Almost every article on changing jobs takes the hard part for granted, namely whether to change, and jumps straight to the easy part: the CV, the interview, the notice period.
This guide does the opposite. It starts from two research findings about the decision rather than the procedure, and they point in opposite directions. Only then does it get to the operational part, including the checks almost nobody spells out.
Two findings that contradict each other (that’s the point)
The first: we’re too conservative. In 2021 Steven Levitt published an unusual study in the Review of Economic Studies. He had recruited tens of thousands of people stuck on a major decision, including “should I quit my job?”, and let a coin toss decide. He then followed up at two and six months. Those assigned to change reported higher wellbeing, and the difference was larger at six months than at two. It isn’t a perfect experiment (participation is self-selected and not everyone follows the coin), but the direction is clear: when a choice feels like a tie, we systematically stay, and we regret it.
The second: we decide at the wrong moment. Lee and Mitchell, in what became known as the unfolding model of voluntary turnover (Academy of Management Review, 1994), showed that most resignations don’t come from slow, reasoned dissatisfaction but from a shock: a new manager, a missed promotion, a reorganisation, a colleague leaving. The shock doesn’t create the dissatisfaction, it makes it visible. And it pushes you to decide fast.
Put together, the two findings give one practical rule:
If you’ve been weighing it up for months, you’re probably underrating the change. If you’ve been weighing it up for three days because something happened, you’re probably overrating it.
The three axes almost nobody separates
“Changing jobs” isn’t one decision. It’s three independent ones, and most changes that go badly moved all of them at once.
| Axis | What changes | Risk | Reversible? |
|---|---|---|---|
| Employer | Same role, different company | Low | Yes, nearly always |
| Role | Different role, same sector | Medium | Yes, with a year’s work |
| Sector | Same role, different industry | Medium-high | Hard to go back |
| All three | Full reskilling | High | Very hard |
The ordering matters because your credibility on the market is made of verifiable evidence. Change one axis and everything else keeps serving as evidence. Change all three and you arrive with no evidence in a context where nobody knows you, which is the worst position of all.
So the right question isn’t “do I want to change jobs?” but: what’s the smallest thing I have to change for the problem to disappear? In most cases the answer is “one thing”, and often not the one it first appeared to be.
A recurring example: people who are exhausted think they need to change industry, when the problem is workload and control over their own work, two things that can change by changing employer alone. If that sounds familiar, the diagnostic part is in burnout: recognising and preventing it.
The cost nobody calculates
Three items that usually don’t make it into the maths.
Notice periods aren’t symbolic. In many countries the length is set by collective agreement or statute and varies a lot by seniority; failing to serve it can mean the employer withholds the equivalent amount. It’s the first number to recover, and it’s written in your contract.
Unemployment benefit usually doesn’t cover resigning. Across most European systems, unemployment insurance presumes involuntary job loss; voluntary resignation is excluded except in defined cases (constructive dismissal, protected periods, and similar). The exact rules are national, and this is one of the few things worth checking on the official site before, not after.
“Plus X gross” isn’t what arrives. A gross increase thins out considerably once tax and local levies are applied, and it has to be netted against what you lose: benefits, meal allowances, bonus schemes, commuting time. The method for the calculation is the same as for the internal negotiation, described in how to ask for a raise. Run it before resigning, because sometimes it shows the move doesn’t pay.
Rules are national: where to check
This is where generic articles go most wrong, because they copy outdated information from a different country.
Resignation formalities differ substantially. In Italy, for instance, resignation from an employment contract must be filed electronically through the Ministry of Labour’s procedure (Legislative Decree 151/2015) and can be revoked within seven days; a handwritten letter alone has no effect. In Spain, baja voluntaria requires the notice set by the applicable convenio colectivo and, as a rule, does not open entitlement to unemployment benefit.
Two consequences:
- Check the source, not the article. For your own country, start from the national labour ministry and the body that pays unemployment benefit. For moving between EU countries, the reference portal is EURES and the rules on coordination of social security are on the European Commission site.
- Get the notice period from your contract, not from a forum. It’s the single number that determines the whole timeline.
The operational sequence, in order
Nine steps, in the order that minimises risk.
- Write the problem in one sentence. If you can’t, it isn’t time to move: it’s time to understand.
- Identify the minimum axis (employer / role / sector) that resolves that sentence.
