances, I work 9:00 to 5:00 Monday to Friday, while Sophie looks after our daughter and the house. Outside of those hours we share the housework and childcare. One downside to this is that Sophie gets to spend a lot more time one-to-one with our daughter than I do.

However, during those first two weeks of Sophie’s recovery, I did all the childcare, and all the house work. My days started at 7:00am, and usually didn’t finish until 8:00pm, when we’d had dinner and I’d done the washing up. I’d get a couple of hours off in the afternoon when our daughter was napping, although that was often spent doing house work. And this was seven days a week.

It was the longest days I’d ever worked, and by far the hardest job I’ve ever had. But it was also the most rewarding.

This new routine also highlighted the effect my usual sedentary lifestyle was having on my health and fitness, as at the end of the first few days of being a full time dad, I’d lay down in bed with aching legs. I’d not done anything specifically strenuous on those days. Simply running round with a very active toddler all day was far more exertion than my puny legs were used to. It’s prompted me to get back into running, as well as look into setting up a sit/stand desk (I did have a standing desk for a while, but ultimately abandoned it. That’s s story for another post).

I loved getting to spend so much time one on one with our daughter, and we definitely connected more. She’s such a joy to be around, at an age where she’s putting a lot of small, adorable, and often hilarious sentences together. This also means she is able to express how she feels, and even engage in something resembling a conversation.

I also had a lot of learning to do, things like what washing machine setting to use when washing bedding, and how to time preparing our daughter’s meals. One of the most important lessons for me was learning more of her abilities, and just how much she is able to do now. As a result, I feel more at ease when she is experimenting and exploring. I’ve always been the more cautious parent, which I think is in part because I don’t spend as much time with our daughter, so I’m less aware of her capabilities.

We also learned new ways of making each other laugh, which often ends up in a positive feedback loop turning from giggles into uncontrollable laughing fits.

I enjoy programming, and I love my job, but at the end of those two weeks, when Sophie was feeling well enough to return to a more normal routine, I honestly didn’t want to go back to work.

I feel incredibly lucky to be able to do the job I do. It’s physically easy (although often mentally exhausting), the pay is great, and I can do it from the comfort of my own home.

That being said, if money wasn’t a factor, I’d chose being a full time dad every time. It’s the best job I ever had.

– Matt

Why I sold my startup for £100,000

2020-04-12T00:00:00+01:00

I launched Saber, my website feedback tool, as a side project in 2011 - back then it was called BugMuncher. Four years later I decided to stop working as a freelance web developer, and put all of my time, and savings, into trying to grow Saber to the point it could pay all my living expenses, and turn a profit.

I documented my progress with complete transparency on Saber’s blog, and after a little over a year of working full time on Saber, I achieved my goal. Saber was profitable.

This is the final entry in my transparent blog series, as from 1st April 2020, Saber is no longer mine. I have officially sold it.

To anyone who’s been following my progress taking Saber from a side project to a profitable startup, my decision to sell Saber may come as a bit of a surprise. After all, a lot of my previous writing has been singing the virtues of being a bootstrapped, solo founder.

For the last couple of years Saber had been stagnating. I’d reached an equilibrium between growth and churn, and had never quite been able to break thought the $5k MRR barrier. This was a little disheartening, as nothing I tried seemed to make any difference, but I was happy to keep trying, as the revenue Saber brought in was enough for me to live fairly comfortably.

That all changed when I became a father at the end of 2018. Suddenly the goalposts had been moved, and the revenue Saber was bringing in wasn’t really enough. We were scraping by, but only just, and were no longer able to put any money aside for savings. Sophie (my wife) and I really wanted to buy our first house, but that was never going to happen if we were unable to save up for a deposit.

The money wasn’t the only problem. The truth is I was bored. I’d been working on Saber for eight years, four of those as my full-time job. I was craving a new challenge. I think this was exacerbated by the fact website feedback was never a passion of mine. I created Saber because it was something other people said they wanted, not because I wanted it. I broke the first rule of starting a business - I wasn’t my own customer. I wasn’t dogfooding.

That’s why In July 2019 I started looking for something else, and very quickly found a perfect looking opening for a senior engineer at FreeAgent. It was a remote job, a new challenge, and a salary of nearly double what Saber was able to pay me. So when I was offered the job, it was a very easy decision - Saber got downgraded to side-project again, and for the first time 8 years, I was working a normal day job.

I’m happy to report it’s a decision I don’t regret. I love the work, the people are awesome, and the salary has allowed us to live comfortably, and save towards buying our first home.

There was only one problem. I was still having to find time to work on Saber. Except with a full time job, a family, and other hobbies, I found myself giving Saber less and less of my time. I’d put Saber in “maintenance mode”, ie: no marketing or new features, just bug fixes and support. Even so, I started to resent Saber, and the time I had to spend working on it. What was once my pride and joy started to feel like burden, a chore. I decided it was time to let it go.

The first thing I did was reach out to the guys at Usersnap, as although they were technically a competitor, I’d been in contact with them throughout my journey. They even sent their congratulations when I achieved profitability, and are just all round great guys. Unfortunately Usersnap weren’t looking to make any acquisitions at the time, and recommended I try FE International, a service that brokers the selling of online businesses.

