---
title: "What onboarding a new hire costs: 2026 study | ekory"
description: "Onboarding a new hire costs an Italian company €5,300 to €9,300: €16 to €28 billion a year nationally. An ekory study on Excelsior and INPS data."
url: https://ekory.ai/blog/the-cost-of-starting-from-scratch
language: en
italian: https://ekory.ai/it/blog/the-cost-of-starting-from-scratch.md
updated: 2026-09-17
---

Research 17 September 2026 16 min read

# The cost of starting from scratch

Onboarding a new hire takes about four months before the person is fully productive, and around a hundred hours of colleagues’ time. For an Italian company that is worth €5,300 to €9,300 per person, €16 to €28 billion a year across the country. How we got there, figure by figure.

- **€5,300-9,300** the cost of onboarding each new hire, ekory estimate
- **4 months** average time for a skilled new hire to become fully productive
- **€16-28 bn** a year: what Italian companies spend bringing new hires up to speed

## Key takeaways

- **Onboarding a new hire costs an Italian company between €5,300 and €9,300**: six to eleven weeks of labor cost, split between the output lost while the person learns and the hours colleagues spend helping. An ekory estimate built on Eurostat data and three employer surveys.
- **It takes about four months** for a skilled new hire to perform like their colleagues: 3.8 months in Germany, 82 working days in Switzerland. People who change sector or come straight from education take two to three times as long.
- **Each new hire absorbs about 100 hours of colleagues’ time**, two and a half weeks spent explaining procedures, tools and special cases instead of doing their own work.
- **In 2025 Italian companies planned 5.8 million hires**, one in three to replace someone who left. About 3 million were hires meant to last, and onboarding them costs between €16 and €28 billion a year.
- **What brings the cost down is well documented**: structured onboarding, a go-to colleague with real time set aside, and documents that answer questions without someone having to ask a colleague every time.

## How many people Italian companies hire in a year

To know what starting from scratch costs, you first need to know how often it happens. There are two sources, and they are best read together.

The first is the **Excelsior Information System** run by Unioncamere and the Ministry of Labour, which asks companies every month how many people they plan to hire and on what contract. For 2025, private companies with at least one employee planned **5,807,270 hires**, counting contracts of at least 20 working days, 2.3% fewer than in 2024. The figure is reliable: for 2024, the Excelsior forecast missed the INPS outturn by just 1,500 contracts out of 5.66 million.

The second is the **INPS labor market observatory**, which counts the hires that actually happened in the private sector, very short ones included: 8,031,456 in 2025 (provisional data), of which 1,224,285 open-ended and 291,433 apprenticeships. In the same year 775,354 fixed-term contracts were turned into open-ended ones, and about 1.2 million resignations were recorded on open-ended contracts.

Hires planned by companies in 2025, by contract

| Contract | Hires | To replace someone who left |
| --- | --- | --- |
| Open-ended | 1,011,130 | 37.2% |
| Apprenticeship | 258,490 | 32.0% |
| Fixed-term | 3,478,380 | 32.9% |
| Other temporary contracts | 1,059,260 | 34.2% |
| Total | 5,807,270 | 33.8% |

Source: Unioncamere and Ministry of Labour, Excelsior Information System, La domanda di professioni e di formazione delle imprese italiane nel 2025, table 5. “Other temporary contracts” covers on-call work, agency work and freelance collaborations.

**One hire in three replaces someone who has left**: about 1.96 million hires in 2025. That is starting from scratch in its purest form: the position existed, someone knew how to do it, and now someone else has to learn it. The other two in three add people to roles that already exist (48.0%) or bring in new profiles (18.1%).

Excelsior also asks why companies use temporary contracts. Some are seasonal (33%) or cover peaks in activity (23%), but **nearly four temporary contracts in ten, 1,756,840 in 2025, are used as a trial period**: the company hires someone on a fixed term with the intention of keeping them if things work out. Those people have to be onboarded properly, just like open-ended hires.

2026 is running at the same pace. For September 2026 companies planned 565,000 hires, 0.6% fewer than a year earlier, and 1.5 million over September to November. In 65% of cases they ask for specific experience: the ideal new hire already knows the trade, but not yet how it is done in that company.

