What Is a GEDCOM File? The Family Tree Format Explained
Someone sends you a file. It ends in .ged. You double-click it and nothing useful happens. Or a family history site offers you a download button and calls the file a GEDCOM. Either way, you are left with the same question: what is this thing, and what can you do with it?
A GEDCOM file is a plain text file that holds a family tree. It stores people, families, dates, places, and notes in a set form that many programs can read. That is the whole point of it. Your tree is not trapped in one app. If a program can export a GEDCOM file and another program can import it, you can move your work from one place to the next.
This guide explains what is inside the file, where the name came from, and what the file can and cannot do. When you want to see your own file checked and turned into tables you can read, the GedConvert tool on the home page does that in one pass. If the file will not open at all, start with how to open a GEDCOM file.
Where the Name Comes From
GEDCOM is short for Genealogical Data Communication. The name tells you the job. It is not a program. It is a way for programs to talk to each other about family history data.
The format was first made by The Church of Jesus Christ of Latter-day Saints. The church runs FamilySearch, one of the large family history groups in the world. The first public form came out in the 1980s. New forms came out over time as computers and family history tools changed.
Two forms matter most today. GEDCOM 5.5.1 was the common form for about twenty years. Most files you meet now use it, even if the file was made last week. GEDCOM 7.0 came out in 2021. It is the current form. It cleans up old rules, requires one clear text encoding, and adds a few new parts. The guide on GEDCOM 5.5.1 vs 7.0 explains the changes in full.
The spec is now kept in the open. You can read it at gedcom.io, the home of the GEDCOM standard. That matters for you. The rules are not a secret held by one company. Any tool maker can read them and build a tool that reads and writes the same files.
What the File Looks Like Inside
A GEDCOM file is plain text. You can open it in any text editor and read it with your own eyes. That is rare for computer files, and it is a gift. A file you can read is a file you can check, fix, and keep for a long time.
Each line has the same shape. First comes a number. That number is the level. It shows how the line fits into the tree of data. Next comes a short tag in capital letters. The tag says what the line is. Last comes the value, the real data. Some lines also hold a short ID code that other lines can point to.
Here is a small part of a real file, made simple:
| Line | What it means |
|---|---|
| 0 HEAD | Start of the file header |
| 1 SOUR FamilySearch | The program that made the file |
| 1 GEDC | The GEDCOM form used |
| 2 VERS 5.5.1 | Form 5.5.1 of the spec |
| 1 CHAR UTF-8 | The text encoding used |
| 0 @I1@ INDI | Start of one person, with ID I1 |
| 1 NAME Mary /Smith/ | Her name, with family name in slashes |
| 1 BIRT | Her birth event starts here |
| 2 DATE 12 MAR 1904 | The date of that event |
| 2 PLAC Smithfield, Ohio, United States | The place of that event |
Read it top to bottom and you can follow it. Level 0 lines start big blocks. Level 1 lines sit inside the block above them. Level 2 lines sit inside the level 1 line above them. The file is a tree of lines, and the level numbers draw the branches.
Names have one odd rule worth knowing. The family name goes between slashes, like Mary /Smith/. Programs use the slashes to tell which part is the family name. If the slashes are missing or in the wrong place, a program may sort or show the name in a wrong way.
The Main Parts of a File
Most of a GEDCOM file is made of records. Each record is one thing in your tree. The main kinds are few, and you will meet them again and again.
- Individual records (INDI). One record per person. It holds the name, sex, and links to that person's events and families.
- Family records (FAM). One record per family group. It holds the partners and the children, and events like a marriage.
- Source records (SOUR). One record per source, such as a census page, a birth record, or a book. Other records point to sources to show where a fact came from.
- Note records (NOTE). Longer text that does not fit in one short field, such as a life story or a research note.
- Media records (OBJE). Links to photos, scans, and other files tied to a person or event.
Events such as birth, death, marriage, and census are not top records of their own. They sit inside a person or family record, each with a date, a place, and often a source link.
Each record has an ID, shown between @ signs, like @I1@ or @F3@. Other records use those IDs to point at each other. A child record points to a family. A family points back to its children and partners. Those points are the links that hold the tree together. When a point aims at an ID that is not in the file, the link is broken. That is a common fault, and the guide on broken links in a GEDCOM file shows how to find and fix them.
What a GEDCOM File Holds Well
A GEDCOM file is good at the core facts of family history. Names, dates, and places move well. So do the links between parents, partners, and children. Sources move too, though the detail can vary by program. Notes move as plain text.
