Recipe · Apify Store · runs in a terminal
What 602 EU TED competition notices show about who buys what
Construction led the German sample, medical products led Poland’s, and 57 notices still open at collection were due within a week.
Germany and Poland had almost the same number of competition notices in this Tenders Electronic Daily (TED) sample: 98 and 100. But Germany had 38 in construction, while Poland’s largest category was medical products, with 21. If I were setting up a supplier alert, I’d start with what the buyer is asking for and use country to route the matches.
I used version 1 of the public EU Public Procurement Notices and Tenders dataset. It contains 1,238 unique TED publication numbers from October 20 and November 17, 2026. The first date accounts for 1,200 rows selected from a larger archived run. I counted the 602 competition notices for the buying and deadline analysis. A procedure can produce multiple notices, and the sample includes buyers outside the EU. Kaggle dataset · automatic/ted-eu-tenders-scraper
Who buys what
| First buyer’s country | Competition notices | Services | Supplies | Works | Most common main CPV division |
|---|---|---|---|---|---|
| France | 103 | 65 | 29 | 9 | Architecture and engineering (71), 14 |
| Poland | 100 | 39 | 48 | 13 | Medical equipment and pharmaceuticals (33), 21 |
| Germany | 98 | 39 | 21 | 38 | Construction (45), 38 |
| Spain | 37 | 26 | 10 | 1 | Medical and architecture/engineering (33, 71), 6 each |
| Italy | 33 | 22 | 8 | 3 | Financial and insurance services (66), 6 |
The table shows the five countries with the most competition notices in this export. Services, supplies and works come from contract_nature. The last column groups the first two digits of main_cpv_code using the EU’s Common Procurement Vocabulary (CPV). Each row uses the first buyer reported by TED and one main code. Other buyers and secondary categories can still matter to a supplier. TED CPV code list
The buyer names show why a category alert helps. Germany’s Wohnbau Stadt Coburg GmbH published three construction notices for the same address, covering building automation, ventilation and roofing. In Poland, Uniwersyteckie Centrum Kliniczne in Gdańsk published two medical-equipment notices under the same CPV division with different main codes. Both patterns are visible in the notice titles and codes in the CSV.
How much time was left to bid?
| Deadline measure | Competition notices |
|---|---|
| With a reported deadline | 542 of 602 |
| Flagged open at collection, with a deadline | 503 |
| Due within seven days of collection | 57 of 503 |
| Median days from publication to deadline, among open notices | 30 |
| Median days from collection to deadline, among open notices | 23 |
The archive records the earliest reported lot deadline. I compared UTC calendar dates and counted deadlines on the collection date through seven calendar days later. A scan that ran a week after collection would have had little or no time to act on those 57 notices. Use the live TED notice and the deadline for your lot when setting a reminder.
What the reported values cover
| Contract nature | Competition notices | With a positive EUR estimate | Median reported estimate |
|---|---|---|---|
| Services | 291 | 108 | €881,616 |
| Supplies | 217 | 72 | €678,398 |
| Works | 94 | 19 | €2,793,840 |
Works had the largest median estimate among the rows shown, with positive EUR estimates on just 19 of 94 notices. Across all 602 competitions, 285 reported an estimate in some currency and 199 had a positive EUR estimate. I kept currencies as published and left blank, zero and non-EUR estimates out of the medians. Of the 533 result notices, 475 reported an award value in some currency. Result notices describe a different stage and amounts can cover different lots or scopes, so read the value field and the notice together before budgeting.
Reproduce the tables
Download version 1 from Kaggle and extract the CSV.
curl -fsSL -o ted.zip \
'https://www.kaggle.com/api/v1/datasets/download/kirkins/eu-procurement-tenders?datasetVersionNumber=1'
unzip -p ted.zip data.csv > data.csv
Save the Python below as ted.py and run python3 ted.py data.csv. The script uses only the standard library.
import csv
import sys
from collections import Counter
from datetime import date
from decimal import Decimal
from statistics import median
with open(sys.argv[1], encoding="utf-8", newline="") as file:
rows = list(csv.DictReader(file))
assert len(rows) == len({row["notice_id"] for row in rows})
print("Notices by publication date:", Counter(row["publication_date"] for row in rows))
print("Notice forms:", Counter(row["form_type"] for row in rows))
competition = [row for row in rows if row["form_type"] == "competition"]
natures = ("services", "supplies", "works")
countries = Counter(row["buyer_country"] for row in competition)
for country, count in countries.most_common(5):
group = [row for row in competition if row["buyer_country"] == country]
nature = Counter(row["contract_nature"] for row in group)
divisions = Counter(row["main_cpv_code"][:2] for row in group)
highest = max(divisions.values())
leaders = sorted(code for code, n in divisions.items() if n == highest)
print(country, count, *(nature[key] for key in natures), leaders, highest)
for country, buyer in (("DEU", "Wohnbau Stadt Coburg GmbH"),
("POL", "Uniwersyteckie Centrum Kliniczne")):
for row in competition:
if (row["buyer_country"], row["buyer_name"]) == (country, buyer):
print(buyer, row["buyer_city"], row["notice_id"],
row["main_cpv_code"], row["notice_title"])
dated = [row for row in competition if row["bid_deadline_utc"]]
open_rows = [row for row in dated if row["open_for_bids_at_collection"] == "true"]
deadlines = [date.fromisoformat(row["bid_deadline_utc"][:10]) for row in open_rows]
published_days = [(end - date.fromisoformat(row["publication_date"])).days
for row, end in zip(open_rows, deadlines)]
remaining_days = [(end - date.fromisoformat(row["collected_on_utc"])).days
for row, end in zip(open_rows, deadlines)]
print("Dated and dated open:", len(dated), len(open_rows))
print("Due within seven days:", sum(0 <= days <= 7 for days in remaining_days))
print("Median days from publication and collection:",
median(published_days), median(remaining_days))
estimates = [row for row in competition if row["estimated_value"]]
eur = [row for row in estimates if row["estimated_value_currency"] == "EUR"
and Decimal(row["estimated_value"]) > 0]
print("Reported and positive EUR estimates:", len(estimates), len(eur))
for nature in natures:
count = sum(row["contract_nature"] == nature for row in competition)
amounts = [Decimal(row["estimated_value"]) for row in eur
if row["contract_nature"] == nature]
print(nature, count, len(amounts), f"{median(amounts):,.0f}")
results = [row for row in rows if row["form_type"] == "result"]
print("Results with reported awards:", sum(bool(row["awarded_value"]) for row in results),
"of", len(results))
Data source
Every finding uses data.csv from version 1. The CSV’s SHA-256 is 1cf19fd1c8500771be86bb06a3bb8d5f73587ae0573a5211325f52def99ca5b8. Source data from the Publications Office of the European Union and Tenders Electronic Daily (TED), © European Union. The archive selects and normalises fields and removes personal information. I grouped the published rows by buyer country, procurement category, deadline and value. Kaggle dataset · TED legal notice
Affiliate disclosure: The Apify Store link carries our referral code. If you sign up for a paid plan through it, ssh.autos may earn a commission at no cost to you.
Sources in this recipe
-
TED EU Tenders — EU Public Procurement Notices
Scrape EU public procurement notices from TED (Tenders Electronic Daily) for any country, CPV code, notice type or publication window.
Machine-readable: openapi.json · llms.txt