## How long it takes a new hire to become productive

Nobody is fully productive on day one, not even someone with experience. They have to learn the tools, the procedures, the customers, the products and above all the exceptions: the things this company does in a particular way for reasons nobody wrote down. For a few months they produce less than their colleagues, and the company pays them in full.

Nobody measures this period systematically in Italy. There are, however, three solid employer surveys from nearby countries that ask the same question: how long until a new hire performs like an established colleague, and how much do they produce in the meantime.

How long a new hire takes to become fully productive: the available estimates

| Survey | Who it covers | Time to full productivity | Output lost |
| --- | --- | --- | --- |
| Germany, BIBB, 2007 data | Skilled workers, 1,001 companies | 3.8 months | 3.6 weeks of pay |
| Switzerland, 2009 data | Skilled workers, 3,947 companies | 82 working days (about 4 months) | 4.7 weeks (29% lower output) |
| United Kingdom, Oxford Economics, 2014 | Workers earning over £25,000, five sectors | 15 weeks from the same sector, 32 from another sector, 40 for new graduates | 3.8 to 8 weeks from the same sector, 11 to 20 from another, 16 to 24 for new graduates |

Sources: Muehlemann and Pfeifer (IZA, 2013); Mühlemann and Strupler Leiser (IZA, 2015; Labour Economics, 2018); Oxford Economics for Unum (2014). For Switzerland the output lost is the product of the published averages (82 working days at 29% lower output).

The three estimates converge. **A skilled worker takes three to four months to perform like their colleagues.** In Germany new hires start out 42% less productive than an established colleague; in Switzerland their output stays 29% lower on average for the whole adaptation period. In a company’s accounts, that means losing the equivalent of three and a half to eight weeks of work for every hire, even when the person comes from the same sector.

The time stretches a lot for people who start further away. In the British survey, someone coming from another sector takes 32 weeks, a new graduate 40, someone returning from unemployment about a year. This matters in Italy: **for more than one hire in three (36.3%), companies do not ask for experience** in the profession or the sector.

There is also a surprising difference by company size. According to Oxford Economics, companies with fewer than ten people bring a same-sector hire up to speed in 12 weeks, those with more than 500 in 18. The authors suggest proximity matters: in a small company the newcomer works next to the owner and gets answers straight away.

## How much colleagues’ time each new hire takes

The output a new hire does not produce is only half the bill. The other half is paid by colleagues: they explain, correct, answer, double-check. The cost shows up nowhere, because those hours are already paid for and are simply taken away from someone else’s work.

The Swiss survey is the only one that measures it directly. **Each skilled new hire absorbs an average of 99 hours of informal instruction from colleagues**, about two and a half weeks of work: 95 hours in companies with fewer than 50 employees, 112 in larger ones. During those hours, the authors note, colleagues cannot do their own work.

Lay out everything a hire costs and the part that comes after the signature weighs far more than the search. In Switzerland, finding the candidate is 21% of the total cost, the new hire’s adaptation 53% and colleagues’ time 26%. In Germany a more recent study puts the costs incurred after the hire at 84% of the total.

In Italy the prevailing model is exactly this: learning by working next to someone. In 2024, 51.3% of companies trained their staff, and 18.9% did it through **shadowing a colleague**, nearly twice the share that organized internal courses (9.8%). And when a profile is hard to find, the most common answer among companies is to hire someone with similar skills and train them afterwards.

The colleague who does the shadowing is often the most experienced person in the team, the one everyone already asks. A hundred hours of onboarding come on top of the questions that person answers every day, and in the same years companies are getting ready to replace the people who retire ([3.1 million by 2029](https://ekory.ai/blog/the-cost-of-retiring-know-how)).

## What onboarding a new hire costs

With these elements the sum can be done. The cost of onboarding is two items added together: the output the new hire does not produce while learning, measured in weeks of the role’s labor cost, and the hours colleagues give them, at their hourly cost. For labor cost we use the most cautious Eurostat figure available for Italy, companies with 10 to 49 employees, where in 2024 a full-time employee cost €43,048 a year: €828 a week, €24.27 an hour.

