UK tech industry pushing up salaries – but UI devs out of luck Despite job losses among vendors and high profile companies, those with IT roles in UK companies saw their salaries increase by as much as 30 percent in the past year, a survey has found. While other sectors of the country's economy struggle to achieve above-inflation pay rises, IT roles have remained relatively resilient, …
Looking at infrastructure roles, rates haven't really risen since 2007.
Wintel Engineer jobs at 3rd line Senior Level still being advertised at around the 45-55k mark which is LESS than what I was earning in that role in 2007.
There is a constant whinge from employers that everyone wants to contract or wants to retire early. OR the evil words "productivity"...but why would you? CEO's have got on average 25% rises/year yet a simple look at the Bank of England inflation calculator will show you what those roles should be paying.
There was an article the other day saying CISO's are leaving in their droves...again WHY would you take that responsibility on for £120k or similar when you know the CFO, whose telling you that you can't have the budget that you want is on multiple times of that.
There is a constant whinge from employers that everyone wants to contract or wants to retire early.
These will be the same reorganisation-happy halfwits who spent the last 20 or 30 years demanding that their staff are on call 24/7, do business travel in their own time, accept unachievable bonus targets in lieu of pay rises and so on. They treat their staff like mugs and are suddenly all surprised when nobody's all that keen on continuing to work for them. Funny, that.
But then some companies are really pushing out the boat to retain their talent.
One global embedded RTOS (deployed across the Solar System) company is giving employees two weeks paid leave (to coincide with US holidays, but hey-ho) on top of their ordinary holiday entitlement as they want to keep their employees fresh and interested. And pay rises have been inflation+ for many years.
That's likely because the "wintel" route is the most common route into tech. Windows engineers are ten a penny. Everyone coming out of University with a BSc in Computer science is essentially a Wintel engineer. My nephew just graduated, I helped him with his course, it was overwhelmingly Windows based. The content of the course was about one or two modules short of what used to be an MCSE...it fell short on network infrastructure and database management (what would have been the 70-216 "BEAST" and the 70-228)...anyone graduating wouldn't be able to pass those exams (or their modern equivalents).
Wintel work is getting less common as well...since most "classic" wintel customers (SMEs) have moved a lot of stuff to Office 365...it's become way less common to have an on-prem Exchange server for example...they might have an on-prem file and print server and that's it.
Some of the more niche businesses (like logistics) might have more complicated setups, loads of different printers, a database server with some legacy looking shitty front end and a bunch of warehouse kit...but your usual bread and butter firms that aren't very complicated...accountants, estate agents etc etc...the kind that deal in documents...they are likely to be mostly cloud now, with a local file server.
Outside of that, you might have a few WAPs, a router and a switch...up to maybe 100 workstations / laptops.
Most of the folks that used to look after these firms, which was a lot of people, are now competing for far fewer jobs...because it is a lot easier for a single person to look after more customers.
I started out in "wintel" in around 1999 (which I think was probably the golden era for Microsoft based work) and it wasn't uncommon for a business to have several servers back then. Even if they were relatively small...then we hit 2003 and SBS became a lot more common...quite a few businesses dropped down to just a single server...and here were are now in 2023...on the verge of your average business having no servers.
I started moving away from wintel around the SBS era...because I couldn't stand SBS...there was a period of time where businesses ran just SBS on an under-specced server. Most people here probably know the setup. I wanted out of Windows stuff anyway, because I wanted to get into Linux anyway, but Wintel was the path of least resistance to get into "the game" as it were.
8GB RAM, Quad Core CPU, RAID1 for Windows, RAID5 for file storage and LTO-1 for backups etc etc...those days.
Move into Linux and you'll see a dramatic pay bump...because the reality is, most experienced wintel engineers won't make the jump
For Windows work (as a freelancer) I might be able to charge £200 a day (and I've got got decades of experience)...on the flipside, installing an SSL certificate on NGINX in a Linux environment (which will take me about 10 minutes, if that), will also bag £200, but it won't take me a day.
In fact anything related to NGINX...to this day, it seems most professionals still give NGINX a wide berth for some reason in favour of Apache...most Pros I've spoken to on the matter seem to think NGINX is way harder to work with than Apache...which in my experience, just isn't true...I think there is just a general perception that because NGINX works differently, it must be harder to use...but those of us that do use NGINX, know that it isn't that far detached from Apache.
