People often frame this as a debate: DNA testing or genealogy records? But that's like asking whether a map or a compass is better. They answer different questions. DNA testing shows you where your genetic lines came from geographically. Paper records show you who your actual ancestors were — names, towns, occupations, the whole story. You need both to build anything real.
The confusion comes from marketing. DNA testing companies advertise with phrases like "discover your ancestry" and "find out where you really come from." That sounds like genealogy research. But what a DNA test actually delivers — ethnicity estimates, DNA match lists, migration pattern theories — is fundamentally different from what you get when a researcher pulls a 1900 census record with your great-grandfather's actual name and occupation on it.
This guide breaks down exactly what each approach gives you, where they overlap, and how to use both as part of a coherent research strategy.
When you spit into a tube and mail it to a DNA testing company, the lab extracts your DNA and compares it against reference panels — collections of genetic data from people with deep roots in specific geographic regions. The result is an ethnicity estimate: a breakdown of what percentage of your DNA matches populations in specific regions.
This is genuinely useful information. If you've always wondered whether the family story about Irish ancestry was real, a DNA test can confirm or challenge that claim. If you know nothing about your family because records were lost — as is often the case with African American genealogy — DNA testing can point you toward geographic origins that historical records can't reach.
But the ethnicity estimate is not a family tree. It's a probability distribution. The same company will often update your results as their reference panels improve, which means your "29% Irish" can become "34% Irish" six months later — not because your DNA changed, but because their algorithm changed. These are estimates, not facts.
The more immediately actionable part of a DNA test is the match list: other people who have taken the same test who share measurable amounts of DNA with you. The larger the shared segment, the closer the relationship. A parent-child match will share roughly 50% of DNA; full siblings share around 50% too; first cousins share around 12.5%; second cousins around 3.1%; and so on, halving roughly each generation back.
DNA matches are powerful — they can help you find biological family if you're adopted, or connect you with distant relatives who have already done the genealogical work and can share what they've found. But the match list only tells you that someone shares DNA with you. It doesn't tell you how you're related until you do the genealogical work to figure it out. A second cousin match could point you to the right family — or to a completely different family if your genealogical tree has errors in it.
Genealogy records are different in kind, not just degree. They are documents created for practical purposes — census-taking, property transfers, military enlistment, marriage licensing — that happen to contain personal information about real people.
A 1900 census record gives you a specific person: their name, age, birth month and year, birthplace, father's and mother's birthplace, year of immigration, occupation, whether they owned or rented their home, and the names and ages of everyone in their household. This isn't an estimate — it's a record that a government enumerator wrote down on June 1, 1900. You can go find that page and read it yourself.
The same is true for immigration records: a ship passenger manifest lists your great-great-grandfather's name, age, occupation, and the town he said he came from — on a specific date when he arrived in New York harbor. Marriage records show who married whom, when, and where. Military discharge papers show service dates and rank. Deed records show what property he owned and what he paid for it.
Each record is a data point. A family tree is the structure you build by connecting those data points across time. When you find a Kowalski in the 1900 census with three children, the 1910 census with five children, and a naturalization record from 1895, you've documented four generations of a specific family — not an estimate of where their DNA came from, but the actual people who lived and worked and raised children in a specific place.
Records have a disadvantage that DNA doesn't: you have to know what you're looking for. If you don't know your great-grandfather's name, the 1900 census won't help you find him. DNA testing can surface matches that point you toward a family you didn't know existed. But once you know the name, records are how you build out the full picture.
Neither column is "better." They're different tools for different jobs. The question isn't DNA vs. records — it's which one moves your research forward at each stage.
DNA testing earns priority in two specific scenarios:
You have no existing family tree and don't know your biological family. If you were adopted, or your family has no genealogical records, or you were otherwise disconnected from your biological family — DNA testing can surface biological relatives through the match list. A parent, sibling, or first cousin match will be immediately recognizable and actionable. The closer the match, the more useful the information. If your biological father has tested, you'll find him in the match list. If he hasn't tested, you'll find his relatives — and they may be willing to talk.
For African American genealogy in particular, DNA testing is often the most practical starting point because the paper trail was deliberately disrupted by slavery and discrimination. DNA can point you toward geographic origins (West African populations, for example) that the historical record can't reach. See our African American genealogy guide for the full picture of how DNA fits into that research context.
You have a documented tree but have hit a wall. If you've built back to a great-grandparent and found nothing in the census, DNA matches with other researchers who have worked on the same family can point you toward records or brick-wall-breaking connections you couldn't have found otherwise. The match list is most useful when you're already working within a specific family surname or region — you know what you're looking for, and the match narrows your search.
