One of the reasons my Ancestry.com tree is public is so that I can make connections to “cousins” I don’t know. The relatives I have connected with have been a fantastic help in researching our common ancestors.

What is a reasonable way to determine how far I should go with the tree connections? For instance, should I stop with second cousin 1x removed, 1st cousin 2x removed, etc.? My concerns with have a very large tree:

  • -it will be meaningless to my descendants
  • -I won’t have time to research and source it correctly
  • -I still have so many gaps in my direct ancestors

  • -It can get expensive quickly to obtain records such as birth, marriage and death – yet I want my information to be as accurate as possible

  • Tom makes the excellent point that your tree should be as large as it needs to be to serve the purpose you want but not so large that you cannot manage it. In its present form, this not a question that will draw on expertise. Can I suggest that you amend it ask about alternative strategies for finding collaborative cousins that do not require the continual expansion of your published ancestor web?
    – Fortiter
    Nov 1, 2012 at 13:05
  • @Fortiter - That might be too much of an amendment, and would reduce the value of Tom's answer. I don't recommend this question be changed.
    – lkessler
    Nov 1, 2012 at 14:35
  • Does a tree (they are not trees, but let's let that one lie!) ever become too large to manage? In what way do they become unmanageable? Unmanageable by genealogical software or unmanageable for our brains? For me the only issue with large trees is having to do a "context switch" in my brain as I review the persons I haven't thought about for a time. My "master tree" has about 16,000 person records, some personas, most conclusion persons, and it could grow to a million or more as far as I'm concerned. I've never had management problems. Nov 1, 2012 at 15:29
  • @TomWetmore - Well, if you took 15 minutes to do research on each person in your tree and verify and document the conclusions for that person, then you can do 32 people in an 8 hour day, 160 people in a 5 day week and 6400 people in a 40 week year (allowing for time off to avoid insanity). It would take 2 1/2 years of doing just this to modestly (not even thoroughly) research 16,000 people. You could not do a million in a lifetime. That would need a collaborative effort.
    – lkessler
    Nov 1, 2012 at 15:38
  • 1
    I am not a fan of editing people's questions without permission. The original question was not primarily about quality, so I believe the editing was wrong. Isn't it backwards to edit a question to make it better match answers? Nov 3, 2012 at 0:06

6 Answers 6


Nobody can answer this for you. You are collecting genealogical information for your own purposes. If you are interested in your own ancestry you should concentrate on your direct lines. If you are interested in your surname in general you should concentrate on records with that surname. If you are interested in all descendants of a particular key ancestor you should concentrate on finding them. If you are interested in all of these (or something else) you have to prioritize your time and efforts in the different areas. There is no right or wrong answer on who to put into your "tree," which obviously is not just a tree.


I am pleased to see that you are concerned with the quality of the data in your tree.

For the purpose of leaving a meaningful tree to descendants, a limited but fully sourced tree is much more valuable than an enormous one that has not been verified with original documents. A public tree is a form of publication, so why would you not include source citations and explanations of how you drew your conclusions? Perhaps you should consider publishing such finished work in other forms (e.g. book, e-book, wall chart) and make provisions for maintaining your online tree after you are gone.

A private online tree could be used for all the work in progress that may or may not be correct.

For making contacts with cousins to collaborate, there is a place for publishing unfinished and speculative trees. In my opinion, many public online trees fall in this category. Sadly, as far as I know, none of the online tree platforms support this distinction.


One thing you can try is to only include your ancestors and their siblings in your online tree.

If someone has a connection to one of your ancestors or sibling, then they may be a cousin or have information on one of your lines. When they contact you, you can then share information with them just about that family line.

The further back your ancestor is (e.g. great-great-grandfather) that they connect to, the further back their relationship (e.g. 3rd cousin) will be, and the less of your entire tree that you will have in common.

You may then let that cousin be the researcher for that side of the family for you. That will allow you more time to spend on your ancestors and the other parts of your tree where you have less information.


Having gone through much of the same thought process as you describe, I trimmed my tree a few years back.* I liked the focus this brought to the work I did enough that I later trimmed again.

A journal-like approach

My database is now focused on my direct lines (ancestors), their children and about four generations of the descendants of those children. As a graphic, it is shaped a little like a Christmas tree.

Occasionally the lines extend a little further (for example, if I am chasing or tracking a little DNA or want to record something about a particular descendant/author/writer).

The software I use allows me to add associates/witnesses and assign roles, so that beyond those in the family tree, I have unrelated individuals in my database.

Different entries/different character

I am not an equal opportunity genealogist, so not every person in my database has been researched to the same extent. I spend more time on what are unproven or conflicted families than I do on other ancestors/families. My interest in associates tends to be even more specific.

*I did back up and save the database before I trimmed. In the new database, I set out to summarize what I was cutting in a comment tag added to the ancestors of those being "trimmed."

UPDATE: Tom writes, "I've never found a reason to trim a tree."

I didn't destroy the file, and probably could have kept the other individuals in a separate dataset rather than a separate file.

In my case, I had drifted further and further from the truck of my tree. When I was conducting work at a FamilyHistory library, I had done a better job of focusing and prioritizing the work, largely because I had to order films (ala, time and money) to further the research. That work had been focused on those more closely related to me. When more wonderful records became available online, I found myself working to solve problems based on readily available records. This meant that I entered more and more extended branches to my family tree, mostly based on "low lying fruit" (BDM and census). That changed when I set up my four or five generation rule of thumb.

Trimming a tree might not be the right decision for everyone, but it was the right move at the right time for me.

  • 1
    I have never found a reason to "trim a tree." There are no disadvantages to large databases anymore. One can depend upon software to limit reports and displays and other outputs to the sub-set of persons you are interested in at any particular point in time. On the other hand I can see many disadvantages to trimming a tree, the most obvious being coming across the same persons again during later research. Nov 1, 2012 at 15:23
  • 1
    +1 for acknowledging our attraction to "low lying fruit". Solid argument for trimming. Nov 2, 2012 at 21:53

Jeni asked: How do I Keep my Public Tree size reasonable?

I suggest having 2 data files.

  • the details for use by your relatives
  • an all encompassing file to attract cousins.

I have a file an a public site that has any connections I have identified as highly likely a match. From those, several individuals have contacted me advising they have the same person. On further contact with them, I have found they have much more on other of my relatives that have not been published, as they were too far from their needs to do so.


My concern with large public trees designed to find cousins is protecting the privacy of living individuals. I see too much criminal abuse (mainly social engineering and targeted phishing) in my day job in computer security so I'm probably more paranoid than most about finding my family's private information in other people's trees. The way I try to walk my own talk (ie not accidentally release someone else's info) is:

  1. use a pando like werelate for the master copy of all the information I don't mind being 'in the public domain'
  2. use myheritage with privacy set for my family and relatives, and only allow close relatives access. Myheritage allows you to have your data private but still find collaborators.
  3. Divide the world into family (people I actually know and see at family reunions), relatives (more distant cousins), and everyone else. I treat each separately wrt what to share.

It's still not perfect but best I could come up with.

One recommendation I would make is that you be sensitive to entering information on living people (which is a tradeoff since you are looking for collaborators). I have found private information about my family in the huge trees and have had a hard time getting it removed. Be sensitive and respectful of other people's request for their privacy.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.