The SEC’s Financial Statement Data Sets are a reporting aid, not a substitute for the filing itself. They provide the numeric information from the face financial statements extracted from corporate reports filed with the Commission using XBRL, and the information is presented without change from the “as filed” reports in a flattened format so users can compare disclosures across time and across registrants[1].
For reporters, the practical lesson is that the datasets are useful for fast screening, comparisons, and trend checks, but every important finding still needs to be verified against the underlying filing before publication[2][3].
The SEC documentation organizes the Financial Statement Data Sets into three core files that reporters should understand together: TAG, NUM, and PRE. The table below summarizes how the SEC describes each file and how a reporter should use it.
| File | Definition | Why it matters for reporters |
|---|---|---|
| TAG | The TAG file is the dataset of all tags used in submissions, including both standard and custom tags.[4] | Use it to understand which XBRL concepts are present in the submission and to interpret the labels attached to reported facts. |
| NUM | The NUM file is the dataset of all numeric XBRL facts presented on the primary financial statements.[5] | This is the core facts table for amount comparisons, and the SEC notes that the NUM file now includes a segments field in the reprocessed data sets.[6] |
| PRE | The PRE file describes the presentation and organization of the financial statements and their tagged facts. | Use it to see how the filing structures line items and relationships before drawing conclusions from the numeric rows. |
The SEC also says the data sets are flatter and more compact than other data products that include additional disclosures, which makes them easier to scan but also narrower in scope[7].
Two timing rules matter for reporters. First, the data sets are updated quarterly[8]. Second, documents filed after the last business day of a quarter are included in the subsequent quarterly posting[9].
The documentation also says that documents filed after 5:30 p.m. Eastern on the last day of the month will be included in the subsequent posting[10][11]. Reporters should treat that cutoff as an operational detail of the posting cycle and the quarterly rule as the broader update schedule.
Use the data sets to find leads, then verify with the underlying filing before concluding anything material. The SEC explicitly says the data sets are intended to assist the public in analyzing filings, but they are not a substitute for the filings themselves, and investors should review the full Commission filings before making any investment decision[21][22].
That caution matters because the SEC also says the data can omit information, can carry extraction or compilation errors, and may not reflect all filing metadata[23][24]. For reporters, that means the safest workflow is: identify the numeric fact in NUM, check how it is organized in PRE, confirm the tag context in TAG, then open the original filing and read the note and statement context before publishing.
The Financial Statement Data Sets are best treated as a structured screening tool for XBRL-based face financial data, not as the final record. They are updated quarterly, have cutoff rules that affect when filings appear, and carry limits that make filing-level verification essential[25][26][27][28].
Get more accurate answers with Super Pandi, upload files, personalized discovery feed, save searches and contribute to the PandiPedia.
Let's look at alternatives: