Beermat Software All articles
Startup Strategy

Gone by Lunchtime: Why Your Best Developer Told LinkedIn Before They Told You

Beermat Software
Gone by Lunchtime: Why Your Best Developer Told LinkedIn Before They Told You

Photo by Photo by Bench Accounting on Unsplash on Unsplash

It usually happens on a Monday. You open LinkedIn with your morning coffee, as one does, half-expecting the usual carousel of thought leadership and people announcing they are 'humbled' by something. Instead, you see it. Your senior engineer — the one who knows where all the bodies are buried in the codebase, the one you were planning to promote next quarter — has posted that they are 'thrilled to be starting a new chapter' at somewhere that is not you.

You were not consulted. You were not warned. You found out the same way their university flatmate found out. Via a post with a rocket ship emoji.

This is the new shape of developer attrition in Britain, and it is happening quietly, repeatedly, and at significant cost to companies that thought everything was fine.

The Silence Before the Rocket Ship

The conventional wisdom around staff retention focuses on exit interviews — that awkward forty-five-minute conversation where someone who is already leaving tells you carefully edited reasons why they are leaving. It is, to be frank, nearly useless. By the time you are conducting an exit interview, you have already lost.

What is less understood is the period before the LinkedIn post. The months — sometimes years — during which a talented developer has mentally checked out while physically remaining at their desk. They are still shipping code. They are still attending standups. They are also updating their CV, doing take-home tests for other companies at weekends, and making a considered, rational decision that you are not party to.

Talking to developers across the UK — from Bristol to Manchester, from Edinburgh to London — a pattern emerges. The decision to leave is rarely sudden. It is the product of accumulated small disappointments, most of which were entirely preventable.

What Actually Drives Them Away

Spoiler: it is not, primarily, money. Or rather, money is rarely the first cause. It becomes a cause when someone else offers more, at which point it acts as the final push on a door that was already open. But the door was opened by something else.

The work stopped being interesting. This comes up constantly. Developers are, at their core, problem-solvers. The moment the problems stop being interesting — when the job becomes maintaining a legacy system with no greenfield work in sight, or when every interesting technical decision gets overruled by someone who last wrote code in 2009 — the psychological exit begins.

Nobody asked for their opinion. There is a particular frustration, articulated with remarkable consistency, around technical decisions being made without technical input. Architecture choices, tool selections, infrastructure decisions — made in meetings that developers were not invited to, then handed down as done deals. 'I found out we were migrating to a new cloud provider in the same all-hands where they announced it to the whole company,' said one backend developer who recently left a mid-sized fintech. 'I'd been there four years.'

The process became the product. British tech companies have, in many cases, imported the worst of Silicon Valley's meeting culture without importing any of the good bits. Developers who want to build things find themselves in sprint planning sessions, backlog refinement, story point poker, and retrospectives about retrospectives. The ratio of doing to discussing tips the wrong way, and the people most affected are the ones who came to do.

Career progression was invisible. Not absent — invisible. Many developers cannot tell you what the path forward looks like at their current employer, not because there isn't one, but because nobody has ever sat down and explained it clearly. Ambiguity about the future is a surprisingly powerful driver of attrition.

What the LinkedIn Post Is Actually Saying

When a developer announces a new role on LinkedIn before telling their manager, it is not rudeness. It is, in its own way, a data point. It suggests the relationship with the employer had already deteriorated to the point where professional courtesy felt unnecessary or even uncomfortable. It suggests the developer did not feel safe having an honest conversation about their future. It suggests, in short, that the exit had already happened emotionally long before it happened officially.

The LinkedIn post is not the problem. It is the symptom of a culture in which developers do not feel heard, valued, or safe enough to say 'I'm thinking about leaving — can we talk?'

What the Good Ones Are Doing Differently

The British tech companies with genuinely low developer attrition are not, in the main, the ones paying the most. They share a different set of characteristics.

They have regular, honest one-to-ones. Not performance reviews. Actual conversations, initiated by managers, about what the developer wants from their career, what they find frustrating, and what would make their job better. Boring? Yes. Effective? Extraordinarily.

They protect maker time. Deep work — the kind that makes a developer feel they have actually accomplished something — requires uninterrupted blocks of time. The companies that understand this structure their calendars accordingly, cluster meetings, and treat a developer's focus time as a resource worth protecting.

They involve engineers in decisions that affect engineers. Revolutionary stuff. When a technical choice needs to be made, the people who will live with the consequences are in the room. They may not always get what they want, but they were asked, and the reasoning was explained.

They talk about the future explicitly. 'Here is where we see you in two years. Here is what we would need to see from you. Here is what we are building toward.' Clarity about trajectory is remarkably cheap to provide and remarkably effective at making people stay.

The Cost of Getting This Wrong

The average cost of replacing a mid-level software developer in the UK — accounting for recruitment fees, lost productivity during the notice period, onboarding time, and the institutional knowledge that walked out the door — sits somewhere between £30,000 and £60,000. For a senior engineer, higher.

That is not a people problem. That is a financial problem with a people-shaped cause.

The developers are not being unreasonable. They want interesting work, genuine input, and a manager who knows their name and occasionally asks how they are getting on. In return, they will build your product, maintain your systems, and — if you are very lucky — tell you when something is going wrong before it becomes a crisis.

Or they will tell LinkedIn. The choice, largely, is yours.

All Articles

Related Articles

Your Napkin Knew Best: Why the First Scribble Beats the Fiftieth Slide

Your Napkin Knew Best: Why the First Scribble Beats the Fiftieth Slide

One Page, No Nonsense: The Radical British Art of Keeping It Stupidly Simple

One Page, No Nonsense: The Radical British Art of Keeping It Stupidly Simple

Charging Peanuts for Gold: The Quiet Financial Tragedy of British Indie Software

Charging Peanuts for Gold: The Quiet Financial Tragedy of British Indie Software