The cost of onboarding each new hire, in three scenarios

| Scenario | Output lost | Colleagues’ hours | Cost per hire |
| --- | --- | --- | --- |
| Cautious | 3.6 weeks (Germany) | 95 (small firms) | €5,300 |
| Central | 4.7 weeks (Switzerland) | 99 (average) | €6,300 |
| High | 8 weeks (United Kingdom, same sector) | 112 (large firms) | €9,300 |

ekory calculations. Labor cost: Eurostat, Labour Cost Survey 2024, Italy, companies with 10 to 49 employees. Colleagues’ hours: Swiss survey, 2009. Rounded to the nearest hundred.

**Onboarding a new hire therefore costs between €5,300 and €9,300, with a central value of €6,300**. That is six to eleven weeks of labor cost: a month and a half to two and a half months paid in full before the person performs like everyone else. A small company that brings in ten people a year spends between €53,000 and €93,000 on it, and the figure appears under no budget line.

These figures are cautious by construction. They leave out the search for the candidate (between 16% and 35% of the total cost of a hire, depending on the survey), external courses, mistakes made while learning, and the case of people who leave before they ever become fully productive. Above all, they give everyone the path of someone arriving from the same sector: for people who change trade or start their first job, the time is two to three times longer.

## What it costs Italy: €16 to €28 billion a year

Moving from one company to the country means deciding which hires to count. We counted cautiously: **only hires meant to last**, that is open-ended contracts, apprenticeships and fixed-term contracts used as a trial period. In 2025 there were 3,026,460 of them. The 2.8 million seasonal, peak and temporary-cover contracts are left out, even though a seasonal waiter still has to be shown the ropes.

The yearly cost of onboarding in Italy, by scope and scenario (€ billion)

| Hires counted | Number | Cautious | Central | High |
| --- | --- | --- | --- | --- |
| Open-ended and apprenticeship | 1,269,620 | 6.7 | 8.0 | 11.9 |
| Plus fixed-term contracts used as a trial | 3,026,460 | 16.1 | 19.1 | 28.2 |

ekory calculations on hires planned for 2025 (Excelsior) and labor cost in 2024 (Eurostat, companies with 10 to 49 employees).

The result is that **Italian companies spend between €16 and €28 billion every year bringing new hires up to full productivity**, about €19 billion in the central scenario. Even counting only open-ended contracts and apprenticeships, the figure stays between €6.7 and €11.9 billion.

Two checks suggest the estimate leans low. The stable hires that actually happened in 2025, according to INPS, were 1,515,718, more than the 1,269,620 planned in Excelsior: with the INPS figure the narrow scope is worth €8.1 to €14.1 billion. And using the average labor cost of companies with 10 or more employees instead of the small-firm figure, the total rises to €19.5 to €34 billion.

## Sector by sector: where the cost falls

Excelsior publishes, sector by sector, the share of hires that are stable: open-ended or apprenticeship. Applying each sector’s own labor cost shows where the money goes. To stay cautious, here we count only stable hires, because the share of fixed-term contracts used as a trial is not published by sector.

The cost of onboarding stable hires, by sector (2025)

| Sector | Stable hires | Cost per hire (€) | Yearly cost (€ bn) |
| --- | --- | --- | --- |
| Manufacturing and utilities | 295,400 | 5,500-9,600 | 1.6-2.8 |
| Construction | 157,300 | 5,200-9,100 | 0.8-1.4 |
| Retail and wholesale | 182,000 | 5,500-9,700 | 1.0-1.8 |
| Tourism, hotels and restaurants | 174,300 | 4,000-7,100 | 0.7-1.2 |
| Business and personal services | 443,100 | 5,500-9,600 | 2.4-4.3 |
| Total | 1,252,000 | | 6.6-11.5 |

ekory calculations. Stable hires: 2025 hires times each sector’s open-ended and apprenticeship share (Excelsior). Labor cost: Eurostat 2024, companies with 10 to 49 employees (NACE sections C, F, G and I; for services the average of sections K-N and P-S weighted by hires). The primary sector is excluded (16,000 stable hires).

Sectors look more alike than one might expect: onboarding costs between five and ten thousand euros almost everywhere, because almost everywhere it is worth about two months of work. Tourism is the exception, where labor costs less and, according to the Swiss survey, adaptation is also shorter (40 working days instead of 82), so the tourism row is probably on the high side. What really makes the difference is volume: services and manufacturing account for more than half of the spending.

