Increase Your Ancestor’s DNA Coverage and Expand Your Opportunity for Discovery

Finding the right group of autosomal DNA matches is critical in moving your genealogy forward and breaking through stubborn brick walls. However, sometimes these important matches are not found within your match list but in the lists of other cousins who share the same ancestor. Search as you might, you can’t find a match that doesn’t exist no matter how deeply you review your match list.

Because DNA inheritance is random from generation to generation, it isn’t assured that enough of an ancestor’s DNA remains in our genome to confidently match other cousins who share the same ancestor. This is why it’s important to test as many cousins as possible or ask them to share their match list with you – this concept is referred to as coverage.

In this post, DNA coverage is defined and an example from my own research is provided, which accentuates the importance in taking the time to review your other cousins’ match lists. Methods for expanding an ancestor’s coverage are additionally outlined.

DNA Coverage

According to genetic genealogist Paul Woodbury, coverage is defined as “the amount of a research subject’s DNA that is represented in a DNA database through their combined tested descendants”.[1]

The importance of increasing the DNA coverage for an ancestor comes from the randomness in which DNA is inherited. As introduced earlier, we may not have inherited the same DNA segment(s) from an ancestor that others in the DNA databases have inherited. As such, we will not find the group of matches helping us discover the identity of our elusive ancestor. However, if we review the match lists of other cousins who similarly descend from the ancestor being researched, we may find those matches.

When investigating more recent ancestors like a grandparent, there’s not likely many match list differences between first cousins. However, for older ancestors like 2x great grandparents, third cousins may share no DNA segments with us or as much as 234 cM.[2] So, as our research pushes further back into our family trees, DNA coverage is not just a nice-to-have but critically important.

To improve the efficacy of your research, consider reviewing the match lists of cousins who descend from different child lines for a researched ancestor than the one you descend as it maximizes the possibility in discovering different matches not in your match list.

For those interested in writing genealogical proof arguments or research reports where DNA coverage is an important evidentiary contribution toward establishing an ancestor’s identity or relationship, you may wish to report the percentage of your target ancestor’s DNA coved by your analysis of the match lists for multiple cousins. Paul Woodbury provides detailed explanations for calculating these percentages,[3] but reporting these percentages are not necessary for the average genealogist not intending to publish or disseminate their research more broadly.

I have many examples from my past research demonstrating the importance of DNA coverage, including how I discovered my paternal Wilson line was from County Fermanagh, Northern Ireland and that my 5x great grandmother’s father was Thomas McMasters. However, the most recent efforts with DNA coverage come from my Hill line, which demonstrates the importance of checking the match lists for more than one cousin.

Coverage Comparison between First and Third Cousins

Readers of my blog are likely familiar with the research on my Hill ancestry from Dorchester County, Maryland. Here, I used the DNA match list of a first cousin once removed to discover a genetic network, which permitted me to theorize that my Hill line from Lycoming County, Pennsylvania in 1805 originated from Dorchester County, Maryland in 1669.

The genetic network included several unlinked family clusters[4], which are a large group of matches who all descend from a single ancestor but for whom you are unable to establish a genetic relationship. These clusters included ancestral couples with the following surnames: Clark, Harris, Linn, Matkins, McKeel, Rumley, and Vickers.

To provide greater confirmation of the Dorchester County hypothesis, I replicated the initial genetic network analysis by finding evidence of the same network with other Hill cousins. While I found evidence of several of the previously mentioned unlinked family clusters (especially Clark) in many of my first through fourth cousins’ match lists, their representation was not as great as it was in the cousin’s match list I initially reviewed.

As a case in point for the unequal distribution of important matches across cousins, I provide visual representations of the genetic networks found in three of my cousins’ match lists (see image below). The network graphs are for my first cousin once removed (tester A), whose DNA I originally used to find the Hill genetic network, and his first cousin (tester B) and his third cousin (tester C). Network graphs organize autosomal DNA matches into clusters based on the number of connections a match has with one another. Clusters correlate to specific ancestral lines of the DNA tester and are more densely clustered together.

The network graphs were created with Gephi software and visualize Ancestry DNA matches ranging from 15 cM to 100 cM (see an earlier post for how the graphs were created). Displayed matches are limited to the parental side of their respective trees where their most recent common Hill ancestor, Susan [Hill] Wilson (1817-1859), is found.

Highlighted below in the abbreviated family tree are the ancestral lines associated with my Hill line represented in the Gephi network graphs above. My previous research, which isolated the genetic network to the Hill side of the Hill/Winland union, produced seven previously mentioned unlinked family clusters, of which the origins of the Matkins, McKeel, and Rumley lines have subsequently been tied to Dorchester County, Maryland from the late 1600s to the early 1700s.

To put a finer point on the Gephi network graph figure presented earlier, take note that the graphs for tester A and tester B, which are first cousins. Despite their close relationship, tester B has no evidence of the same Hill genetic network comprising of the Clark, Harris, Linn, Matkins, McKeel, Rumley, and Vickers unlinked family clusters.