Load testing cloud infrastructure also drags in quite a bit of money. All those folks using outsourced developers with no QA, eventually reach a point where their system underperforms for some reason or another...so they ultimately get someone in to perform load tests to figure out where the problem lies. A day spent analysing a website for slow loading elements and reviewing the code, database queries etc etc can get you £1000+.
Then comes the consultancy to oversee the devs in implementing proposed changes...that is easily £500 a day.
You need to look for areas of tech that you can dramatically reduce overheads in, that's where the money is. I can usually bring the costs of Cloud infrastructure down by at least 50%...my average client usually spends in the region of £50,000-£100k a year in cloud costs...because they over scale to compensate for crappy code / shabby database optimisation and poor decision making on the types of cloud infrastructure they actually need.
Just finding "that one massive query" can allow a business to drop down an RDS tier on Amazon quite easily. Shave £25k off a £50k bill and suddenly your rates don't look expensive. They're essentially getting you for free. Potentially, most of what they pay you, they could recoup in the first bill after you optimise things.
This all said, to get yourself into these sorts of positions, and effectively negotiate, you need to think less like an engineer and more like an executive...because you're not there to produce the latest shiny, you're there to help optimise the business.
Anyway, TL;DR, wintel is being cannibalised by Microsoft, there are tons of Wintel engineers, Linux is where the action is.
you need to think less like an engineer and more like an executive...because you're not there to produce the latest shiny, you're there to help optimise the business.
100% agreed. You need to think in terms of money and how you can add value.
There are tons of highly skilled software engineers - many of whom are probably Reg readers - who seem to fail to make big money. But it's because they can only articulate things in tech terms, which the people paying for services frankly know little about.
Why are they interested in hiring you? To reduce their costs, or help increase their revenue. There aren't any other reasons.
The case about finding a slow query is a great example. If it's an online business you might charge £1k to find and optimise that. They might save 10 times that in what would otherwise be lost orders from people abandoning their site because of how unresponsive it was. You need to explain it in the latter terms and stop going on about some bullshit in MySQL (or whatever they use) that they know nothing about. It's as simple as "you're losing money, and I can help stop this... for a handsome fee... which is still significantly lower than what you'd lose otherwise".
"The only role category to see earning fall was mid-level UX and UI designers"
Hmm, while good UI and UX are indeed very important things (as Apple back in the 80s once knew), could this possibly be related to the fact that many of the self-proclaimed current emperors usurping in these roles, wearing their magnificent new clothes, have finally been seen through as the empty bluster and bullshit that they are, and everyone is getting heartily fed up with (un)discoverable UIs, the removal of menu bars, garish and horrible flatso 'design', etc, etc?
having UI design on embedded systems being one of my least favorite tasks, and doing cool device control things being a LOT more interesting, I'd have to conclude that UI design (I never figured out why anyone calls it 'UX' unless they are from Marin county, related light bulb joke below) doe not require nearly the talent nor competence of "other kinds of coding". Or, how a kernel + device control C programmer like me can easily do the UI design also, given a search engine for docs and "how do I do this in XXX", but the converse is most likely NOT the case without a lot of low level coding experience.
That being said the UI with a web interface can be the fastest and simplest solution, especially for an RPi with a touch screen in 'kiosk mode'. The hard part is the architecture on the system side, aka how do you turn button clicks into "things happening". Lots of solutions for that exist, with PHP 'shell_exec()' being at the heart of it.
If you specialize in UI design you should probably learn how to do OTHER things too, from SQL to device control. An RPi or Arduino can help with that. Design a vending machine maybe? Or a retro game console? You would learn a lot of useful things, and learn where to get the necessary parts.
At that point you would be hired to do the harder stuff but could assist with or even take over the UI part later (or design the glue that makes it all work).
Light bulb joke: How many people from Marin County does it take to change a light bulb? At least 3. 1 to change the bulb and the rest to "share in the experience"
(which reflects what I think of the term 'UX', yeah)
> (which reflects what I think of the term 'UX', yeah)
The only valid "User Experience" (barring certain video games whose purpose is to have an "experience") is: None At All. After a (hopefully a quick and easy) learning period, if I am "experiencing" your program instead of my workflow, then your program has failed.