## What brings the cost down, according to research

Time to productivity is not a constant. It varies from company to company, and research on onboarding new hires, which academics call organizational socialization, is fairly consistent about what it depends on.

- **Understanding the role, feeling capable, feeling accepted.** The meta-analysis by Bauer and colleagues on 70 samples of newcomers showed that these three states are what connect good onboarding to performance, satisfaction and staying with the company. The 2025 update, covering 183 studies, confirms how much acceptance by colleagues matters.
- **Structured onboarding, not left to chance.** The SHRM Foundation guidelines distinguish four building blocks, the “four Cs”: compliance with rules and paperwork, clarity about the role, the company’s culture, and connections with colleagues. Almost every company covers the first; according to the same source, about 30% stop there or not much further. In one case it cites, Texas Instruments, people who went through an improved onboarding program were fully productive two months sooner.
- **A go-to colleague, with real time.** In Microsoft’s program, 56% of new hires who met their onboarding buddy at least once in the first 90 days say the buddy helped them become productive quickly; among those who met more than eight times, the share rises to 97%. It is the new hires’ perception, not a measurement, but the direction is clear.
- **Contact with people who know.** The British finding on small companies (12 weeks against 18) suggests that closeness to someone who really knows the work and can answer immediately makes a difference.
- **Documents that answer.** A new hire often asks colleagues because they cannot find what they need on their own. In [our previous study](https://ekory.ai/blog/the-cost-of-knowledge-that-cant-be-found) we used the estimate that white-collar workers spend 19% of their time looking for information the company already has; for someone who has just arrived and does not yet know where to look, it is reasonable to expect more.

The room for improvement is there, because the starting point is low: according to Gallup, only 12% of employees strongly agree that their company is excellent at onboarding new people. No comparable measure exists for Italy.

## What a company can do, in order of feasibility

1. **Measure time to productivity.** How long does it take a new hire in your company to work on their own? Ask the managers of the last five hires: once you have seen the number, you will not forget it.
2. **Prepare before day one.** Access, the materials of the role, a plan for the first weeks and a go-to colleague with hours set aside in the calendar. Every day spent waiting is a paid day with no output.
3. **Make the documents answerable.** Most of a new hire’s questions already have a written answer somewhere. If they can find it on their own, with the source, the most experienced colleague only answers the questions that truly need them.
4. **Turn answers into material.** Every new question a colleague answers is a missing page. Writing it down once saves the next hour, and the hour of the next hire.
5. **Take stock at 30, 60 and 90 days.** Not as a formality: to understand what the new hire is missing, and what the company is missing to onboard people better.

There are new tools for the first three points. An AI agent working on the company’s documents answers the new hire citing the document and page, and from the same materials it can prepare a course with lessons, podcasts and quizzes, with the manager seeing who has completed it. That is what ekory does: see how it works for [fast-growing companies](https://ekory.ai/scaleups) and on the [product page](https://ekory.ai/product). The practical, week-by-week guide is in [AI onboarding: the first 90 days](https://ekory.ai/blog/onboarding-with-ai-first-90-days).

## Methodological note and limits

The estimate is built so that anyone can redo it with a calculator. For each hire: cost of onboarding = weeks of output lost × weekly labor cost + colleagues’ hours × hourly labor cost.

