Most people wait for the raise to arrive on its own, as recognition of their effort. It usually doesn’t, and not out of malice: an organisation adjusts pay when it has a reason to, not when someone deserves it in a moral sense.
This guide treats the request as what it is, a negotiation, and tries to give you the three things usually missing: the public data to anchor the request, the rights EU legislation is granting you right now, and the arithmetic of what a gross raise is actually worth in your pocket.
Two numbers that change the conversation
Worth starting here, because it shifts the request from “I want more” to “there’s an objective problem”.
First figure. OECD international comparisons of average annual wages show Italy as one of the few advanced countries where real wages have stayed broadly flat over a multi-decade horizon, while most other countries saw growth. That isn’t a political opinion: it’s a time series.
Second figure. ISTAT regularly publishes an index of contractual pay, measuring how much collectively bargained minimums grow. In recent years contractual pay growth has often lagged inflation, which means something specific: if your salary has only grown through contract renewals, your purchasing power has in all likelihood fallen.
That’s a usable premise for the conversation, and it isn’t aggressive: you’re not accusing anyone, you’re observing that automatic adjustment hasn’t kept pace. It’s a much stronger opening than “my rent went up”.
What you can ask for by law (and it’s changing now)
The EU Pay Transparency Directive 2023/970 is the most significant development on this topic in years, and almost nobody talks about it. The transposition deadline for member states was June 2026: check how far it applies in your country. The points that concern you directly:
- Right to information when applying: candidates must be able to know the starting pay or the range for the position, before the interview or at any rate before negotiating.
- Ban on asking for pay history: an employer can’t ask what you earned before. This matters because pay history is historically the mechanism that perpetuates disparities.
- Right to information once employed: workers can request information on average pay levels, broken down by sex, for categories of workers doing the same work or work of equal value.
- Pay secrecy clauses: agreements cannot prevent workers from disclosing their own pay.
That last point deserves a practical note: many people believe discussing salary with colleagues is forbidden. In most cases it isn’t, and the directive explicitly strengthens that position. Knowing what people doing your job at your company earn is the single most useful data point you can have, and it’s now harder to stop you getting it.
Do check the state of transposition when you read this: until national rules are in force, exercising these rights depends on the existing framework.
Building a market figure that holds up
“I read online that people earn more” doesn’t hold. You need a source your counterpart can verify. Three tiers, from most to least solid:
| Source | What it gives you | Limitation |
|---|---|---|
| Your applicable collective agreement (CNEL holds the official Italian archive) | The minimum for your grade and the correct grade for your duties | It’s the floor, not the market |
| Public statistics (ISTAT, Eurostat) | Average pay by sector, qualification, region | Aggregated: doesn’t capture your specific role |
| Private salary surveys and job ads | Ranges for comparable roles | Often opaque methodology, self-selected samples |
The move almost nobody makes and that’s worth the most: check your job grade. It frequently happens that duties have grown and the grade hasn’t. If you’re consistently doing work that your collective agreement places at a higher grade, you aren’t asking a favour: you’re flagging a contractual mismatch. That’s an entirely different conversation, and much harder to dismiss.
The arithmetic almost nobody does: gross versus net
Here’s a piece of reality that changes strategy and rarely appears in articles on this subject.
Italy’s tax and social contribution wedge is among the highest in the OECD. Translated: of a gross raise, a substantially smaller portion reaches your payslip, and the share depends on your income tax bracket. A gross raise of €100 a month, at an average income, typically lands as something in the region of €55-65 net, with variation from deductions and local surcharges.
Two operating consequences:
1. Ask in gross annual terms, think in net monthly terms. The figure you name should be gross (it’s the company’s language), but your target should be calculated backwards from the net you want.
2. Consider non-cash components, which are taxed more favourably. Meal vouchers, corporate welfare, paid training, expense coverage, remote working days, devices. They often cost the company less than an equivalent raise and are worth more to you after tax. They aren’t a consolation prize: in many negotiations they’re where the deal actually closes.
Timing, without superstition
Favourable moments share one logic: when your contribution is recent, visible and documented.
- right after closing a project with a measurable outcome
- during the window when the company sets next year’s budgets (asking once budgets are approved means asking when there’s no room left)
- when you’ve permanently taken on responsibilities you didn’t have before
- when you have a concrete external offer (with the caveat in the final section)
Unfavourable moments: during cost-reduction plans, right after a bad quarter, in the same conversation where you’re reporting a problem, and in the heat of the moment after something made you angry. In that last case it isn’t the timing that’s wrong: it’s that the request becomes something else and gets read as such.
