This post is for people new to using DNA for genealogy who are trying to figure out which DNA matches to contact.
I don't provide information for people trying to find the parents of a living person or dealing with sensitive situations. This post, and all my DNA posts, are aimed at hobbyist genealogists working on "historical" DNA projects.
Sometimes the techniques differ for "living" vs. "historical" problems but sometimes you just need to approach the topics differently, especially when discussing contacting matches. If you need help dealing with a problem involving the parentage of a living person or a sensitive situation, please look for help specifically for your situation.
A great free site to start with is DNAAdoption (even if your situation is not actually an adoption).
You've got a lot of DNA matches!
How do you know which DNA matches to contact?
There is no point in trying to contact EVERY match. Especially when you're getting started, this isn't worthwhile because you won't know what to do with most of the information you might get.
When you contact a match, it is vital to ask them for something specific. Just contacting them and saying "Hi! You're my DNA match." usually just leaves them wondering why you're contacting them.
There isn't some magical secret you've missed that they will know. If you think you'll just contact your matches and they'll know what to respond back, they won't.
That means deciding who to contact starts with asking yourself why you're contacting them.
Before we go on, let me tell you that response rates to messages to DNA matches are really low. The less clear your message, the lower they will be. They are so low to start with, an unclear message easily means you won't get any responses. If you've received messages and didn't know how to respond, there's an example for you right there.
Also, I'm going to give some pretty specific recommendations depending on your "why." You need to determine your why but you can always keep things simple by contacting your closest DNA matches. Once you identify the reason you want to contact DNA matches, you might realize you can pick and choose which matches to contact, though.
Know Why You are Contacting Your DNA Matches
To figure out why you're contacting a match, you may need to start with why you even took a DNA test.
Perhaps it was curiosity in general and now you want to pursue your family history.
You may have had a specific family history question in mind, though.
There are a few basic reasons people contact their DNA matches and you need an idea of these before you decide who to contact.
Which DNA Matches to Contact for Family Photos or First-hand Stories
You might want to connect with cousins from known family. Maybe you just want to connect. Maybe you're wondering if they have family photos or information you can share.
In this situation, you'd want to contact matches where you know how you are related. You most likely want these to share a more recent relative. Think about it this way, if you're looking for family photos, how far back will those exist? For poor families, not very far back (In my family we have a few photos from the 1920s but most start in the 1940s, my husband's family, who wasn't poor, has older photos. I know some families have many early photos).
You might find 19th-century photos owned by distant relatives but is that where you want to start? Probably not. There aren't a lot of 19th-century photos in existence so you'd have to contact a lot of DNA matches to get a response from someone who has such a photo. Even a family Bible is similar. A Bible might exist but a family can have many photos but usually only has one Bible (I know, that's not exactly true but close enough).
So, if you want information or photos, you are probably looking for people who share your known family and family that lived in the 20th century.
Determine who to contact by starting with your own family tree. You can either go through your matches comparing the attached tree to your tree or…
Figure out what types of relationships (aunt/uncle, first cousin, second cousin, first cousin twice removed, etc.) are good candidates to have what you want. Now you can focus on contacting your closest matches of those types (not worrying about the more distant matches) and ask for information on your shared ancestor(s).
Note that DNA matches are "grouped" under headings that don't always represent the exact relationship and even for the sites that show a very specific estimate, this is always an estimate. Tools like "ThruLines" and "Theory of Family Relativity" also estimate the relationship and that info is presented separately.
Also note, a cousin that is three or four times removed, particularly first cousins three or four times removed, might share a 20th-century ancestor. These types of cousins share far less DNA with you. If the test results you work with are for a test taker that is very young or much older, you are likely to have useful cousins that share very little DNA.
Learn to find these cousins deep in your match list if your goal is first-hand information or family photos. At AncestryDNA you can try using ThruLines to find these matches but at other sites it can be harder. This is a goal where searching for a shared surname will be helpful but if the surname is very common, you still need to use other methods to narrow which matches you review.