I have stuck to one keyboard layout and one text editor since the early 90s because, for all the basic second-by-second operations, I can drive both without any conscious thought. I don't expect any other programs to be that enjoyable to use, but it defines the goal: just get out of the way and, when you really can't (perhaps because I've not yet learnt that feature) be clear, discoverable and as blindingly obvious as is possible![1]
Anyone else who tries to explain why their UX is going to "kindle joy" had better have a high tolerance for invective (and preferably to do so in a public place where one's courtesy to innocent bystanders will be a restraint).
[1] anything I can click must be *obviously* and consistently demarcated, with tooltips when I hover[2] (preferably with a "do you want to know more?" link into the documentation sitting in the same directory as the executable. If you *have* to have menus (and not all do) have menus you can walk through using the keyboard; with unambiguous textual descriptions, not random icons chosen solely because they don't look the same as a competitor's icons - and tooltips on the menu items, even - especially - the greyed-out ones so you can find out *why* they are greyed-out! Shower scorn on any GUI toolkit that doesn't automatically let you have all those menu tips (it is a pain under Windows but that is why you used a toolkit).
[2] yes, this does mean a pointer and hence mouse or equivalent - and no, once we get beyond the simple calculator app, I've yet not found a phone/tablet app that is up to snuff: hold your finger down and try to wobble, maybe you'll get the "cut/copy/paste" popup menu this time is - aaaargh[3]
[3] my lawn, bugger off, you know the drill by now
1a: Grey out inactive options, DON'T REMOVE THEM FROM THE DAMN INTERFACE!!!! I have mother***** had it with these mother****** snakes programs that remove options they think you don't want to know about, so it's impossible to discover that they are there until you have done some unexplained esoteric operation that magiclly adds the option to a menu.
It is easy, relatively speaking.
Someone specialising in UI/UX was far more limited in what they could do 10-20 years ago than they are now.
There was a time where UI was mostly dominated by Frontpage / Dreamweaver people and it was far less granular...
Just look how far UX/UI has come in terms of websites.
I built my first site in 1995...and I've pretty much kept up with each change in the "meta" between then and now and we seem to be on course to go full circle.
This is a very rough outline of how things have moved, at least for me...I haven't put the years in, because I can't accurately remember specific years I jumped from one thing to the next...but earlier on, it was shorter time frames.
1. Monolithic, pure HTML. No real layout structure. Links at the top and bottom of each page. Hella static.
2. iFrames...menubar in the left iframe. Separate pages. Static.
3. IFrames with Tables. Separate pages. Whole page. Semi-dynamic. CGI. Fucking Java Applets.
4. Tables with CSS. Separate pages. Whole page. Dynamic server side. PHP. Flash is now a thing, but I don't use it.
5. Tables with CSS and JS. Separate pages. Whole page. Dynamic. Server side. PHP.
6. Divs and Spans with CSS and JS...UI frameworks begin (Bootstrap, Angular, React). Monolithic, pretty much one page. Mostly static, dynamic elements.
Even though as time goes on, the sites become more complicated, the route to getting there has become a lot less clunky, and inefficient...a rough overview of the "setup" at each stage as per my very rough memory...not complete, but you'll get the gist.
1. FTP client, Notepad. Basic folder structure, painstakingly coded by hand...slooooooooooooooooooow!
2. FTP client, Notepad. Maybe Frontpage / Dreamweaver with integrated FTP, WYSIWYG...slooooooooow!
2.2 Geocities bitch!
3. FTP client, Notepad. Maybe Frontpage / Dreamweaver with integrated FTP, WYSIWYG...fiddly CGI-BIN...sloooooooow and annoying.
4. Seamless SFTP mount, Netbeans (or equivalent) or just Notepad (or Linux equivalent)...much faster to iterate. Data in flat files.
5. SVN, Netbeans (or equivalent) or just Notepad (or Linux equivalent)....much faster, data now in a database...bit more fiddly.
6. Git, Sublime / Atom / VSCode or Nano...tools like Bootstrap Studio for semi-WYSIWYG...themeforest exists, no need to design a UI from scratch when you can buy one that is almost there for peanuts then adjust it....
...and now, time required to get a decent finished result then the work required by the "UI" guy...