In every other scenario — and this is most scenarios — start with records. Here's why:
Records give you anchors: specific names, dates, and places you can work outward from. If you know your great-grandfather was Stefan Kowalski, born around 1861, working in steel, living in Chicago in 1900, you can find that specific record and use it to find the 1910 census entry, the immigration manifest, the naturalization record, and eventually a village in Galicia where his parents lived. Records provide the skeleton. DNA confirms whether the bones are in the right place.
Records are also free in many cases. FamilySearch.org covers US census records 1790–1940 for free. The National Archives has the 1940 census at 1940census.archives.gov, free. County courthouses hold deed records and vital statistics. The free genealogy research guide covers the full toolkit.
A $99 DNA test gives you one useful data type. A free afternoon with FamilySearch can give you names for four generations of ancestors.
Here's the research sequence that gets you the most information:
This sequence isn't arbitrary — it's how the two approaches work best. Records give you a tree. DNA confirms the branches are connected correctly and finds new branches you didn't know existed.
Father's Day is June 21, 2026. If you're still deciding between a DNA kit and a genealogy research service, consider what you're actually giving the father in your life.
A DNA test gives him percentages: he's 28% Irish, 19% Scandinavian. Interesting at a cocktail party, but not a story he'll come back to. A genealogy research report — built from the actual census records, immigration manifests, and property deeds that document his ancestors — gives him names, towns, occupations, and a narrative. His great-grandfather who arrived in New York in 1892, worked in a Chicago steel mill for 30 years, and raised five children in a neighborhood where his neighbors were from the same village in Poland. That's a story he'll tell for the rest of his life.
KinLore builds that report — a 3,000+ word narrative based on the actual documents we find. One-time purchase, no subscription. Order at kinlore.ai/gift and it arrives before Father's Day.
The question isn't which is better — it's which is more useful for where you are in your research. If you don't know your ancestry at all, DNA can open doors records can't. If you want to build a documented family tree, records are faster, cheaper, and more specific than any DNA test.
For most people, the right approach is: start with records, use DNA to verify and extend, and treat your ethnicity estimate as interesting context rather than actionable genealogy.
And if you want the full picture — the actual people, the actual documents, the actual story — without spending months learning to navigate census databases and immigration archives yourself, that's exactly what a KinLore genealogy report delivers. See what we found for other families →
A DNA test tells you which populations your genetic material most closely matches, based on reference panels of people with deep regional roots. It identifies DNA matches — people who share measurable amounts of DNA with you — and estimates how you're related. It does NOT tell you who those ancestors were by name, where they lived, what they did for a living, or when they arrived in the country. Think of it as a map of regions, not a census of people.
Genealogy records — census returns, birth and death certificates, immigration manifests, military discharge papers, deed records — tell you specific, named individuals: where they lived, what they did for a work, who their neighbors were, when they arrived in the country. A 1900 census record tells you a specific person named Kowalski lived in Chicago, worked in a steel mill, immigrated in 1888, and had a wife and three children. DNA can't tell you any of that. Records build stories; DNA confirms you're on the right branch.
Start with genealogy records. The research path through documents gives you names, dates, and places — the scaffolding your family tree is built on. Once you have a tree with 2–3 generations documented, a DNA test becomes a verification and extension tool: it can confirm you've identified the right ancestors and occasionally surface new matches you didn't know about. If you do a DNA test first without any genealogical context, you're looking at a map with no street names — interesting, but hard to navigate.
Yes, but with important caveats. If you were adopted or have no existing family records, a DNA test can help you find biological family — biological parents, siblings, half-siblings, and extended matches who may be willing to share what they know. However, the success of this approach depends entirely on whether your biological family has also taken a DNA test. If they haven't, you may find lots of distant matches with minimal usable information. In this scenario, DNA is most useful when combined with the paper trail from your adoptive family records.
DNA testing companies have varying policies on how your genetic data is stored, shared, and used for research. Some share de-identified data with pharmaceutical partners; some have sold access to law enforcement; all retain your biological sample indefinitely unless you specifically request destruction. If privacy is a concern, read the company's current privacy policy carefully before submitting your sample. You can upload raw DNA data to Gedmatch for research purposes, but that platform also cooperates with law enforcement under certain conditions.
See what KinLore found for your family
We search census records and more across 3,233 U.S. counties and build a 3,000+ word narrative from the documents we actually find.
See what KinLore found for your family — kinlore.ai/gift
Related: How to Start a Family Tree: Complete Beginner's Guide · What Is a Family History Narrative (And Why It Beats a DNA Test) · 5 Free Ways to Research Your Family History · How to Research African American Genealogy · How to Find US Census Records