Don't worry about finding cousins that are several times removed, initially, if you don't have specific information you need from them. This is why you need to know the purpose of contacting your matches.
For a purpose like finding family photos or 20th-century stories, your concern is what relative you share regardless of how much DNA you share with the match. For other purposes, you might still care about specific relationships but only if you share enough DNA.
Your match list is ordered by the amount of DNA you share (remember, the relationships provided are estimates). You can share more DNA with a third cousin than a first cousin three times removed but those are very different shared ancestors. Working with DNA matches is time-consuming so you want to prioritize what you do and that means knowing why you're doing it.
Recommendation
If your goal is finding photos, stories, or otherwise contacting matches who share your 20th century known ancestors:
- Determine which shared ancestors you are interested in. Treat each distinct ancestor/couple as a separate project.
- Easiest Option: Focus on your closest matches who have an attached public tree and simply compare your tree to the match's tree until you find someone who shares the desired ancestor.
- More Involved Option:
- Write down the types of relationships you are most likely to have with people sharing the desired ancestor.
- Use a chart or other reference to determine how much shared DNA you could have with people of these relationships (given as a % or shared cMs).
- Write down known surnames, including surnames that might not be part of your tree but would be in your cousins' trees. Consider married names, 2nd marriages, etc. If you have very common surnames, try and identify uncommon surnames that could help you search for matches sharing a specific branch/ancestor.
- Determine the combination of surname searches and shared DNA you can use to find the matches that share the desired ancestor(s).*
- Stay organized.
*Part of determining the combination of surnames/shared DNA is so you do NOT get overwhelmed. You can add more surnames or shared DNA later to identify more matches.
I can't get too specific on exactly what to do because it depends on what your relationship to the ancestor of interest is. If you're looking for people who share your grandparents and you're pretty young, you won't have a lot of matches fitting that situation so you might be able to manually review all your closest matches. In such a situation, it may not be possible for you to have a first cousin three times removed that shares your grandparents (you might have a first cousin three times removed that shares your 3rd great-grandparents, they are the match's grandparents).
Conversely, if you're much older (or the test taker for the test you are using is) AND you were the baby of a large family, you might have lots of matches that share your grandparents. You might start by manually reviewing your closest matches and later develop a search strategy using a combination of surnames and shared DNA.
Learning to determine the types of cousins of interest is a skill you need to develop. Use a cousin chart and just keep practicing. Start simple and add on to your project as your skills improve.
Which DNA Matches to Contact for Family Story Clues
So the next thing you may be trying to get from matches are clues. This can be clues to determine if a family story is true or clues to finding a missing relative or clues to busting a brick wall.
The distinction for this "why" is you're looking for clues from KNOWN relatives and it's in a specific branch of your tree.
If you're just getting started using DNA, you need to focus on matches with trees. Just as with the previous example, you need an idea of what types of relationships those matches would be to you (or to the test taker for the test you're working with).
This can help you ignore matches that are clearly too distant but also helps you develop a quicker eye for determining if a match is likely helpful based on their tree (you'll start to know if you don't see a shared ancestor in a tree of a certain size if it's likely the shared ancestor is too far back to help with this problem---this takes practice but you start by knowing what your goal is).
When you're first getting started, you can't really focus on matches without trees if your interest is in a specific branch.
You can absolutely do this but it takes time. To get started, start with your matches with trees.
Besides starting with the matches with trees, you need to be organized.
If you don't know how else to organize matches, create lists of which matches belong to each branch and/or enter notes in the notes field at the testing site (I highly recommend entering brief notes at the testing site no matter what. It makes it easier to see the situation/relationship while looking at lists of matches or shared matches at the site).
You likely want to create a list of matches related to a specific branch so you can see at a glance who you've contacted or not and who has responded or not. This becomes more complex as you have other goals.