- **Hires**: hires planned by companies for 2025 (Excelsior), contracts of at least 20 working days, private companies with at least one employee. Wide scope: open-ended (1,011,130) plus apprenticeship (258,490) plus temporary contracts used as a trial period (1,756,840) = 3,026,460. Narrow scope: the first two only, 1,269,620.
- **Output lost**: 3.6 weeks in the cautious scenario (Germany: average cost of lower productivity, €1,975, over a monthly wage of €2,365, that is 0.84 months); 4.7 in the central one (Switzerland: 82.2 working days at 28.7% lower productivity, divided by five working days a week); 8 in the high one (United Kingdom: top of the range for hires from the same sector).
- **Colleagues’ hours**: 95.5 (Swiss companies with fewer than 50 employees), 98.6 (average) and 111.7 (companies with 50 or more).
- **Labor cost**: Eurostat, Labour Cost Survey 2024, Italy, industry, construction and services excluding public administration, companies with 10 to 49 employees: €43,048 a year per full-time equivalent employee (€827.8 a week, dividing by 52) and €24.27 an hour. It is the lowest value among the size classes available; we did not update it to 2025.
- **The arithmetic**: cautious scenario 3.62 × 827.8 + 95.46 × 24.27 = 2,997 + 2,317 = €5,313; central 4.72 × 827.8 + 98.60 × 24.27 = 3,907 + 2,393 = €6,300; high 8.00 × 827.8 + 111.73 × 24.27 = 6,622 + 2,712 = €9,334. Multiplied by 3,026,460 hires: €16.1, €19.1 and €28.2 billion; by 1,269,620: €6.7, €8.0 and €11.9 billion.
- **Sectors**: Excelsior’s stable share by sector applied to each sector’s hires; Eurostat labor cost of the corresponding sections; the same weeks and hours for every sector. Because labor costs differ by sector and the primary sector is left out, the sector total (€6.6 to €11.5 billion) differs slightly from the national narrow-scope figure.

The limits need stating plainly. First: the times to productivity come from German, Swiss and British surveys, with data from 2007, 2009 and 2014; there is no comparable measure for Italy, which is one of the reasons we publish this estimate. Second: the German and Swiss surveys cover skilled workers, so applying their average to every hire overstates the cost of the simplest roles, while it understates it for people changing sector or starting their first job. Third: Excelsior counts contracts, not people, and includes part-time work (33% of open-ended hires according to INPS). Fourth: in the Swiss survey the published average adaptation cost (8.4 weeks of pay, training included) is higher than the product of the averages we use in the central scenario. Taken together, the choices we made keep the estimate on the low side.

## Frequently asked questions

### How long does it take a new hire to become productive?

For a skilled worker it takes three to four months on average: 3.8 months according to the German survey, 82 working days according to the Swiss one. People coming from another sector or from education take much longer: in the British survey, 32 weeks for someone changing sector and 40 for a new graduate.

### How much does onboarding a new employee cost?

According to the ekory estimate, onboarding costs an Italian company between €5,300 and €9,300 per person, with a central value of €6,300. That is six to eleven weeks of labor cost, split between the output the new hire does not produce while learning and the roughly one hundred hours colleagues spend helping. The cost of finding the candidate is not included.

### How many hires are planned in Italy in 2026?

Excelsior publishes monthly forecasts: for September 2026 companies planned 565,000 hires, and 1.5 million over September to November, close to the year before. Across 2025 the planned hires were 5.8 million, about one in five of them stable, open-ended or apprenticeship.

### How many people change jobs in Italy every year?

In 2025 private employers recorded 8 million hires and 7.7 million terminations, short contracts included (INPS). Resignations from open-ended contracts were about 1.2 million. In companies’ own forecasts, one hire in three replaces someone who has left.

### How can time to productivity be reduced?

Research points to three levers: structured onboarding with a plan and clear milestones, a go-to colleague with time set aside, and materials that answer questions without anyone having to ask each time. In one documented case, Texas Instruments, a better program brought new hires to full productivity two months sooner.

### Why does onboarding cost colleagues too?

Because almost everything a new hire learns, they learn from someone. In Switzerland each new hire absorbs an average of 99 hours of informal instruction from colleagues, 26% of the total cost of a hire. In Italy shadowing a colleague is the most common form of training after external courses.

