How to Prove a Research Dataset Existed Before a Date

Flat illustration of a data server, a stack of research tables and a padlock, with chained blockchain blocks

Researchers sometimes need to show that a dataset existed in a particular form before an analysis was run, a result was published or an allegation was made. Data files are poor at recording their own history. Here is what common formats store, why it is weak, and how to freeze a dataset and timestamp it with TimeBinder® so you can show when it existed, without sharing the data.

Short answer: Dataset files record very little reliable history, and their dates come from your own systems. To show a dataset existed before an analysis or publication, freeze the exact files (and a README or codebook) and timestamp them with TimeBinder at that point. A TimeBind shows that the exact file existed no later than the recorded time and still matches its fingerprint. It doesn’t show who made it or what a court will make of it. See timestamping research data and protocols.

Disclosure: I’m Inna Hall, co-founder of TimeBinder, the file timestamping tool this guide uses. This page is general information, not legal advice.

Table of Contents

Where data files keep their dates

FormatDates stored insideNotes
CSV, TSV, JSON linesNoneOnly file-system dates, which change when copied
Excel (.xlsx)Created and last-saved dates, last saved byIn the document properties; you can view them with our file date and metadata inspector
ParquetNo standard date; a created_by field names the writing libraryCustom key-value metadata may include whatever the writer added
SPSS, Stata, R data filesVaries by format and versionTreat as unreliable unless documented
Databases and exportsWhatever columns you storeRow timestamps are data, so they are as editable as the rest

Why those dates are easy to challenge

Data is meant to be edited: cleaned, recoded, merged and re-exported. Each step can rewrite file dates and metadata, and anyone with the file can change them. In a dispute about whether data was altered after the fact, the question is not what date the file shows but whether you can show this exact version existed at that time. For files that may later be disclosed or reported, see a checkable timeline for sensitive data files.

What to freeze

  • The raw data as collected or received, before any cleaning.
  • The analysis-ready dataset used for the results you report.
  • A checksum manifest: a text file listing the SHA-256 of every file in a large dataset. Our free SHA-256 file hash checker shows the hash of any file.
  • The analysis plan and code: scripts, notebooks and the protocol. See proving a Jupyter notebook existed.
  • The data dictionary or codebook that explains the variables.

Once the files are frozen and timestamped, here is how to show a dataset wasn’t altered after a date.

A simple routine

  1. When data collection ends, make a read-only copy of the raw files.
  2. If there are many files, zip them once, or create a manifest of their SHA-256 hashes.
  3. Open TimeBinder and TimeBind the zip or manifest together with the protocol and codebook (up to 10 files per TimeBind).
  4. Repeat at the analysis-ready stage and before submission.
  5. Store the certificates with the frozen files and your data management plan.

Which plan fits

The Free plan is $0 and gives you 1 Token a month. A Pooled TimeBind costs 1 Token, covers up to 10 files and is usually recorded within 24 to 48 hours, and often sooner, depending on Bitcoin network activity. An ASAP TimeBind costs 3 Tokens and gets its own Bitcoin transaction straight away. Checking a file is always free, and no credit card is needed for the Free plan.

If you TimeBind more often, Essential is US$20 a month for 5 Tokens ($4 per Token, against $5 pay-as-you-go) and Expert is US$50 a month for 15 Tokens (about $3.33 per Token). Paid plans replace the Free plan’s monthly Token, so Essential adds 4 Tokens for US$20 and Expert adds 14 for US$50. See pricing.

What a timestamp does and doesn’t do

TimeBinder anchors each file’s SHA-256 fingerprint to the Bitcoin blockchain and gives you a Proof of Time Certificate. That is verifiable evidence of when a file existed: it shows the file existed no later than the anchor time, and any later change to the file is detectable. Your file is hashed in your browser and never uploaded. ASAP TimeBinds can be checked independently from the certificate and the Bitcoin blockchain; Pooled TimeBinds are verified through TimeBinder.

It doesn’t show who created the file, whether its contents are true, or whether AI was used to make it, and it can’t tell you how a court or anyone else will treat it. It also doesn’t cover versions you didn’t TimeBind. Pooled TimeBinds are verified through TimeBinder; ASAP TimeBinds can also be verified independently from the certificate and the Bitcoin blockchain. A timestamp doesn’t show data was collected honestly or analysed correctly. Pre-registration and data repositories serve different purposes and can work alongside it. See blockchain timestamping for research and research forgery protection.

Frequently Asked Questions

  1. Does a CSV file store when it was created?

    No. A CSV is plain text with no place for metadata, so the only dates are the file-system dates on whatever computer holds it. Those change easily when files are copied or downloaded.

  2. What about Excel and Parquet files?

    An .xlsx file stores created and last-saved dates in its document properties, written by the saving app. Parquet files record the library that wrote them in a created_by field but have no standard creation date. All of these depend on the computer that saved them.

  3. Can I timestamp a very large dataset?

    Yes. Hashing happens in your browser, so large files simply take longer. For datasets with thousands of files, make a checksum manifest and TimeBind the manifest with a zip or a sample of the files.

  4. Is it better to deposit data in a repository?

    If you can publish, a repository that issues a DOI gives you a public, dated record and makes the data citable. A timestamp is for data you can’t share yet: under embargo, under review, or containing personal information.

  5. Is my data uploaded?

    No. TimeBinder hashes files in your browser and only the fingerprint is anchored to Bitcoin, so confidential or personal data never leaves your device.

Plan details were checked against timebinder.io/pricing on 29 October 2026.