When are new IT freelance contracts posted in Germany? Weekday, time of day and quarter boundaries
Most new IT freelance contracts in Germany are posted from Tuesdays to Thursdays; on workdays 90% appear between 08:00 and 18:00 German time. Seeing new projects early means you can respond early. Here is how new projects spread over weekdays, hours and the month – and whether more appear around the turn of a quarter.
On which weekdays are most IT contracts posted?
Tuesday to Thursday lead, level with each other: each brings 20% to 21% of a week’s new listings. Monday (18%) and Friday (18%) bring fewer. Saturday and Sunday together bring 3%. The shares are averages over 19 complete weeks without a public holiday.
At what time of day are IT contracts posted?
From 08:00 to 18:00, 90% of a workday’s listings appear, 2.7% to 11% per hour. Most are added between 11:00 and 12:00 and between 15:00 and 17:00. 8.6% appear between 18:00 and 06:00; they are in the data at 06:00 in the morning. Between 06:00 and 08:00 hardly anything is added: 1.2% together. All times are German time; a listing counts to the hour in which it appeared.
The table shows the same for each weekday: every cell is the share of all new listings of a week. Whatever appears after 18:00 is in the first row of the next day – on Monday, what appeared on Sunday evening.
| Hour | Mon | Tue | Wed | Thu | Fri | Sat | Sun |
|---|---|---|---|---|---|---|---|
| 18–6 | 0.3 | 1.7 | 1.8 | 2.6 | 1.8 | 1.6 | 0.1 |
| 6–7 | 0.0 | 0.0 | 0.1 | 0.2 | 0.1 | 0.0 | 0.0 |
| 7–8 | 0.1 | 0.2 | 0.1 | 0.1 | 0.2 | 0.0 | 0.0 |
| 8–9 | 0.4 | 0.6 | 0.5 | 0.6 | 0.5 | 0.1 | 0.0 |
| 9–10 | 1.4 | 1.6 | 1.8 | 1.4 | 1.3 | 0.1 | 0.1 |
| 10–11 | 1.6 | 2.1 | 2.4 | 2.0 | 1.7 | 0.0 | 0.1 |
| 11–12 | 2.3 | 2.2 | 2.0 | 2.5 | 1.9 | 0.1 | 0.1 |
| 12–13 | 1.5 | 1.7 | 1.8 | 1.6 | 1.7 | 0.0 | 0.1 |
| 13–14 | 1.7 | 1.9 | 1.8 | 1.6 | 1.5 | 0.0 | 0.1 |
| 14–15 | 2.1 | 1.9 | 2.2 | 2.0 | 1.7 | 0.1 | 0.1 |
| 15–16 | 2.0 | 2.5 | 2.0 | 2.1 | 2.1 | 0.0 | 0.1 |
| 16–17 | 2.5 | 2.1 | 2.2 | 2.2 | 2.1 | 0.1 | 0.1 |
| 17–18 | 1.6 | 2.2 | 1.7 | 1.8 | 1.1 | 0.0 | 0.1 |
Are more IT contracts posted at the start or end of the month?
New projects spread evenly over the month: no part deviates from the usual level beyond the usual variation; the deviations lie between −4% and +5%. Each workday is compared with the same weekday in the 4 weeks before and after; 0% means as many new projects as usual.
Are more IT contracts posted at the turn of a quarter?
A common belief: shortly before and at the start of a new quarter, especially many projects are posted, because budgets for the new quarter are available and projects are meant to start with it. Is that so? Not so far: in none of the 4 periods examined around the quarter boundaries did more or fewer new projects appear than the usual variation explains.
Both quarter ends together: 4% fewer new projects than usual – within the usual variation. Both quarter starts together: about as many new projects as usual – within the usual variation. Not counted are 5 bulk uploads, in which far more listings than usual appeared on one board within a short time. Counting them, the end of June 2026 would be at +6% (about 200 more listings on 24 June 2026 and 26 June 2026), the start of July 2026 would be at +8% (about 100 more listings on 6 July 2026), the end of September 2026 would be at +6% (about 300 more listings on 30 September 2026) and the start of October 2026 would be at +2% (about 80 more listings on 2 October 2026).
| Quarter boundary | last 5 workdays | first 5 workdays |
|---|---|---|
| June → July 2026 | −5% | +2% |
| September → October 2026 | −3% | −1% |
| all together | −4% | 0% |
An arrow (↑/↓) and color mark a deviation beyond the usual variation; grey values lie within it.
The end of March cannot be tested yet: the data begin in May 2026. Every regeneration includes all quarter boundaries in the data – once a March is among them, it appears here.
Best day and time to look for IT contracts
- Look several times a day. 90% of a workday’s listings appear between 08:00 and 18:00. If you search only once in the morning, you see everything posted during the day only the next morning.
- On Mondays, check the weekend too. The weekend brings 3% of a week’s new listings together – few, and easily missed among Monday’s new ones.
- Search continuously rather than waiting for a date. New projects are posted on every workday, at the turn of the month and the quarter too. Waiting for a particular day means missing the projects of all the others.
Projectmatch scans the project boards every hour from 06:00 to 18:00 every day, matches each new project against your profile and, if you wish, applies automatically. So you see a matching project in the hour it appears, not hours later. Try it now or browse the current listings on the projects page.
See also: The monthly IT freelance market report for Germany, IT contract length in Germany: start and extension, How remote are IT freelance projects in Germany? and IT contract & freelance job boards in Germany.
Method
What is counted. Every listing that Projectmatch finds for the first time on one of the German IT freelance project boards it covers – also when its description could not be read completely afterwards, for instance because the project had already been closed again: it was posted all the same. The same position, posted by several agencies or on several boards, counts several times; for a contractor, each listing is a separate way into the project. When this analysis was introduced (October 2026), such duplicates were spread about evenly over weekdays and hours. In October 2026, 625 new listings were added per workday on average.
How precise the time is. Projectmatch checks the boards once an hour from 06:00 to 18:00 German time every day, each time in the first minutes of the hour. A listing counts to the hour before the check that first found it: what the 10:00 check finds new appeared between 09:00 and 10:00. The time is no more precise than the hour. Whatever appears after 18:00 is found by the 06:00 check the next morning and counts to that day. What is measured is when a listing became visible on the board, not when it was written. Hours are available since 20 June 2026.
Why shares and comparisons rather than absolute numbers. How many listings are counted on a day depends on which boards are covered and how completely – and that has changed since May 2026: boards were added, the checks of single boards failed for a while. So the analysis never compares days across such changes. Weekdays and hours are shares within the same week. Parts of the month and quarter boundaries are compared, per board, with the same weekdays of the same board in the weeks before and after – they are the “usual level”.
Cleaning. A board’s first 7 days are its initial import – everything already posted there – and do not count. An hour of a board with more than 2 times its usual count plus 30 counts as a bulk upload and is trimmed to the usual count; usual is the median of the same hour on the same weekday in the 4 weeks before and after. Such peaks occur when an agency posts or re-posts many projects at once, or when Projectmatch catches up after an outage; so far 27 peaks with 5,045 listings in total were trimmed. A workday of a board with less than 50% of its usual count counts as an outage and is left out of all comparisons (checked for boards that usually list at least 10 projects a day; so far 21 days of single boards). Public holidays are left out too: the nationwide ones, Corpus Christi and All Saints’ Day (holidays in Bavaria, Baden-Württemberg and North Rhine-Westphalia, among others) and Christmas Eve and New Year’s Eve. Weeks with a holiday are left out of the weekday and hour comparison.
What counts as a difference. A weekday or an hour only counts as weaker than the busiest one when its share is at least 1 percentage point lower on average and the gap is at least 2 standard errors, measured over the single weeks (19 weeks, with hours 17). A part of the month or a period at a quarter boundary only counts as above or below the usual level when it deviates by at least 5% and by at least 2 standard errors, taken from the variation between months or between single workdays. Everything else is “within the usual variation”.
Quarter boundaries. The last 5 workdays of a quarter and the first 5 of the next are compared with the workdays of the 4 weeks before and after, per board and weekday. A board only takes part when it has at least 10 usable workdays in those weeks and its level afterwards differs from the one before by less than a factor of 2 – otherwise its coverage changed in between. Every regeneration includes all quarter boundaries in the data; the row “all together” tests them jointly, and that test grows more conclusive with every quarter. A month is included once it is complete in the data.
The figures begin in May 2026 and are updated monthly. They are published under the CC BY 4.0 license: you may use them freely, also commercially, crediting “Projectmatch (projectmatch.io)” and, online, linking to this page.