## Sources

1. Unioncamere, Ministry of Labour and Social Policies, Excelsior Information System, [_La domanda di professioni e di formazione delle imprese italiane nel 2025_](https://excelsior.unioncamere.net/pubblicazioni/2025/domanda-di-professioni-e-di-formazione-delle-imprese-italiane-nel-2025), 2026 (tables 4 and 5, figure 27, annex table 30, appendix 1).
2. Unioncamere, Ministry of Labour, [“Lavoro: a settembre 565mila entrate programmate dalle imprese”](https://excelsior.unioncamere.net/sites/default/files/documenti/11092026_com_Excelsior_Settembre26.pdf), press release, 11 September 2026.
3. Unioncamere, Ministry of Labour, Excelsior Information System, [_Previsioni dei fabbisogni occupazionali e professionali in Italia a medio termine (2025-2029)_, second update](https://excelsior.unioncamere.net/pubblicazioni/2025/previsioni-dei-fabbisogni-occupazionali-e-professionali-italia-medio-termine2025), 2026.
4. INPS, [_Osservatorio sul mercato del lavoro: assunzioni, trasformazioni e cessazioni_, January-December 2025](https://www.inps.it/content/dam/inps-site/pdf/dati-analisi-bilanci/osservatori-statistici/osservatorio-precariato/OssMercato_Lavoro_GEN-DIC2024-2025.pdf), provisional data processed on 10 March 2026.
5. S. Muehlemann, H. Pfeifer, [_The Structure of Hiring Costs in Germany: Evidence from Firm-Level Data_](https://docs.iza.org/dp7656.pdf), IZA Discussion Paper 7656, 2013; published in _Industrial Relations_, 55(2), 2016.
6. S. Mühlemann, M. Strupler Leiser, [_Ten Facts You Need To Know About Hiring_](https://docs.iza.org/dp9363.pdf), IZA Discussion Paper 9363, 2015; published as [“Hiring costs and labor market tightness”](https://doi.org/10.1016/j.labeco.2018.04.010), _Labour Economics_, 52, 2018.
7. C. Linckh, S. Mühlemann, H. Pfeifer, _Beggars cannot be choosers: labor market tightness and hiring standards, wages, and hiring costs_, Leading House Working Paper 217, 2024.
8. Oxford Economics for Unum, [_The Cost of Brain Drain_](https://www.oxfordeconomics.com/resource/the-cost-of-brain-drain/), February 2014.
9. Eurostat, [Labour cost, wages and salaries by NACE activity (lc\_ncost\_r2)](https://ec.europa.eu/eurostat/databrowser/view/lc_ncost_r2/default/table), Labour Cost Survey 2024, data updated 15 September 2026.
10. T. N. Bauer et al., [“Newcomer adjustment during organizational socialization: A meta-analytic review of antecedents, outcomes, and methods”](https://doi.org/10.1037/0021-9010.92.3.707), _Journal of Applied Psychology_, 92(3), 2007.
11. T. N. Bauer et al., [“New Horizons for Newcomer Organizational Socialization: A Review, Meta-Analysis, and Future Research Directions”](https://doi.org/10.1177/01492063241277168), _Journal of Management_, 51(1), 2025.
12. T. N. Bauer, [_Onboarding New Employees: Maximizing Success_](https://studentaffairs.uga.edu/wp-content/uploads/2021/08/Onboarding-New-Employees.pdf), SHRM Foundation, 2010 (copy hosted by the University of Georgia).
13. Microsoft WorkLab, [“Strategies for Onboarding in a Hybrid World”](https://www.microsoft.com/en-us/worklab/strategies-for-onboarding-in-a-hybrid-world), 29 June 2021.
14. Gallup, [“Why the Onboarding Experience Is Key for Retention”](https://www.gallup.com/workplace/235121/why-onboarding-experience-key-retention.aspx), 16 May 2018.
15. ekory, [_The cost of knowledge that can’t be found_](https://ekory.ai/blog/the-cost-of-knowledge-that-cant-be-found), July 2026.

## More research and articles.

- [AI in law firms: professional secrecy, Law 132/2025 and the uses that hold up](https://ekory.ai/blog/ai-in-law-firms): Industries 15 September 2026 Secrecy applies to AI providers too, the client must be told, and AI-invented citations have already reached the courts, in Italy as well.

- [How many Italian companies use AI in 2026: the data by size and sector](https://ekory.ai/blog/ai-adoption-in-italian-companies-2026): Research 8 September 2026 16.4% of Italian companies with 10+ employees use AI: 53.1% of large ones, 14.2% of small ones. The data by size and sector, against Europe.

- [All research and articles.](https://ekory.ai/blog): Resources Data, regulation and cases, in English and Italian.

## Tell us about a process you still do by hand. We'll show it to you on your own materials.

For the first call, your website or a catalog is all we need.