1. A month for a good sized site. Plus hours of regular work to keep the site "updated" and adapted to new content (major UI changes here and there, browser wars). Upload, test, repeat methodology.
2. A few weeks, plus, hours of regular, but less so, work to keep the site "updated" and adapted to new content (minor UI changes here and there). Upload, test, repeat methodology.
2.2 Sign my guestbook. Hit counters!
3. Two weeks. Update frequency is still high, but each update requires less time and UI tweaks becoming less common.
4. Under two weeks. UI tweaks even less common.
5. Less than a week. Some updates automated, user generated content. Less time spent updating. UI tweaks a few times a year.
6. A couple of hours. Buy a template, edit it and extend it, drop in the dynamic components...dedicated UI guy largely irrelevant.
We have essentially gone from UI being the bulk of the work, to being something you can quite easily pick up pretty much ready done for less than £50.
That's not to say that UI isn't important, it is, but it's become something that doesn't necessarily require loads of dedicated folks.
Some of the fancier web teams out there might have a dedicated UI person or two, but your typical web shop doesn't.
I think this is mostly because most of what a UI person would have done in the past has been split up and assigned to different people. Content management for example, Dorothy on reception can now do the blog updates etc via a nice web ui, that looks exactly like 100s of other webuis...so the front end guy no longer gets involved in content.
Content has also become dynamic, so there are less UI tweaks need to be made on the fly, developer has that locked down.
Essentially, once a site is up and running...that's it. The UI guy doesn't have a lot to do. Because these days, you can't make dramatic changes on a production site with high volumes of traffic without pissing people off.
In the first 10 years of the web, things moved at breakneck pace...new, whacky shit was happening all the time...because it could, not many businesses relied on the web for business...today, many businesses live and die on the internet...and making changes too frequently can have a massive knock on effect with your customer base. Businesses are less inclined to rock the boat as it were.
The golden age for UI/UX guys was around 2005-2015...the ride is slowing down...and as UI frameworks get easier to use and more homogenised and standardised, UI/UX specialists will become relatively niche.
Personally, as a developer / infrastructure guy...I've always liked having a UI/UX guy to shit on and divert attention to. You're the unsung heroic punching bags of software development that allowed developers to get a breather every now and then while you argued about the company fonts, logo usage guidelines and endless meetings about "story telling", "wire framing" and "user journeys"...but I think the sun is setting on those days...at least for permanent in house roles...you'll always exist as consultants / freelancers...and you'll be all the better for it. <3
But it is easy compared to what it was say 10-20 years ago.
It's possible to throw a prototype together in under an hour now to communicate your ideas...that wasn't possible 10 years ago.
As a result, UI/UX has become a lot easier to plan and implement than it was once...and because of that...cheaper.
You don't need to spend weeks mocking things up in Illustrator / Photoshop anymore...you can just use Sketch Up or something like that...shaves off massive amounts of time...as with most things in tech, yes the pay as a permie is dropping...because the actual role has become cheaper...but as a freelancer, because things are easier, you can have more clients...
So swings and roundabouts...freelance UI/UX folks probably do extremely well because they can juggle far more clients than has ever been possible before, despite charging a bit less.
As a permie, you have one client paying you your salary...a freelancer can have 50 clients paying a grand a piece per year.
Your employer can either keep you on at £30k a year, or pay a contractor £1000 for 4 visits a year to get the same result and have money left over for an extra junior dev.
UI/UX changes aren't that common, you might have to revise things 3 or 4 times a year at most and have a major refresh every 2-3 years to keep things fresh.
Tech is slowly, but surely moving to a more freelance style setup because as tech rolls on and usability increases (thanks to UI/UX guys), you need less dedicated staff.
UI/UX is one of those roles that if you're really good at it, you create less work for yourself...you're better off being mediocre than really good, otherwise it's like you're just building your own coffin.
You mean the ones Microshite don't employ? After all they keep effing up the UI in Windows and other apps everytime they touch them. Almost always to make them far, far worse.
Why? To justify their existence?
They are sort of the IT equivilant of Uncle Ernie in Tommy - they just have to "fiddle about"
The problem Microsoft has is legacy backwards compatibility. They have got themselves into a situation where they can't overhaul certain components / UIs without alienating a lot of people or breaking a lot of tech. It's why the control panel is such a fucking mess and why certain things in Windows are slow as balls.