IF you want to be able to track who you've contacted, who's responded, and the different purposes or goals, keep a separate list for each goal (i.e. a list in addition to any notes entered at the testing site).
The testing sites don't offer features that can handle complex, customized projects. You may have some matches that belong in multiple lists and this can get difficult using the in-house features, only. Using only the in-house note field or other features might work for now but you want to be ready for your later success. In particular, the colored dots at AncestryDNA will allow you to create customized lists but you only get so many dots. As your needs grow, the dots either stop being helpful or have to be adjusted. I would use them but also keep a list. You can change the meaning of the dots so your list serves as your backup.
Just to be clear, the suggestion I've just made is about finding clues from people who you know how they are related to you.
Recommendation
If your goal is clues, i.e. contacting matches who share your 19th or 20th century known ancestors:
- Determine which shared ancestor/branch you are interested in. Treat each distinct ancestor/couple as a separate project.
- Keep a list of matches related to each ancestor/branch. You can use this to also note who you've contacted and who responded as well as their known or suspected shared ancestor.
- Keep brief notes in the in-house notes field for matches where you've determined the shared ancestor, excluded a shared ancestor/branch, or other notes that would help when you review a list of matches online.
- Use the suggestions from the previous recommendation section. Keep in mind that if you're focusing on more distant shared ancestors, you likely have more matches of interest. Remember to start simple and add on so you don't get overwhelmed to start with.
Which DNA Matches to Contact for New Ancestor Clues
The holy grail of DNA use (for genealogy) is finding new ancestors. This actually isn't a good project for someone just getting started using their DNA results as it is all about getting clues from matches you don't know how they are related.
I'd recommend starting by gathering information from people you know how they are related. A lot of identifying a new ancestor is excluding matches that clearly are not related to your goal (yes, you need to focus on finding a specific new ancestor, not just any new ancestor just like you need to know a specific "why" for contacting matches).
I mention New Ancestor Clues here because I've noticed a lot of DNA users totally fixated on identifying their known ancestors and basically throwing out all the matches where they don't know how they are related! Most do want to find new ancestors so this makes no sense.
Recommendation
If your goal is identifying a new ancestor (a specific one or any new ancestor):
- Start by identifying the shared ancestor for your closest matches and making note of it.
- Build trees for your closest matches that don't have a tree or a large enough tree (see below).
- LATER you will need to determine which new ancestor you are trying to identify which involves ignoring matches clearly not related to that branch. You do this by knowing which branch matches do belong to, i.e. you'll need to have already identified the shared ancestor for as many matches as possible.
The number one "thing" I think genealogists need to do to use DNA, regardless of why they are using it, is building trees. Until you're looking for a new ancestor, you can contact matches for other things before you have to build trees, such as the aforementioned family photos or clues involving known ancestors. Building trees helps with these goals as well but if you're only interested in new ancestors, it's a must.
Building Trees
The biggest reason to contact matches is to get information to build a family tree for them. This obviously applies when there is no tree attached, the tree is private, or the tree is very incomplete or obviously wrong (hint: don't bluntly tell a match their tree is wrong, be polite!).
You might be trying to figure out if a match is a known relative (shares a known ancestor) or not (the shared ancestor is unknown and therefore potentially a new ancestor).
Finding new ancestors is not a beginning task but you might not have trees for any relatives of the types to help with the first two, beginner-friendly, goals so you need to build trees.
Before you contact a match to learn about their family tree, you still want an idea of your "why" for using DNA. This is how you narrow down which matches to contact. Also, if they respond, you want to know if you should ask for other information (maybe they look at your tree and respond "oh, you're my first cousin once removed, my grandparents Smith are your great-grandparents" and now you want to ask if they have family photos---be ready to ask about the family photos while they're communicating with you, not weeks later!).
When you first get started it's pretty involved to try and target specific branches of your tree (if you can, it usually means you have several close relatives tested).
