DNA Matches Explained: How to Read Your Match List

A DNA match is another person in the same testing database who shares measurable pieces of DNA with you, which means the two of you almost certainly descend from a common ancestor somewhere. Your match list ranks those people by how much DNA you share, measured in a unit called the centimorgan (cM) — the more centimorgans, the closer the relationship. That number gives you a shortlist of possible relationships, never a single answer.

That last point is the one most people miss on day one. Shared DNA is a measurement; the relationship label next to it is a guess built on top of that measurement. Learn to read the number and the labels stop being confusing. If you already have a cM figure in front of you and want to see the realistic options, drop it into our shared cM / DNA match calculator and you'll get the list in a second.

What Is a DNA Match, Actually?

When you take an autosomal DNA test, the lab reads several hundred thousand specific spots in your genome. The company then compares your reading against everyone else in its database, looking for long stretches where your DNA and theirs are identical. Those stretches are called segments, and a person who shares at least one qualifying segment with you becomes a match.

The logic behind it is simple inheritance. You got half your autosomal DNA from your mother and half from your father, they each got half from their parents, and so on. Two people who inherited the same chunk from the same ancestor still carry a matching copy of it generations later. Finding that chunk in two living people is strong evidence they share an ancestor.

Every company sets a minimum before it will call something a match. Most use a floor somewhere in the range of 6 to 8 cM for the smallest reported segment, and AncestryDNA raised its threshold to 8 cM in 2020 specifically because smaller segments were producing too much noise. That threshold is why your list has thousands of names rather than millions.

One thing your match list is not: a family tree. It's a ranked list of people whose DNA overlaps yours. Turning that into a tree is the work, and it's genuinely fun work once you know how the numbers behave.

What Do Centimorgans Measure?

A centimorgan is a unit of genetic distance, not physical length and not a percentage. It measures how likely a stretch of DNA is to be broken up and reshuffled when it's passed to the next generation: two points 1 cM apart have roughly a 1% chance of being separated in a single generation.

For match-reading purposes, the practical translation is easier. Humans have roughly 6,800 cM of autosomal DNA (closer to 7,400 if you include the X chromosome), and every relationship has a typical share of that total. Parent and child share half of it. Full siblings average half. First cousins average about an eighth. If you want the mechanics in more depth, our guide to centimorgans explained walks through what the unit is actually doing.

Here's the part that surprises people: shared DNA is random past the parent–child link. You and your sibling both got 50% of your DNA from each parent, but not the same 50%. That's why one sibling can share 2,500 cM with you and another 3,100 cM, and why two people who are both your second cousins can show up 200 cM apart on your list.

Total Shared cM vs Longest Segment

Every match shows two numbers worth reading, and beginners usually look at only one.

Total shared cM is the sum of all your shared segments. It's the headline number and the one that drives relationship predictions.

Longest segment is the size of the single biggest unbroken piece. It tells you how recent the connection is likely to be, because segments get chopped smaller with each generation of inheritance. A match with 90 cM in one 60 cM segment is a very different animal from a match with 90 cM spread across twelve tiny fragments — the first points at a real, findable, relatively recent ancestor; the second often points at deep, tangled, population-level relatedness.

Rule of thumb worth internalising: if the total looks like a second cousin but the longest segment is only 9 cM, be suspicious. Genuine second cousins nearly always share at least one chunky segment of 40 cM or more.

Shared cM to Relationship: The Reference Table

These ranges come from the Shared cM Project, the crowdsourced dataset built by genealogist Blaine Bettinger from tens of thousands of documented, known relationships and hosted as an interactive tool at DNA Painter. It's the standard reference in the genealogy community precisely because it's built from real reported results rather than pure theory.

Two features of that table do all the damage to people's expectations. First, the ranges overlap heavily — 250 cM is compatible with at least half a dozen relationships. Second, once you get past third cousins the bottom of the range is zero, which means the relationship exists but the DNA doesn't show it.

For a fuller walkthrough of how to read the grid and squeeze more certainty out of it, see our shared cM chart guide.

Why Your Closest Match Matters Most

If you only do one thing with a new match list, look at the top of it. Your highest match sets the ceiling on what the rest of the list can tell you, and it usually determines your whole strategy.

• A parent, child, or full sibling at the top means you can eventually split your entire match list into maternal and paternal sides, which is the single most powerful sorting move in DNA genealogy.
• A first cousin or closer at the top means the connection to your known family is short and you can usually identify the shared grandparents quickly.
• Nothing above about 200 cM means you're working with distant cousins only. Progress is still possible, but it'll come from clustering many mid-sized matches rather than from one decisive name.

Your top match is also the one worth double-checking against your paper tree first. If it lines up with what you expect, your tree and your test are both behaving. If it doesn't, that's information too — and it deserves careful, unhurried handling, which we cover in how to confirm a DNA match.

Close Matches vs Distant Matches

Think of your match list as three tiers, each of which rewards a different approach.

Close matches (roughly 200 cM and up)

These are almost always identifiable. The relationship possibilities are few, the segments are big, and the shared-match lists are informative. Start here every single time, even if your actual research question is about a distant branch — close matches are what let you label the sides of your tree.

Middle matches (roughly 40–200 cM)

This is the productive middle: real, traceable relationships in the second-to-fourth-cousin band. You'll usually need trees and shared matches to place them, and clustering tools help enormously. Most brick-wall breakthroughs happen in this tier.

Distant matches (under about 40 cM)

Thousands of names, mostly unresolvable one at a time. Many are genuine but connect through ancestors born in the 1700s. Some are false positives created by the algorithm stitching together coincidental similarity. Use them in groups, not individually.

A practical filter: if a match under 20 cM doesn't sit inside a cluster you already understand, don't spend an afternoon on it. There's better return on your time higher up the list.

