DEVELOPERS
Company data as a flat file
Take the data as files: a one-off extract, or a full bulk dataset loaded and refreshed on a cadence. Delivered by download, or automatically to your S3 bucket or over SFTP, in the format your pipeline expects.
3 of 380,000,000 rows
Who it's for
Best for teams that want the data in their own stack.
Load a whole dataset into your warehouse and keep it refreshed, or pull a one-off slice for a project. Take it as a direct download, or have every refresh delivered automatically to your S3 bucket or over SFTP. Either way you hold the data and work it your way, with no live dependency on us.
How it works
Configure your delivery.
A flat file gives you the data as files to process on your side. Set the scope, cadence and destination to see exactly what you get.
Scope
Cadence
Destination
The complete dataset, refreshed on the schedule you set, dropped to your S3 bucket.
Best when the data underpins a product, a model or an internal system that needs to stay current.
Sample file
s3://your-bucket/companies_global_2026-07.parquet · refreshed monthlyTake it as a direct download, or have every refresh delivered automatically to your S3 bucket or over SFTP.
How it's delivered
Download it, or have it delivered to you.
The same files, however suits your stack. Take them as a direct download, or have every refresh delivered automatically to your S3 bucket or over SFTP on the schedule you set. Delivery is encrypted in transit and monitored, so a missed drop is caught and re-run, never silently skipped.
Direct download
Pull each file yourself, once or whenever a refresh is ready.
S3 bucket
Every refresh lands straight in your S3 bucket, ready to load. Suits cloud-native and AWS-based pipelines.
SFTP
Files are delivered over SFTP to your endpoint. Suits teams that prefer a traditional secure file transfer.
Not sure whether S3 or SFTP fits your setup? Talk to a data expert and we will match the delivery to your stack.
Formats and specifics
Pick the format that fits your stack.
The same records, delivered the way you want to read them. Switch formats to see the fit.
For spreadsheets, quick inspection and near-universal compatibility.
Sample record · IKEA Belgium
uniqueID,companyName,tradeName,nationalID,statusCodeName,yearStarted,address1,postCode,city,cityCode,countryCode,latitude,longitude,website,webDomain,internationalCode01,altInternationalCode01,globalLinkageUniqueId,globalLinkageBusinessName,familyMembers
0511525465,IKEA Belgium,Ikea Anderlecht,BE2150382409,Branch,2005,Bergense Steenweg 1432,1070,Anderlecht,BE21001,BE,50.81086425,4.2914458,https://www.ikea.com/be/fr/stores/anderlecht/,ikea.com,5961,4791,0529233333,Stichting INGKA Foundation,948Details worth knowing before you load
- ✓UTF-8 encoding, with a header row naming every field.
- ✓Comma-delimited by default for CSV. Tab-delimited available on request.
- ✓Large datasets are split into multiple files and compressed (gzip) so they move quickly.
- ✓Every field maps to our data dictionary, so column names mean the same thing across datasets and across refreshes.
Loading your files
Files drop straight into the tools you already use.
pd.read_csv, pd.read_json or pd.read_parquet).Because the schema is consistent and every field is documented, there is no guesswork about what a column means or how to map it, and it stays stable across refreshes.
How do you want the data?
Flat file or API — the same records either way.
FAQ
Frequently asked questions
Parquet, CSV or Excel. You choose. Parquet is best for loading into a warehouse or analytics engine at scale, CSV is the most portable, and Excel opens straight into a workbook for business users and analysts.
Either. Take a full bulk dataset, or a one-off slice filtered to the markets, sectors and fields you need.
Your choice. Take a one-time delivery, or a recurring refresh on a schedule you set.
By direct download, or delivered automatically to your S3 bucket or over SFTP on the schedule you set. The files are identical whichever route you choose.
Both deliver the same files on the same schedule. S3 suits cloud-native and AWS-based pipelines, landing files straight into object storage. SFTP suits teams that prefer a traditional secure file transfer into an existing endpoint.
It depends on the scope you request. Large datasets are split into multiple compressed files so they transfer quickly and load cleanly.
A flat file gives you a set of records to hold and process yourself. The API returns individual records in real time. Many teams use both.
All five: identity, group structure, tax ID, points of interest and financials. You can combine them in a single delivery.
Yes. Every field maps to our data dictionary, so column names are consistent and defined across datasets and across refreshes.
Yes. Request a sample matched to your criteria and see the real structure, fields and values before you commit.
Get the data in your stack, your way.
Request a sample matched to your criteria, and see it for yourself.