What is a Concordance DAT File?Load File

📖 What is a Concordance DAT File?

A Concordance DAT file is a load file used in eDiscovery. It is a plain-text, delimited file that lists documents in a legal production, one document per line, with a column of metadata for each. Review platforms read the DAT file to load a set of documents in the right order with the right information attached.

The name comes from Concordance, the litigation-support software that first used this layout as a standard. The format stuck, and today a Concordance DAT file is the most common way document data moves between eDiscovery tools. When one law firm produces documents to another, or a vendor hands a data set to a review team, the metadata almost always travels in a .dat file.

The extension is .dat, which is a generic name many programs use, so a Concordance DAT file is really a specific type of delimited text file. What makes it a load file is its structure: a header row of field names, then one row per document, with values separated by special delimiter characters chosen because they almost never appear inside real metadata.

⚡ Quick Facts

🔣 The Special Delimiters

A normal CSV uses commas to separate values. That breaks the moment a document's metadata contains a comma, and legal metadata is full of them: names, addresses, subject lines. Concordance solved this by picking delimiter characters that almost never appear in real text.

Two characters do the work. The field separator marks where one column ends and the next begins. It is a non-printable control character, ASCII 20 (byte 0x14), which is why you never see it on screen. The text qualifier wraps each value so its contents are read as one piece, even if a stray separator appears inside. The qualifier is the thorn character þ, ASCII 254, stored as byte 0xFE.

Because these characters are so rare in ordinary writing, a value almost never collides with a delimiter, which is exactly why the format has lasted for decades. You will not see the field separator at all in a plain text editor, since ASCII 20 is invisible, and the thorn qualifier shows as a small box or odd symbol. That mix is the first sign you are looking at a real Concordance DAT file rather than a comma file.

🗂️ What a DAT File Holds

Open a Concordance DAT file and the shape is simple once you know the delimiters. The very first line is the header row, a list of field names. Common ones include DOCID, BEGBATES and ENDBATES (also written BEGNO and ENDNO), CUSTODIAN, AUTHOR, DATESENT, plus path fields like NATIVEPATH, TEXTPATH and OCRPATH. Every line after that describes one document, with a value under each field.

The fields fall into a few groups. Identifiers like the document id and the Bates numbers uniquely label each record. Metadata fields carry the facts about the document: who sent it, when, the subject, the file name. Family fields, named BegAttach and EndAttach, record relationships so an email and its attachments stay linked as one family. Every document in a family shares the same BegAttach and EndAttach values. Path fields point to the actual files that go with each record. The extracted document text is stored in a separate .txt file per document, not inside the DAT, and a path field points to it, alongside paths to the native file and images.

# a Concordance DAT file, delimiters shown as [F] field and [Q] qualifier
[Q]DOCID[Q][F][Q]BEGBATES[Q][F][Q]CUSTODIAN[Q][F][Q]DATESENT[Q][F][Q]SUBJECT[Q]
[Q]DOC0001[Q][F][Q]ABC000001[Q][F][Q]Alex Morgan[Q][F][Q]2024-05-14[Q][F][Q]Q2 report[Q]
[Q]DOC0002[Q][F][Q]ABC000002[Q][F][Q]Sam Lee[Q][F][Q]2024-05-15[Q][F][Q]Re: Q2 report[Q]

In a real file the [F] marks are the ASCII 20 field separator and the [Q] marks are the thorn qualifier. A review platform reads the header, then maps each value on every line to the matching field.

🧩 The Files That Come With It

A DAT file rarely travels alone. It is the metadata half of a production, and it usually arrives with companion files that hold the images, text and page mapping.

The most common partner is the OPT file, an Opticon cross-reference file that lists one image page per line and links each page to its Bates number. The LFP file plays a similar role for page-level image loading. Alongside these sit the actual content: single-page TIFF images named by Bates number, single-page text files holding the extracted text, and sometimes the native files themselves.

Put together, a typical production folder holds a VOL001.DAT for metadata, a VOL001.OPT for the image cross-reference, and folders of TIFF, TXT and native files. The DAT file is the map that ties all of it into reviewable documents. Concordance DAT paired with an Opticon load file is the de facto standard for productions, though sibling formats exist: Summation uses a DII load file, and the EDRM XML format is defined but rarely used in practice.

📂 How to Open a Concordance DAT File

How you open a .dat file depends on whether you want to read it, edit it or load a whole production for review.

