National Genealogical Society Home Study Course Discounted Bundled Offer
Cyndi's List - Ten Years Old

Your Legacy Database - Should You Split It?

At some point, users of any genealogy computer program, including Legacy Family Tree, will ask the question, "Should I split my database?"

The reasons given for wanting to split a database are many:

  1. you may want to keep your spouse's genealogy in a separate database
  2. you may want a separate database for each line you're working on
  3. you may want to share just a portion of your database with another relative/researcher
  4. you feel that splitting your database would make more room in your main database

Before making the plunge and splitting your database, consider these cautions:


Suppose you separated your genealogy from your spouse's, and they are now in two separate family files. As your research progresses, you get further and further back in time. You begin researching a family that seems all too familiar - maybe you've researched it before. In fact, you had previously researched it - on your spouse's side.

The further back you trace, the chances are higher that you and your spouse have common ancestry. Such is the case with my wife and me. Back in the 1580s, we share a common ancestor - Stephen Hopkins is my 12th great-grandfather. He is also my wife's 12th great-grandfather. If I had split my database so that my wife's database was not included in mine, I may not have picked up on the fact that we were 13th cousins. I may have even duplicated our research.

This is one of the reasons why I keep all my genealogy in the same database.

Duplication of data entry

When I first started my research, I had eight separate databases. One for each of my eight great-grandparents. I thought it'd be easier to research this way. The descendants of these great-grandparents number into the thousands. In fact, because I am a descendant of each of these great-grandparents, I entered myself into each database.

The challenge this presented was that every time I needed to update my information, or I became a new dad, or even if I needed to update information on any of my cousins, I had to update the same information in at least two of the databases. If I needed to update my personal information, I would have to update it in all eight. Chances are greater that I would make a typo in at least one of them.


If my database were separated from my wife's, printing a pedigree chart or other reports becomes challenging. For example, I could not print cascading pedigree charts beginning with my son, Evan, because on the first chart, only his paternal side of the family would appear. If all my genealogy were combined into one family file, this would not be a problem.

But won't it get too big?

We have "test databases" of Legacy where we've linked over a million individuals. There are two potential challenges when working with databases of this magnitude. First, navigation becomes a bit slower. Second, the size of the database is considerably larger than one of just a few thousand. However, we haven't run into too many researchers that claim to have researched a million individuals - I'd love to see their documentation....

It's still okay...

There may still be good reason to split your database. I've created a separate research database to help me in tracking all the different Alanson Browns. Once I have positively identified my Alanson Brown, using Legacy's split screen tool, I can drag and drop him and his family into my personal database.

I've also created a separate database where I indexed the tax lists of Washington County, Tennessee from 1778-1850. Now doesn't that sound like an exciting way to spend a weekend?

So before you go ahead and make your split, carefully consider the above implications.


Feed You can follow this conversation by subscribing to the comment feed for this post.

When doing researching, prefer the single master file. When doing reports, want to stay within a single family line. Would be "nice" if when posting new information to the master list that it automaticly roll to the separate file as well. One can dream.

I agree with everything you said, and that is how I run my research as well. I keep my main database file in Legacy, and the only GEDCOM files I import are my own.

I keep research databases for some branches of the family, where I put in speculative connections etc., and move it to my main file when I'm more certain about it.

I too do something very similar in that I have my own master database and also another which is in fact a copy of my brother in law's. These two I manually link to create a joint website.

Additionally I have another database which contains many individual trees some of which may someday connect to mine!

I really enjoy your tips on how to use Legacy more effectively. By nature,we tend to not read/review throughly therefore missing an easly way to accomplish a task.
By you pointing out helpful hints, the program becomes more valuable.

Thankyou for the timely hints. Several years ago I went down the seperate database path for different family lines and ran into a number of problems, some of which you mention. I ended up combining them into one master database, creating a few problems for me along the way. All is now well, however, and I now use a similar system to what you do. I highly recomend a single database file.

Been there, done that. I don't split any more.

I agree with maintaining a single database, and maybe a second for "not yet certain" researches; one problem with this decision is how to give paternal (or maternal) side relatives what they care about (pictures, documents, etc.); the alternative is giving everything to everybody ...

This is valuable advice! I learned the hard way that splitting was not the way to go, and that one large file would be better. I'm sure we all have "overlapping" ancestors--especially if families stayed in the same area for several generations. With Legacy, it is easy to split off one segment if it becomes desirable. Also, having one correctly-formatted master source file is much easier than several!

Your article is very informative. I, too, had separate databases for my parents' families and found it hard to remember to make changes/additions in both so I merged them and am much happier.

Hi Everyone,

Great topic! I had my wife's and my family separate and decided to merge them specifically for my kids' sake. They are asking more questions about both sides of the family and having the files merged into one, makes explaning and demonstrating easier.