While I was looking into FE International in December 2019, I received an unexpected email from a long time follower of my progress with Saber named Steve. In the email, Steve asked if I’d be interested in selling Saber to him. The timing couldn’t have been better, and of course I told him would be delighted to sell as long as the price was right. I instantly liked Steve, and his plans for Saber. He is also a solo founder of a bootstrapped startup, and he wanted to keep running Saber as it is, with the goal of growing it.

Before any money was mentioned, I already had a number in my head. That number was £100,000 - roughly 3 x Saber’s annual revenue. I arrived at this number because 3-5 x ARR is usually quoted as the standard guide price for a SaaS business, and as Saber had been stagnating for a while, I felt the low end of that scale was more realistic. £100,000 would also allow us to put a deposit down on a house, and have good portion left over for future projects.

Knowing that, I could scarcely believe it when Steve’s opening offer was in fact 3 x Saber’s annual gross profit. As Saber has super low overheads, that worked out to be $134,151, or roughly £103,000 at the time. Not only had the offer to buy Saber come at exactly the right time, but the offer was also for exactly the right price. It was almost enough to make me believe in fate…almost. I loved that Steve made such a reasonable offer, and didn’t try to low-ball me. Perhaps I could have got a little more with some negotiating, but the offer was very fair, and even slightly exceeded what I was looking for, so I was more than happy to accept it.

Over the next three months there was due diligence, contracts, and asset transfers, which takes us to today. Saber is no longer mine, and I couldn’t be happier. Although my life isn’t totally Saber-free quite yet, I’ll be working with Steve over the next 6 months to help with the transition, and of course I’ll be interested to keep up with Steve’s progress in the future. Full disclosure - the deal actually worked out to be worth around £110,000 in the end, as the pound had weakened against the dollar over that three months. A nice little bonus, but as I was willing to accept £100,000, that’s how much I feel I sold Saber for, even though an extra £10,000 ended up in my business bank account.

So what now? The plan for the money is:

20% will be saved to pay my next corporation tax bill (it’s gonna be a doozy)
40% will be taken out in dividends to use as a deposit on a house.
40% will be kept to fund future projects.

I’ve already got my next project lined up, and those of you who were readers of the Saber blog may be surprised to know it doesn’t involve programming, and isn’t any kind of tech startup. Keep an eye on this blog if you’re interested to see what that is, and how it goes.

- Matt

August 2019 Income Report

2019-09-05T00:00:00+01:00

August was my first full month with Saber in “maintenance mode”, and it went pretty well. I had my first week working for FreeAgent on the 5th August, which I spent working in their offices in Edinburgh. Conveniently there was very little that I needed to do to Saber during that week. However, my first week working from home seemed to be more of a test, as bugs and support requests came in thick and fast. I think I spent at least half an hour every morning working on Saber that week. At least since then it’s all settled down quite nicely.

In terms of the figures, August was a bit of a shit one. I don’t know if there was an increase in cancellations due to the news that I’d no longer be working on Saber full time, or if it was just a coincidence, but there was a lot of churn.

The Figures

	August 2019	July 2019
Monthly Recurring Revenue	$4,030	$4,136
Paying customers	50	54
- Personal Plan	12	16
- Startup Plan	27	27
- Corporate Plan	11	11
Unique users on landing page	589	753
New Free Trial sign ups	11	19
Free Trial sign up rate
New Paying customers	1	0
Lost Paying Customers	5	1
Free Trial to Paying conversion
End to End conversion
Support Tickets Closed	8
Bugs fixed	1

Once more I’ve made some changes to the table. First, as I’m no longer taking a salary from Saber, I’ve removed the profit row. I’ve also added two new rows, to help me track how much work is involved when Saber is in maintenance mode. I’ll also be tracking actual time spent starting in September.

As I mentioned before, I had an influx of support emails around the middle of the month, and one fairly difficult bug to fix, but other than that, it was a fairly easy month.

Thankfully most of the churn was from the bottom plan, so revenue loss wasn’t so bad, and I’ve still managed to stay above $4k MRR. ~~Free trial signups were surprisingly bad, especially considering page views were up.~~ I forgot to exclude the blog-only traffic, so actually traffic was bad, and signups were just ok.

Revenue & Growth

Life at FreeAgent

Of course, I’m still very much in the “honeymoon period”, but my first 4 weeks working full time for FreeAgent have been amazing. I’ve been able to get really stuck into their frankly gargantuan code base, and already been able to deploy a bunch of fixes and updates. I’ve also been able to learn a load of new things, particularly regarding CD/CI and more advanced git stuff, ie: the kinds of things I didn’t need to know when working as a solo developer.

I’ve also loved being able to be just a developer. In the last four weeks I’ve had to do zero marketing, design, server administration, copywriting, or any of the other bullshit I have to do for Saber. What I have done is write code, read code, review code and learn about code. More importantly, even as a remote employee, I’ve always got people to talk to, to learn from, to ask for advice, and to offer my help.

Basically, it’s exactly the job I didn’t realise I was looking for, and I love it.

- Matt