- Verify actual demand, not perceived demand. Useful public data: the Cedefop skills forecasts, Eurostat for employment by sector, and the World Economic Forum Future of Jobs Report. The summary picture is in which jobs are most in demand.
- Close the smallest gap between you and the target role, not all of them. The method for doing it verifiably is in learning a skill in 30 days.
- Rebuild the CV around results, not duties: how to write an effective CV.
- Fix the public profile before going active, including notification settings: optimising your LinkedIn profile.
- Prepare answers as episodes, not as adjectives: interview questions.
- Negotiate on the written offer, never on the verbal promise. Ask for level, salary, variable pay and working arrangements in writing.
- Resign only with a signed contract in hand. And serve your notice: your sector is smaller than it looks.
The two most expensive mistakes
Leaving something instead of joining something. It’s the signature of the shock Lee and Mitchell described: the decision gets framed as “I’ve had enough”, and the selection criterion for the destination becomes “anywhere else”. The test is simple: if you can only describe the new job by contrast with the old one, you haven’t chosen anything yet.
Treating the counter-offer as a win. When you resign and the company comes back with more, it’s solving its own problem: replacing you costs money and time. If what they’re offering now was possible before, the question is why it didn’t arrive before. Sometimes the counter-offer is legitimate and should be accepted, but it has to be assessed as a new offer, with its own timeline, not as a repair.
My take: looking and resigning are two separate decisions
I think the most common piece of advice, “never leave a job without another one lined up”, is right in form and often wrong in use, because it becomes an excuse not to take even the first step. The version I’d use is different: don’t resign without another signed contract, but don’t wait for another contract to start looking. They’re two different things, and conflating them is the most common way to stay put for years.
The less diplomatic part: the cost of changing is visible immediately, the cost of not changing isn’t. It’s spread over years, has no date and no culprit. Levitt’s result says exactly that, that our indifference point is miscalibrated, and it’s worth remembering when we feel we’re deciding “rationally”.
Frequently asked questions
How long should I stay before changing? There’s no threshold. The useful criterion is the overall record: one or two short stints in a career explain themselves in thirty seconds; a sequence of eight-month tenures needs an explanation that holds up, and it will be asked about at interview.
Should I tell my manager I’m looking? No, and there are visibility settings to check before you start, not after. Communication happens when you have a signed contract.
Does changing sector at forty make sense? Yes, but one axis at a time. The most robust path is usually same role, new sector, so the experience stays usable while you learn the context. Only afterwards, if needed, a different role.
If I take a pay cut at first, did I get it wrong? Not necessarily. Economic research on mobility shows a substantial share of wage growth, especially early in a career, comes from changing employer rather than from internal raises. A short step back can be the price of moving one axis well, as long as it’s a decision and not something you just absorb.
How do I tell burnout from simply being in the wrong job? The practical test is recovery: if a real break returns you to normal, it’s load; if it doesn’t, changing employer won’t fix it.
Sources
- S. D. Levitt, Heads or Tails: The Impact of a Coin Toss on Major Life Decisions and Subsequent Happiness, Review of Economic Studies, 2021. Participants randomised into change reported higher wellbeing at two and six months, suggesting a systematic bias towards the status quo.
- T. W. Lee and T. R. Mitchell, An Alternative Approach: The Unfolding Model of Voluntary Employee Turnover, Academy of Management Review, 1994. Resignations are often triggered by a discrete shock rather than gradual dissatisfaction.
- T. R. Mitchell, B. C. Holtom, T. W. Lee, C. J. Sablynski and M. Erez, Why People Stay: Using Job Embeddedness to Predict Voluntary Turnover, Academy of Management Journal, 2001. Embeddedness predicts staying beyond stated satisfaction.
- R. H. Topel and M. P. Ward, Job Mobility and the Careers of Young Men, Quarterly Journal of Economics, 1992. A substantial share of early-career wage growth attributable to changing employer.
- EURES. European job mobility portal, including national labour law and living and working conditions summaries.
- European Commission. Coordination of social security systems for people moving between member states.
- Cedefop. European skills forecasts and occupational demand.
- Eurostat. Employment, tenure and labour market transition statistics.
- World Economic Forum, Future of Jobs Report 2025. Expected shifts in occupations and skills by sector.