Why Some Relatives Don't Show Up

This confuses more beginners than anything else. Four reasons cover almost every case.

They never tested. The obvious one, and by far the most common. Your match list only contains people who bought a kit and consented to matching.

They tested somewhere else. The databases don't talk to each other. Your first cousin at 23andMe is invisible to you at AncestryDNA. Uploading your raw data to MyHeritage, FamilyTreeDNA, and GEDmatch (all of which accept transfers) is the cheapest way to widen the net.

They opted out of matching. At 23andMe you have to actively opt into DNA Relatives; at other companies you can switch matching off. Some testers only want ethnicity results.

The DNA genuinely isn't there. Past third cousins, shared DNA becomes a lottery. The testing companies' own published estimates put the odds at roughly 90% for third cousins, around half for fourth cousins, and well under 10% by sixth cousins. A missing fifth cousin isn't an error — it's normal inheritance.

That last point deserves emphasis because it goes the other way too: absence of a match is not proof of no relationship beyond the close range. It's only meaningful evidence for close relatives, where a match is guaranteed.

How to Work Through a New Match, Step by Step

1. Write down the shared cM total and the longest segment. Both, not just the first.
2. Get the list of possible relationships. Run the number through the DNA match calculator or DNA Painter's Shared cM Tool.
3. Use age to cut the list. A match 30 years older than you can't be your first cousin twice removed downward. Generations are the cheapest filter available.
4. Check shared matches. If your new match also matches three people you've already placed on your father's maternal side, you've found the branch before you've found the person.
5. Look for a tree — theirs, or one you build for them. Surnames and locations do most of the remaining work.
6. Confirm the hypothesis with a second line of evidence. Records, an additional match on the same line, or a targeted test of another relative.

Step 4 is the one people skip and shouldn't. Shared matches are often more informative than the cM number itself, because they tell you which side of your family a match belongs to — something the number can never do on its own.

What Can Distort Your Match Numbers

A few situations reliably push shared cM totals away from the textbook ranges, and knowing about them saves you from bad conclusions.

Endogamy. In populations that intermarried within a closed community for centuries — Ashkenazi Jewish, Acadian, Low German Mennonite, many island and isolated rural populations — everyone shares small amounts of DNA with everyone else. Totals run high, predicted relationships come back too close, and a "second cousin" may really be a fourth cousin twice over. Our guide to endogamy explains how to adjust.

Pedigree collapse. Cousins marrying cousins a few generations back means you're related to a match through two paths at once, and the shared DNA adds up.

Company algorithms. AncestryDNA applies a filter called Timber that removes segments in regions where huge numbers of people match, so Ancestry's totals for distant matches often read slightly lower than the same comparison elsewhere. That's a deliberate design choice, not an error — more on how each company handles this in how accurate is AncestryDNA.

Half relationships. Half relatives share about half the expected DNA of their full counterparts, which slides them down into the next band and makes a half first cousin look like a first cousin once removed.

What DNA Matches Can and Can't Prove

Being honest about the limits is what separates good genetic genealogy from wishful thinking.

DNA can prove that two people share recent ancestry, that a close relationship is or isn't biological, and that a documented line is genetically supported when multiple independent matches line up on it.

DNA can't tell you which specific ancestral couple a segment came from without a tree, which side of the family a match is on without shared matches or a parent tested, or the exact relationship from a cM number alone. It also can't distinguish a full aunt from a grandmother from a half sister on the number by itself — all three land in that 1,300–2,300 cM band, and only age, documents, and context separate them.

And a human note: sometimes a match reveals something unexpected about a family — an unknown parent, an adoption nobody talked about, a sibling who wasn't mentioned. If that's where your results took you, go slowly. Verify before you conclude, and think carefully about who else the information belongs to. There's no rush, and the DNA will still be there next week.

FAQ

What does cM mean on my DNA match list?

cM stands for centimorgan, the unit used to measure how much DNA two people share. Higher totals mean closer relationships: parent and child share about 3,400 cM, first cousins average around 850 cM, and second cousins typically fall between 75 and 360 cM. It's a measure of genetic distance, not a percentage.

How many DNA matches should I have?

Most testers at a large company end up with tens of thousands of matches, the vast majority of them distant. The exact count depends on your ancestry: people with deep roots in a heavily tested population like the United States or Britain get far more matches than people whose families come from under-represented regions.

Why does my match's predicted relationship look wrong?

Predicted relationships are generated from shared cM alone, and cM ranges overlap. A single number can fit half a dozen relationships, so the company picks the statistically most common one. Age, shared matches, and family trees are what turn a prediction into an answer.

Can a DNA match be a false positive?

Yes, mostly at the small end. Segments below about 8 cM have a meaningful chance of being coincidental rather than inherited from a recent shared ancestor. Matches above 20 cM are almost always genuine, and matches above 50 cM effectively always are.

Why do my siblings have different matches than me?

Because you and your siblings inherited different halves of your parents' DNA. Full siblings typically share only about half of their DNA with each other, so each of you carries cousin segments the others didn't get. Testing a sibling routinely surfaces matches you don't have.

Does a bigger shared cM number always mean a closer relative?

Almost always, but not perfectly. Random inheritance means an unusually generous second cousin can out-share a stingy first cousin once removed, and endogamy can inflate totals across the board. Treat the number as a strong signal that still needs confirmation.

Start With Your Own Numbers

Your match list stops being intimidating the moment you realize it's just one measurement — shared centimorgans — dressed up with labels. Read the total, check the longest segment, look at the shared matches, and let age and geography do the rest.

When you've got a number in front of you and want the honest list of what it could mean, run it through our free DNA match calculator. It takes one number and gives you every relationship that fits, so you know exactly which possibilities you're choosing between.