I also have been down the separate surname files, but then I started before the computer when everything was done by hand. Even when genealogy programs and the computer arrived I kept separate files. I have found where I have updated some and not the others. Now I am in the process of making one research database with both my late husband and my side in one file along with other families that came from the same area in Russia as my ancestors. It is not fun to have to look for ancestor when you have to look through many files to find them and you do not remember under which surname to look for them.

Thank you for having a note section either general or research that is unlimited to the amount of information you can type in. I do a transcript of my documents here such as census, Wills, church records, death certificates and etc.

Agree in the main but I am on a second marriage so as my present wife`s tree grew I decided it was time to split it. I believe this is a valid reason.

I have about five different family files. This is from when I first began and did not know how they were all related - just knew that they were. Now I would like one data base, but do not know yet how to successfully merge. That will be my next learning curve.

I have a database on the Belgian royal family,with ancestries of related persons,by now ,there are
399.000 people in it -all
interlocked,splitting thát
up would be foolish,indeed.
But,I wonder:the data files of that base is 560Mb
by now,-according to the Legacy help file,the maximum is 1Gb.
When will I get into trouble -and what error messages can I expect? ("File too large" or something?)


Ken M. suggested that you first do a File Maintenance on your file. This will compact the database and reduce the file size tremendously.

I have kept a single master file and as my wife (second marraige for both of us)is gradually starting to do her family history I had considered separating them. From the comments in this reference I will try to keep them together though on two separte computers. We have a home network. How does one keep the data clean and shared? I would value comments or guidance.
Would alos like to use tabloid sheets for certain reports and have more generations per sheet? Have I missed something?
Ed Somers

I would like to contact the person who wrote the article on whether to split data bases who said, "I've also created a separate database where I indexed the tax lists of Washington County, Tennessee from 1778-1850. Now doesn't that sound like an exciting way to spend a weekend?" We have lots of relatives in Washington County, Tennessee and some of them are Browns.

I run two databases- one on my biolgical father's family, and then one with my mother's family and my adopted stepfather's family. This makes it easier when dealing with the seperate families- while both acknowledge the other, they don't care about the other spouses family, when they want copies of gedcoms, they are all from different areas of the continent, so far I've had no overlap.

I also started a seperate database for my husband's family after we got married, as once we got past his grandparents, we had very sketchy and unverified info. Now that I have verified names, marriages and dates, I am planning on importing it into my primary file, as there is potential for overlap due to families being from same areas in europe.

I used to do four files: father's paternal and maternal and mother's paternal and maternal. Now I'm in the process of combining them.

If a relative wants info on just one or two lines I can export to another file by using the tags.

It took a while to retag everyone but the shortcut in the Name List really helped.

Now I add the family tag to each person as I enter them into the database.

My tagging system is: dad's paternal side is Tag 1, maternal side is Tag 2. Mother's paternal side is Tag 3, maternal side is Tag 4.

Another complication is dad's step-father. So his line is Tag 5.

And I use Tag 9 for the unrelated people who may become related later.

So far this is working well.

With respect to the above article re 'splitting family files'. I would appreciate knowing the exact steps necessary to carry out this action.

I have TEN separate family files. I did not know any better when I started. Is there by any chance a tutorial on how I could merge them all -- or at least all the ones on my family and then a separate one for my husband's family?

Thanks for any advice.

Karen McCain

For a single researcher, a single database is simplest, but the thread's premise begs the question. Once there's more than one hand in the kitchen, there's no escaping the complexities of multiple databases. There's just the glum and growing understanding of the inadequacy of the tools available for working with multiples through Legacy, other genealogy programs and obsolescent exchange standards(?)like GEDCOMs.

Pretty much any large genealogy database overlaps or includes databases being developed by other authors. Often, early circulation of a new family tree triggers further authorship by cousins who have their own priorities, histories, genealogy links and privacy issues. Simple genealogy data can be relatively compact, but that's no longer true when genealogy drifts into family history. Further, manipulating kilobytes of data is one thing, gigabytes quite another. Add video and terabytes are predictable...

What we need is tools and standards to let us very efficiently manage our own patch (in as many versions and pieces as we need) and manage overlap and exchange with the other databases we connect to. Issues include synchronization of versions, standardization and transmissions of rules and publication permissions etc.

Lacking that infrastructure, I personally develop and document a core database of less than 2,000 which connects via carefully maintained interfaces to other databases shared with or authored by other workers, or set up by me for research or security convenience. A few notes, the Intellishare feature of Legacy and sometimes direct GEDCOM editing then let me readily assemble a larger database when and as needed.

Great article. I have buried any half-formed ambition to split my modest database. I like the idea of a database to collect people who may become related, so to speak; but how is that accomplished? Don't people have to be connected to somebody somewhere to find a position in the database? Or did I miss something ...again?

Bruce - no, individuals in the database can be entered as unlinked individuals. Just go to Add > New Unlinked Individual. The Tree Finder at View > Tree Finder gives you a list of these unrelated persons so you don't lose them.

The comments to this entry are closed.