Not to mention the whole smattering of tools that exist in automated scripts and the like to workaround shitty problems of yore...
Anyone remember a tool called "click yes" from about 15 years ago? I do...I still fucking hate it, and I hate that it ever existed.
For the uninitiated...
https://www.contextmagic.com/express-clickyes/
That shit was everywhere for years...especially places that had Blackberry servers. Remember those??
If someone came along with an Exchange integration the way BES worked today, we'd laugh them out of the room.
Again, for the uninitiated...Blackberry server, for a while at least, worked via an Outlook MAPI connector...you had to create a single account that had full access to all the mailboxes you wanted to link to Blackberry devices and the BES software would monitor the mailboxes through this one account and "send as" the appropriate user...shitty right? :)
Because interfacing with Exchange directly so fucking opaque (it probably still is) that was the only way to pull it off. The other option was to use the sync platform provided by your mobile provider...which connected via POP3...yes POP3...to the exchange server to get your email and some providers didn't give you the option to "leave copies" on the server. So if you didn't know what you were doing, you could end up clearing a load of mail (up to 3 months worth) from a users mailbox into their Blackberry with no straight forward way to move it back. You could do it, but it was a fucking chore.
Blackberry was probably great if you were the end user, but it was a fucking nightmarish hellscape for the people that had to support it...this is just one example of how Microsoft buttfucked itself by being opaque and leaving legacy components lying around.
It did slightly improve further down the line, but by then, Blackberrys were basically dead.
That whole era was littered with shit like this...circa 2003-2006 was a bad time to be a wintel engineer...but it was also the same time that having servers in your offices started picking up steam...so a lot of bad habits were formed, loads of shitty macros and scripts were built that users got used to...and now here we are, stuck with the fallout still manifesting as shit UI...it's one of the curses of proprietary tech. They build something everyone wants, keep it closed off which forces people to extend it via workarounds, people build solutions around it and boom, every new version has to be able to support those older "workaround" setups or you lose customers...eventually, you end up with a system full of security problems, that looks like a pig with lipstick...users start to hate it, so you look for alternatives, there usually is an alternative that does exactly what your workaround does...almost...so you switch, implement another small workaround to make up for the shortfall, the original software company dies, and your proprietary journey starts all over again.
It'd be amazing if we lived in a world where UI/UX would continually improve and draw in more customers and just make things better overall...however, we live in a world where some people write scripts to click buttons (VBA devs, I'm looking at you...with your fucking Excel and Sage to Access to SQL and back bullshit), because there is no other way to automate it other than via the UI, and if that button shifts a couple of pixels, its name, label or its hook changes, the script breaks...this can lead to customers considering whether they need this tool in the first place or whether, after a couple of years, something better exists...which it usually does...but then you're up against old Betty in finance who has got used to the way things work and she doesn't want to learn something new...she works 3 days a week, so it's not worth overhauling the system...then the UI changes anyway, which breaks a load of shit...but *phew* they still have "classic" mode as an option...which everyone switches to, even though the new UI is leaps and bounds better...but because the new UI leads to so many breaking changes with workflows that people are used to...it's widely regarded as "shit"...so they keep the classic look forever...which leaves systems administrators flipping between the new UI and "classic" because all the new features and settings are only accessible via the new UI, and the "classic" settings and features are in the old UI.
You can see this in action on Office 365...you can access both the classic Exchange UI and the "New" Exchange UI...classic still exists because there are settings in there that you just can't access via the new UI and vice versa.
It's apparent in Windows 11 with the weird "4 layers" of UI that you can see if you dig deep into the control panel.
Even the original Windows 2000 icons still exist in Windows 11, because that icon Betty uses to trigger a script needs to be that fucking yellow star...because that's what she's used for the last 20 years...if you don't use that icon, she won't remember which icon to click.
Anyway, rant over...UI/UX...yes it's important...but is it essential? No.
2023 and companies are still acting like they're doing us a favour. No, I work for my own benefit. I care professionally about doing a good job but if a company is doomed through poor management, staff "abuse" or simple plan market forces, then time to move on.