DNA tester C, which is a third cousin to both testers A and B, does show evidence of the Clark, Linn, Matkins, and McKeel clusters, but there are not as many matches as found with tester A. Nevertheless, tester C has several other matches within the Linn, Matkins, and McKeel cluster who are also from Dorchester County that I have yet to tie into the above families. They represent new opportunities to advance my Hill research that may ultimately confirm how all the unlinked families connect with one another.

When expanding an ancestor’s coverage, it’s important to seek out cousins of varying degrees of relatedness especially through different child lines of the target ancestor. It seems logical to expect that first cousins would have more similar matches than third cousins. However, in the preceding example, this was not the case emphasizing that DNA inheritance is random. The probability that cousins have more dissimilarities between their match lists becomes greater for common ancestors further back in their family trees.

How to Expand an Ancestor’s Coverage

There are two methods to increase an ancestor’s coverage. The first is to recruit cousins, who share the common ancestor, to take an autosomal DNA test. These cousins might be someone you know, or it might require you to contact a cousin who is more distantly related and for whom you are not personally acquainted. To accomplish the latter, it’s sometimes easier to convince people to take a DNA test who have an interest in genealogy or family history. I contact individuals through FamilySearch’s Relative Finder or those who have an online family tree indicating descent from the ancestor of interest. When at the RootsTech genealogy conference, I also use FamilySearch’s Relatives Around Me feature.

The second method to expand coverage is to ask those who have already taken a DNA test to share their match list with you. It’s not necessary to ask only your matches to share their list with you. In fact, I have discovered a greater number of newer matches for cousins who are not on my match list as they are more likely to have inherited different DNA segments from our common ancestor. Both Ancestry and MyHeritage make it easy to share match lists, but the same is not possible on FamilyTreeDNA.

When contacting individuals to either take a DNA test or share their match list with you, I have several recommendations for making the request:

1. Establish Trust. Build a relationship first before asking others to share or take a test. DNA is personal, so make a personal and meaningful connection first. This takes time.
2. Balance Information Sharing. Don’t overload the message with too much detail or family lineage. Briefly state how you’re related and/or provide a link to your publicly viewable family tree. Consider sharing very brief research summaries or information about the common ancestor that piques their interest. Offer to share more research if they are interested. Shorter messages are more fully read, and meaningful ones are more likely answered.
3. Be Personable. Share a little about yourself to humanize your message, but don’t get too personal. I like to say where I live, what I do for work, and perhaps a small bit about my family life related to the inquiry.
4. Follow up but be Patient. Document when you made contact and follow up in a couple of weeks or months depending on the situation. When contacting Ancestry users, I review their profile page to see when they last logged in. Ancestry typically states whether someone logged in today, 2-6 days ago, last week, last month, 3-11 months ago, or over a year ago. You’ll have greater success receiving a response from those who are more active. That said, be patient when waiting for a response – not everyone logs on daily, and many never respond because they have no interest, no knowledge, or have never logged back in since getting their results.
5. Use Snail Mail. If emails and within app messages fail to generate a response, try sending a letter the old-fashioned way – through the mail. Admittedly, finding a physical address is challenging but not impossible. Sometimes users identify their city or state of residence in their profiles. If their name is less common, you may be able to find them on social media or online directories, such as whitepages.com in the U.S. While the letter should be type-written for legibility, hand write the address on the envelope to differentiate the letter from junk mail.

Conclusion

Increasing the coverage for an ancestor is an exercise in patience and persistence. Genealogy research is incremental and expanding ancestral coverage to find additional DNA matches capable of breaking down stubborn brick walls is worth the time and possible expense. Just like elsewhere in life, not all match lists are created equal.

As visually portrayed in network graphs between first and third cousins, segment inheritance for a particular ancestor can be vastly different. Increasing the coverage of my Hill ancestor has provided me with additional DNA matches to investigate and a reinvigorated research agenda in Dorchester County, Maryland.

---

---

Sources

[1] Woodbury, Paul (2020), “Covering Your Bases: Introduction to Autosomal DNA Coverage.” Legacy Tree Genealogists. Accessed 23 May 2025 at https://www.legacytree.com/blog/introduction-autosomal-dna-coverage?fbclid=IwAR05Igiueb7_pXggPvLwrj5buVp0Y7peV43g1BpUwV6dHLIGkqSNtY0-w5U.
[2] Bettinger, Blaine T. (2020), The Shared cM Project 4.0 tool v4. DNA Painter. Accessed 23 May 2025 at https://dnapainter.com/tools/sharedcmv4. Centimorgans (cM) are units of measure for genetic linkage.
[3] Woodbury (2020).
[4] An unlinked family cluster is a large group of matches who all descend from a single ancestor but for whom you are unable to establish your genetic relationship. See the following blog post for a more in-depth discussion of unlinked family clusters: https://myfamilypattern.com/smallmatches/.