Start by contacting matches (who don't have a tree or not enough information in their tree) sharing about 75 cMs or more. If you have lots of matches you could contact at this level, aim higher (a second cousin on average shares well over 200 cMs---they share your great-grandparents where 75 cMs is more like a 3rd cousin who shares your great-great-grandparents).
You might be in the opposite situation where you have very few closer matches (maybe no one even sharing 100 cMs). In this case, you may not want to focus on contacting matches unless you have your own tree well developed. Your matches are obviously more distant if they mostly share less than 100 cMs. Consider building your own tree and learning more about doing genealogy (if needed) and learn about using DNA before putting in a lot of time contacting matches.
Whether to delay contacting matches is up to you but DNA will not magically give you a family tree. The closer matches you have (and the more close matches) the easier it is to build a family tree from DNA. Fewer close matches doesn't prevent you from using DNA, you just have to be more skilled.
If you have a family tree that goes back to your great-great grandparents, obviously you might be able to fit matches sharing under 100 cMs into your known tree. If it only goes to your great-grandparents, or has many branches missing, getting a tree for a match may or may not help you.
Recommendation
If your goal is building trees:
- Pick a cM level appropriate for your match list.
- Option 1: contact matches without tree information sharing 75 cMs or more
- Option 2: Too many matches over 75 cMs? Move the number up to 100 or 200 cMs. You can also start with just matches without a tree vs. matches with a private tree or incomplete tree to help narrow your options.
- Option 3: Too few matches/no matches sharing 75 cMs or more. First, determine if your own tree is extensive enough to find a shared ancestor sharing less than 75 cMs (you may need to learn more about using DNA to determine this). Build out your own tree, if needed, before focusing on contacting matches or work on both at the same time.
- You need as robust a family tree for yourself (or the test taker, if you're using someone else's test) as possible if you're looking for new ancestors. When you build a match's tree, you want to build as little as necessary. The better your tree, the faster you'll recognize a shared branch in someone else's tree. That means keep building your tree, especially adding siblings and cousins to "flesh out" your tree. Add to your tree when you find a shared ancestor in a match's tree.
- Stay organized. Have a system for building matches' trees and knowing which match the tree belongs to. This is in addition to any organizing suggestions I've already made related to tracking who you've contacted. I use one online tree for all my family's DNA tests. I copied my own tree information and then I build floating trees for matches in that same tree (connecting them to my tree once I find a shared ancestor). Creating a separate tree file for every DNA match makes it too difficult to connect people together (and that's the point).
Have you found your "why?"
I've discussed three main reasons why you might want to contact your DNA matches and which matches you should start with for each reason.
The reasons are:
- Find family photos or more recent family stories (focus on 20th-century ancestors)
- Find clues to family stories or clues related to your known ancestors (focus on 19th and/or 20th-century ancestors---but specifically known ancestors).
- Find clues to new ancestors (unknown ancestors).
If your interest is unknown ancestors or you can't find enough matches with trees showing your known ancestors, I recommend focusing on building trees for your matches. You still need to know which of the three reasons above you are interested in to decide which matches to contact.
Remember, it can take time to be able to target a specific ancestor or branch of your tree within your DNA matches. You have to start by "sorting" your matches in some way so you determine which ones will help with your specific goal (and you have to identify a goal before you can sort your matches). DNA is not an instant answer to a genealogical problem. It is one tool you can use. Sometimes it's the best tool for a specific goal, sometimes it's helpful, and sometimes it's not the best tool. You'll need traditional genealogy skills, regardless, and you'll need to keep growing your skills for traditional genealogy and DNA.
When you're ready, I also have posts with suggestions of what to say to your matches. Find the "beginner's" post, here. Find examples of AncestryDNA messaging templates (for more experienced genealogists), here.
Are you new to genealogy/DNA and trying to use DNA for a "historical" genealogical purpose? If you've got a different goal than one of the three listed above, I'd love to hear about it so I can try and provide more ideas for how beginners can get started contacting their matches. Leave a comment!