Sorry but IT is like a slower version of being a premier league player, you only got so long to work this game before you start to lose your edge and it gets harder to keep up, say around age 55. After that I want to slow down or simply bugger off to happy retirement, sorry but retirement is "look after number one" and screw everyone else, make your cash as fast and as much as you can, then you have control and decide what, where and how you work if you still want to.
>but retirement is "look after number one" and screw everyone else,
That could be a good IR35 test.
Are you doing this to do a good job, benefit the company and enhance shareholder value? Then you are a hidden employee and should be paying full tax and have your ears chopped off (whatever HMRC current punishment is)
But if you claim you are just out to screw the customer out of every last penny and make as much as possible and damn the consequences = then you are behaving like a proper company and get to keep your contractor status.
The problem with being over 50 is that you have enough experience to realise that 80% of everything is crap, the latest fad is not going to change the world (probably), and anyway won't do so next week.
The thing to do is not say these thoughts out loud. Dye your hair, be enthusiastic when the latest craze is being mooted, collect your cheque, and laugh about it at home or at the pub/golf club/tennis club/morris dancing club event.
I don't agree at all. If you do the same old job for years and don't push yourself to learn and develop professionally, then it's no surprise you're not in demand. But that demand is entirely a function of your initiative and committiment. I know this because I've just been laid off (I'm 51) and have never been in such demand, I've also never been offered so much money.
I was a wintel infrastructure engineer with a bit of software engineering thrown in for about 15 years. I saw the demand was declining and what the market was doing, then I stuck my head down and relentlessly worked on renewing my skills in cloud, software engineering, design patterns, cloud security, software architecture, and cloud infrastructure. I'm now a very experienced and competent cloud architect.
It's all down to you as an individual - put the effort in to be at the top of your profession, or don't bother. Claiming you are the victim of ageism simply means you are not prepared to put in the huge effort that it takes to be better than your competitors. I have never seen ageism in tech (ever!), but I have seen many, many older techies give the advantage to young, enthusiastic people who are more flexible and open to change. These younger people are driven, they are committed to learning, and they are adaptable.
It might sound harsh, but if you aren't valued by potential employers, it's because you didn't make yourself valuable enough.
Because the UK sneers at technical skills and pays us dirt, sadly.
(And then wonders why as a country we don't have a productive economy. But, never mind, there's always room for plenty of better paid pencil pushers, sorry, manglers, sorry, managers… And too often what seems to appeal to their egos is "Look how many £Ms I wasted spent on buying in product/lock-in 'service' [whatever] from a US company" (which, too often, ends up being a square peg fitting badly in a round hole) rather than "For only £xM, our developers built a rival product that is just as good, and we now have a market to sell it to, as well".)
Because the UK sneers at technical skills and pays us dirt, sadly.
And then wonders why as a country we don't have a productive economy.
True. But there's actually a much bigger reason. In the UK - at the time of writing - you're pretty much in either one of these two categories (especially when it comes to property).
1. You have rich parents, who helped you get on the property ladder. You had/have some form of inheritance.
2. You don't.
If you're in the former category there's very little incentive to earn more. Because you're always going to be better off than people in the second category.
If you're in the second category you have little incentive to earn more because you're probably never going to be able to retire.
The concept of having a career and reasonable pension provision of the prior to 1990s / early 2000s just doesn't exist anymore. This creates a system where essentially nobody gives a fuck about money as such and that's why "we don't have a productive economy". Because you're either fine financially, or never will be. People - especially under 30's - have wised-up to the fact that working their arses off isn't necessarily going to change any of this. Exacerbated by the fact some teenagers make millions from being social media "influencers".
Sounds like envy. Influencers are a tiny percentage of the overall work force and most people don’t give a toss about what they do.
Your analysis is likely overly simplistic. The other day when banks offered 100% mortgages as a solution to the deposit problem , some dick head said they can’t buy a house because one isn’t available local to them. That’s a massive fail and shows a cringey level of entitlement.
Under 30s are more than capable of being crippled by mortgage debt just like over 30s.
Because the UK sneers at technical skills and pays us dirt, sadly.
I've just been emailed this job:
"Senior Business Support Officer, playing a crucial role in ensuring smooth operations and providing exceptional support to our clients and internal teams. Your dedication, attention to detail, and strong organizational skills will be vital in maintaining the high standards we set for ourselves.
£10.90 per hour"
This is 45p more than minimum wage.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