1. A text editor, to look inside. Any editor such as Notepad or a code editor opens the raw file. You will see the header row and the records, with the delimiters showing as small boxes. This is the quickest way to confirm what fields a production contains.

2. A spreadsheet, to read it as a table. Excel or another spreadsheet can import the file if you set the field separator to ASCII 20 and the text qualifier to the thorn character. Done right, each field lands in its own column and the file reads like a normal table. This is how many people first make sense of a DAT file.

3. An eDiscovery review platform, to work the case. Tools such as Concordance, Relativity, iPro or a hosted review platform ingest the DAT file along with its OPT and image files, then present each document with its metadata, images and text ready for search, tagging and production. This is the full workflow the format was built for.

⚠️ The Encoding Trap

The single most common reason a Concordance DAT file will not parse is character encoding, and it catches experienced people. The thorn text qualifier is stored as one byte, 0xFE, in the old Windows-1252 encoding. In UTF-8, the same character is stored as two bytes, 0xC3 0xBE.

A DAT file carries no marker that says which encoding it uses. So if a tool reads a Windows-1252 file as if it were UTF-8, it looks for the wrong byte pattern. It either misses the delimiters and merges every field into one, or it trips over the lone 0xFE byte and reports an error. The qualifiers and field boundaries silently fall apart.

The fix is to detect the encoding before parsing, not assume it. Files created by older tools tend to be Windows-1252, newer ones UTF-8, and some productions even mix the two line by line. When fields look scrambled after import, the encoding is almost always the reason, and re-importing with the correct setting restores the columns. A UTF-8 file may also carry a byte-order mark, or BOM, that some parsers show as stray characters at the very start of the first field.

⚖️ DAT vs CSV

A DAT file and a CSV file are both delimited text, so people often ask why legal work uses the DAT rather than a plain CSV. The answer is reliability with messy real-world metadata.

A CSV separates fields with commas and wraps values in quotation marks. Both characters appear constantly in document metadata: a subject line with a comma, a body quote with quotation marks. Each one can break a CSV parser and shift every following field. A Concordance DAT file uses an invisible ASCII 20 separator and a thorn qualifier instead, characters that almost never appear in real text, so the columns stay aligned even across millions of records.

The trade-off is readability. A CSV opens cleanly in any spreadsheet, while a DAT file needs the right delimiter setting first. For casual data that is a nuisance, but for a legal production where a single misaligned field can misattribute a document, the DAT file's reliability is worth the extra step.

🛡️ Is It Safe to Open?

A Concordance DAT file is plain text. Being plain text, it contains no executable code, so nothing runs when you open it and the format poses no direct threat of its own. Viewing one in a text editor or spreadsheet is perfectly safe.

The care needed is about the contents. A DAT file usually lists legal evidence and often contains sensitive metadata: names, email addresses, dates and subjects tied to a case. Treat it as confidential material, keep it inside your secure review environment, and avoid uploading a production's metadata to an unknown online tool. When the file points to native files, those native files should get the same caution as any outside document.

🛠️ Common Issues

🏛️ A Short History

Concordance began as litigation-support software for organizing and searching large document collections. As legal teams needed a reliable way to move document sets between tools, Concordance's delimited load file became the shared standard, and its name stuck to the format even as other software adopted it.

The choice of an invisible separator and a thorn qualifier was deliberate. In an era before consistent Unicode support, the designers needed separators that would never clash with the commas, quotes and tabs found in real document text. Those odd characters solved the problem so well that the format still uses them today.

LexisNexis later acquired Concordance, and although review platforms such as Relativity now dominate day-to-day work, the Concordance DAT load file remains the common currency of eDiscovery productions across the industry.

❓ Frequently Asked Questions

📝 Summary

A Concordance DAT file, extension .dat, is the standard load file of eDiscovery. It is a delimited plain-text file that lists documents in a legal production, one per line, with a column of metadata for each. Fields are separated by an invisible control character, ASCII 20, and wrapped in a thorn text qualifier, ASCII 254, characters chosen because they almost never appear in real text.

The DAT file is the metadata map of a production. It travels with companion files, an OPT or LFP for image pages and folders of TIFF, text and native files, and a review platform reads it to rebuild each document with its data, images and text. To read one quickly, import it into a spreadsheet and set the field separator to ASCII 20 with the thorn text qualifier, watching the encoding if the text looks scrambled.