How to Merge Two GEDCOM Files Without Losing Data

Why merging GEDCOM files is risky

Merging two GEDCOM files sounds simple — load file A, import file B, save the result. In practice it is one of the fastest ways to corrupt a family tree.

Common problems include:

• Silent duplicates: The same ancestor appears twice with slightly different name spellings (e.g. "John Smith" and "Jno. Smith"), so the merge tool treats them as different people
• Conflicting dates: Two files give different birth years for the same person; one overwrites the other with no record of the change
• Lost source citations: Some merge tools strip source references from the secondary file
• Broken family links: A spouse in file A may link to a different family in file B, leaving orphaned children
• Encoding mismatches: One file uses UTF-8, the other Windows-1252, producing garbled accented names after the merge

A cautious workflow — analyse first, merge second — prevents most of these issues.

Step 1: Analyse both files separately

Before merging anything, upload each GEDCOM to GEDminer one at a time and note:

• Total individuals and families — your baseline counts
• Earliest and latest birth years — establishes the time range each tree covers
• Top surnames — surnames that appear in both files are merge hot-spots
• Top birth locations — overlapping locations suggest shared ancestors
• Tree Health Score — a low score in either file means you should clean it up before merging

Write these numbers down. After the merge you'll compare them against the combined file to verify nothing was lost.

Step 2: Identify overlapping ancestors

Open both files in your genealogy software side by side and look for overlap. The fastest way is to:

1. Pick the most distinctive surname that appears in both trees (use GEDminer's Surname filter on each file)
2. List every individual with that surname born in the same decade in both files
3. Compare birth places — exact matches are likely the same person; nearby parishes may also be the same person

Make a written list of suspected matches *before* you merge. This becomes your verification checklist afterwards.

Step 3: Clean each file before merging

A merge amplifies existing errors. Fix them in each source file first:

• Run Error Detection (Integrity tab) on each file and resolve impossible dates, missing vitals, and broken links
• Run the Duplicate Finder on each file individually — it is much harder to spot duplicates after merging
• Use the Location Standardiser to normalise place names so "London, Eng." and "London, England" don't survive as separate entries
• Run Vital Sharpener to tighten estimated dates that might cause false-positive duplicates

A clean source produces a clean merge.

Step 4: Choose your merge strategy

There are three common approaches:

A. Software-side merge (recommended for small overlap) Import file B into the program containing file A. Most desktop genealogy programs prompt you to confirm or skip each potential duplicate. This is slow but gives you full control.

B. Manual reconciliation (recommended for large overlap) Keep both files separate. Pick the better-sourced individual for each shared ancestor and copy facts from the weaker file across by hand. Tedious, but produces the cleanest result.

C. Tree-as-reference (recommended for "borrowed" data) Don't merge at all. Keep your tree as the master and treat the second file as a research source — cite it as evidence rather than absorbing its individuals.

Never use auto-merge tools that combine without confirmation.

Step 5: Verify the merged file

After merging, export the combined GEDCOM and upload it to GEDminer. Compare against your baseline notes:

• Individual count should be approximately *(A + B) − overlap*. A much higher number means duplicates survived.
• Family count should follow the same logic
• Tree Health Score should be similar to or higher than the original files. A sudden drop means broken links or missing vitals.
• Run the Duplicate Finder again — it will surface any merges you missed
• Check Tree Connectivity to confirm there are no orphaned branches

If anything looks wrong, restore from your backup and try a different strategy.

Always keep a backup

Before any merge, make at least two backups:

1. The original .ged for tree A
2. The original .ged for tree B
3. (Optional but wise) A copy of your genealogy program's native database file

Store these somewhere outside your working folder — a separate cloud drive, USB stick, or email to yourself. If the merge goes wrong, you can roll back without losing weeks of work.

A bad merge is recoverable. A bad merge with no backup is not.