The script, and why it’s built this way
“Thanks for the time. I wanted to talk about my pay, and I’ve brought a few things.
Over the last twelve months I’ve [result 1, with a number] and [result 2, with a number]. Compared to when I started in this role, I’ve taken on [new responsibility] and [new responsibility].
I’ve checked my grade and the reference pay for this type of role: my current level sits below what I see for comparable duties.
On that basis I’d like to propose an adjustment to [gross annual figure]. How can we move on this?”
Three deliberate choices inside that text.
The subject is announced up front. No preamble: the listener knows immediately what this is about and doesn’t spend two minutes wondering where you’re going.
The figure is precise. The anchoring effect, described by Tversky and Kahneman in Judgment under Uncertainty: Heuristics and Biases (Science, 1974), is one of the most replicated findings in decision psychology: the first number on the table measurably conditions the rest of the negotiation. A precise number anchors better than a round one, because it implies a calculation behind it.
It ends with an open question. “How can we move on this?” forces a developed answer; “would that be possible?” invites a yes or a no.
Then the hard part: stop talking. The silence after the request is uncomfortable, and the instinct is to fill it with justifications that weaken everything you just said. Leave it there. It’s their turn.
If the answer is no
An immediate no is almost always a budget no, not a judgement on you. The point is not to leave the room empty-handed:
“Understood. Then I’d like to clarify two things. First: which specific results would need to happen for the answer to change? Second: when can we revisit this, with those goals as the reference?”
If you get explicit goals and a date, you’ve turned a refusal into a path, and you’ve also set up a test: if by that date the goals are met and the answer is still no, that’s the information you were looking for. Not about your worth, but about the company.
Ask for the goals in writing too, even just as a summary email you send yourself. It isn’t distrust: it’s that in six months nobody will remember the details the same way.
The real leverage: having alternatives
In the negotiation literature, starting with Fisher and Ury’s Getting to Yes (1981), the central concept is BATNA: the best alternative to a negotiated agreement. It isn’t a rhetorical technique, it’s a factual condition. People with real alternatives negotiate differently, and the other side senses it even when nothing is said.
Building alternatives means, in practice: keeping your LinkedIn profile current even when you’re not looking, knowing which skills the market is paying for (the data is in most in-demand skills and most in-demand jobs), and periodically adding to what you can do, with the method in learning a skill in 30 days.
A note of caution on using an external offer as leverage: it works once, it has a relational cost, and it should only be used if you’re prepared to accept it. Counteroffers made to retain someone who then stays resentful have a modest success rate, and either way they put you in a different category in the eyes of whoever decides.
Frequently asked questions
How often is it reasonable to ask? Once a year is a sensible cadence, or sooner if the content of the role changes substantially. Asking every three months devalues the request.
Email or in person? In person or on a video call, with an email beforehand to book the meeting and state the topic. Knowing in advance what the conversation is about stops your counterpart answering off the cuff without checking the room they have.
How much can I ask for? It depends on the starting point. If you’re correcting a documented grading mismatch, the contract determines the figure. If you’re asking for a market-aligned adjustment, asking somewhat above your target leaves room. A request disconnected from the data you brought, though, undermines everything else you said.
What if I’m afraid to ask? Extremely common, and it’s often about not being able to recognise your own contribution rather than fear of a no. If that sounds familiar, the mechanism is described in impostor syndrome: writing your results down before the conversation is, not coincidentally, the same countermeasure.
Can asking for a raise hurt me? A prepared, evidence-based, non-ultimatum request isn’t seen as a problem in healthy organisations. If it is, you’ve just collected relevant information about where you work.
Sources
- EU Pay Transparency Directive 2023/970. Right to pay information, ban on asking for pay history, invalidity of pay secrecy clauses. Transposition deadline: June 2026.
- OECD. Time series on average annual wages and on the tax wedge (Taxing Wages) across member countries.
- ISTAT. Index of contractual pay and pay statistics by sector and qualification.
- CNEL. National archive of Italian collective agreements, for checking grades and minimums.
- Eurostat. Comparative statistics on pay and labour costs across the EU.
- A. Tversky and D. Kahneman, Judgment under Uncertainty: Heuristics and Biases, Science, 1974. Anchoring effect in numerical decisions.
- R. Fisher and W. Ury, Getting to Yes, 1981. The BATNA concept, the best alternative to a negotiated agreement.



