Under the delegated act to DSA Article 40, each Very Large Online Platform and Search Engine
must make a data catalogue available on its online interface. It describes the non-public data
assets that vetted researchers can request in their access applications under Article 40(4) for
research on systemic risks, including their structure and metadata.
Why do we monitor platforms' data catalogue pages?
Only the catalogues show what non-public data a platform holds, and researchers' DSA data access
applications must specify the data they concern. Platforms can edit these pages anytime, with no
changelog, so incomplete listings and later changes would pass unnoticed. For accountability and
transparency, the monitor snapshots every catalogue on a fixed schedule and publishes each change.
How to use our monitor
The monitor has two main functions, each shown below with real data from Amazon's catalogue.
Review the monitoring timeline
Timeline of monitored changes
Each dot is a monitoring run. Colours show how significant the detected changes are, and numbers indicate how many platforms had changes.
15 Aug
21:02
3
17 Aug
21:08
2
19 Aug
21:13
8
21 Aug
21:18
8
23 Aug
21:22
2
25 Aug
21:27
6
27 Aug
21:31
5
29 Aug
21:37
3
31 Aug
21:40
6
2 Sep
21:45
8
4 Sep
21:52
11
6 Sep
22:01
2
8 Sep
22:06
5
10 Sep
22:12
6
12 Sep
22:19
3
14 Sep
22:25
1
16 Sep
22:29
6
21 Sep
09:39
2
23 Sep
18:00
7
25 Sep
18:00
27 Sep
18:00
1
29 Sep
18:00
4
1 Oct
18:00
3
3 Oct
18:00
AI summary of changes
An AI model reviews the changes across all platforms and writes a summary for each monitoring run. Each summary also includes an AI assessment per platform. Before any of this reaches the dashboard, a project team member reviews it in an admin queue — the AI's classification is what's proposed, not what's automatically published.
🤖significant2026-09-29
One significant change: Amazon's contact-point sentence now also covers help with data-mining and robot tools. Booking's capture was a sign-in page and was removed; X's new page layout and the two Google Shopping changes were insignificant.
AmazonsignificantXinsignificantGoogle Google Shopping Data CatalogueinsignificantGoogle DE Google Shopping Data Catalogue DEinsignificant
Review past monitoring runs
Click the dot of a past run on the timeline and the whole dashboard switches to its results. Read the AI summaries and the individual platform overview cards of that run.
Nothing is overwritten — every past run stays queryable exactly as it looked on the day.
Change assessments with colours
An AI model assigns each detected change one of two assessment types. Each type corresponds to a colour, which is shown on the timeline and the platform overview cards.
significant
A real, worth-your-attention change: new or removed datasets, changed API endpoints, or meaningfully rewritten text.
insignificant
No real change worth flagging: small wording or formatting tweaks, dynamic tokens, session IDs, or invisible metadata.
Filtering of monitoring results
The platform overview cards can be filtered by the language of the data catalogue or the latest change assessment, and sorted in multiple ways.
Show all catalogues or only the English or German versions.
Filter
Narrow the cards by the latest assessment: all platforms, significant changes, insignificant changes, or fetch errors.
Sort
Order the cards alphabetically, by status, by date of the most recent change, or by significance of the latest assessment.
Explore monitoring runs per platform
Detailed results for each platform
Each platform overview card leads to the current version of the platform's data catalogue page, all monitoring snapshots, and an explorable change history.
Aliexpress
Degraded FetchT2
Tiktok
UnchangedT2
Amazon
ChangedT1
7 changes
Every version is archived
Each platform page starts with the source URL of the monitored catalogue, the time of the last monitoring run, and its content fingerprint, so any changes in the runs can be compared.
URL: https://trustworthyshopping.aboutamazon.com/digital-services-act-dsa-amazon-eu-store-data-access-for-vetted-researchersLast run: 3 Oct 2026, 18:00 (Berlin)Fingerprint: 455c364662d11f34480c9bc4…
Explore the latest catalogue content
The Current Content panel shows the text of the platform's data catalogue page as captured in the latest snapshot, together with the number of outgoing links.
“Approach Robust Proactive Controls Innovative Tools and Technology Holding Bad Actors Accountable Protecting and Educating Customers Collaborating Across Partners and Industry Focus Areas Product Safety Anti-Counterfeiti…”
28 outgoing links
captured in Amazon's latest snapshot.
History of monitored changes
The change history lists every monitoring run of the platform with monitoring timestamp and summary of whether and how the page changed. An AI model assesses the most recent detected change and writes a short explanation.
🤖AI Assessmentsignificant2026-09-29
Amazon's page now includes information about a dedicated point of contact for researchers who need assistance with data mining and scraping tools, which is significant because it clarifies the support available to researchers when using these tools.
Timestamp
Significance
Links ±
2026-09-29
significant
One row of 7 on its full history table.
Filtering by change type
Two change types are available for filtering the monitoring history.
insignificant
No real change (a token refresh or session ID), or only a minor wording/formatting tweak with no research impact.
significant
A real outgoing link was added or removed, or a substantive text change — worth a researcher's attention.
Compare any monitoring runs
Clicking on any monitoring run in the change history opens a comparison of that snapshot's data catalogue page to the previous one. It shows an AI summary of the changes, the changed links, and a line-by-line comparison of both snapshots' data catalogue texts.
AI assessment of the changes
An AI model assesses the changes between the monitoring runs and writes a short explanation. The assessment explains the change type and its significance.
🤖AI Assessmentsignificant2026-09-29
Amazon's page now includes information about a dedicated point of contact for researchers who need assistance with data mining and scraping tools, which is significant because it clarifies the support available to researchers when using these tools.
Line-by-line comparison
The comparison shows the data catalogue texts of both monitoring runs. Select the side by side view, with removed lines marked in the previous snapshot and added lines in the current one, or the unified view that merges both texts into a single column with the changes marked inline.
Previous · 27 Sep 2026, 18:00 (Berlin)
Current · 29 Sep 2026, 18:00 (Berlin)
**Publicly accessible data**
**Publicly accessible data**
−
Article 40.12 of the DSA establishes that very large online platforms must give access without undue delay to publicly accessible data to researchers who meet the requirements of Article 40(8), points (b), (c), (d) and (e) of the DSA, and who use the data solely for performing research that contributes to the detection, identification and understanding of systemic risks in the European Union, pursuant to Article 34 of the DSA. As stated in Amazon’s Conditions of Use, researchers who meet these requirements are permitted to use data mining, robots or similar data gathering and extraction tools in respect of publicly available data on the Amazon EU Store. Researchers who have any queries in relation to this can contact our dedicated point of contact for the data access process.
+
Article 40.12 of the DSA establishes that very large online platforms must give access without undue delay to publicly accessible data to researchers who meet the requirements of Article 40(8), points (b), (c), (d) and (e) of the DSA, and who use the data solely for performing research that contributes to the detection, identification and understanding of systemic risks in the European Union, pursuant to Article 34 of the DSA. As stated in Amazon’s Conditions of Use, researchers who meet these requirements are permitted to use data mining, robots or similar data gathering and extraction tools in respect of publicly available data on the Amazon EU Store. Researchers who have any queries in relation to this, or who need assistance with regard to the use of data mining, robots or similar data gathering and extraction tools, can contact our dedicated point of contact for the data access process.
### **Vetted researchers**
### **Vetted researchers**
Added and removed links
The comparison lists the links that changed between the monitoring runs. All added and removed links are shown with their full URLs.