This is why the format has lasted so long. The core of a family tree is simple. A person was born. A person married. A person died. Two people are the parents of a child. Those facts fit the form well, and they are the facts you most need to keep safe.
The format also holds facts you may not expect. It can store adoptions, name changes, and many kinds of events. It can store the place of an event in a set form. It can store a citation that says which page of which book a fact came from. Good programs use all of this. Weak programs use only part of it, which leads to the next point.
What a GEDCOM File Does Not Hold Well
No format is perfect, and it helps to know the weak spots before you move a tree.
Photos and scans do not travel inside the file. A GEDCOM file holds links to media files, not the files themselves. If you move only the .ged file, the photos stay behind. You must move the media files too, and the links must still match.
Each program adds its own extras. The spec lets programs add custom tags. Those tags start with an underscore, like _MILT or _TODO. Your old program reads them fine. A new program may not know what they mean and may drop them on import. Data stored only in custom tags is data at risk when you switch tools.
Some details have no home. Shared events, complex source notes, and some place details do not map cleanly between programs. Most of your tree will move. A small part may arrive in a simpler form, or land in a note field.
The file holds no layout. A GEDCOM file has facts, not charts. The look of your tree — the colors, the chart style, the way photos are placed — lives in the program, not the file. You move the facts. You rebuild the look.
None of this makes the format weak. It makes it honest. It is a data format, and data is what it moves. Knowing the limits helps you check the right things after a move.
Why This Format Still Matters
You might ask why an old text format still runs the world of family history. The short answer is that nothing else does the same job. Each big site and program has its own store of data. GEDCOM is the one bridge they all share.
Three traits keep it in place. First, it is open. Anyone can read the spec and build for it. Second, it is plain text. A file from 1999 can still be read today with no special tool. Third, it is the export that every major service offers. When you leave a service, the GEDCOM file is what you take with you.
That last trait is the one that protects you. Services change prices. Services close. Programs stop being sold. If you hold a current GEDCOM copy of your tree, none of that can take your work away. The guide on why genealogy data gets locked in tells that story in full, and the backup guide shows how to keep a copy on a set schedule.
How to Check Your Own File
You do not need to read a whole file line by line. A quick check tells you a lot. Open the file in a text editor and look at the first lines. You should see a HEAD block, a line that names the program that made it, and a CHAR line that states the encoding. Then search for TRLR. That tag marks the last line of a complete file. If it is missing, the file was cut short.
Next, count the scale of the file. Search for " INDI" to see how many person records it holds, and " FAM" for family records. Those counts should be close to what you expect from your tree. A file with far fewer people than your tree is a partial export, not a full one.
For a full check, use a validator. The GedConvert tool reads your file and gives you a validation report with each problem listed by line. The free plan includes the full report plus CSV tables of your individuals and families, so you can see your data in a form that is easy to scan. Paid plans add event, source, and note tables, a repaired file, and version conversion. See the pricing page for the full list.
Frequently Asked Questions
Is a GEDCOM file the same as a family tree? It holds the data of a family tree: the people, families, facts, and links. It does not hold the chart or the look. Think of it as the facts behind the tree.
Which programs can make a GEDCOM file? Almost all of them. Ancestry, MyHeritage, FamilySearch, Gramps, RootsMagic, and Family Tree Maker all export GEDCOM files. The steps differ, but the result is the same kind of file.
Can I edit a GEDCOM file by hand? Yes, since it is plain text. Small fixes by hand are fine if you keep the line shape exact. For big files, a tool is safer, since one wrong level number can break a whole record.
What is the difference between .ged and .gedcom? Both names are used for the same kind of file. The short .ged name is older and more common. The longer name shows up in newer tools and in GEDCOM 7.0, which uses .ged as well in most cases.
How big can a GEDCOM file be? There is no small limit in the spec. Real files range from a few kilobytes for a small tree to many megabytes for a tree with tens of thousands of people. Big files are still plain text and still follow the same rules.
Does a GEDCOM file hold DNA data? No. DNA results and matches are not part of the spec. Your DNA data stays with the testing company, and no export file carries it.
References
- https://gedcom.io/ — The home of the GEDCOM standard, with the current 7.0 spec and the older 5.5.1 spec.
- https://www.archives.gov/research/genealogy — The U.S. National Archives guide to genealogy research and records.
- https://www.loc.gov/ — The Library of Congress, whose collections include census, local history, and family history material.