A private read of your DNA matches
An unexpected close match, a parent who may not be the one you grew up with, a grandparent nobody can place. Send us your closest matches. We’ll tell you what each one can and can’t be, what it most likely means, and the next step most likely to settle it. Privately, and in plain English.
Take your time. You don’t have to decide anything today.
Send us your matches ↓ · nothing to pay until we’ve looked
Example read · an invented case
D.R. is very likely your half-sibling: it fits the number, the ages and the shared matches. The open question is which parent you share.
You mentioned none of your dad’s cousins appear in your list. That leans toward a different biological father, but only weakly if few of them have tested.
If a cousin of the father you grew up with tests and matches you at a cousin’s level, that strongly points to him being your biological father; if that cousin also matches D.R., D.R. is very likely your half-sibling through him. If the cousin matches neither of you, the answer lies in your biological father’s family, which D.R. shares. If your dad is alive and willing, his own test settles it directly; whether to ask is entirely your choice. Free first: upload your raw data to GEDmatch, where D.R. may appear under a full name.
Some of this can be hard to read. We write it plainly and gently, and what you do with it is yours to choose.
What you get
$49 · only after we’ve looked and told you it will help.
How we work
$49 for the full read. But you don't pay anything up front: send us your matches, and within about a day we'll write back with what they can and can't show and whether a full read would help. If it would, we send a payment link. If it wouldn't, we'll say so, and that first note is free.
Claude, an AI made by Anthropic, works through your matches and the records; Patrick White runs the desk. Our notes are signed by both of us. The odds come from the Shared cM Project, the standard reference built from tens of thousands of known relationships.
No. Never. We read what you send us and the public historical records, and we write to you. If you want to reach out to a match, we can draft a first message for you to send yourself, in your own name, when and if you choose.
No, and please don't. We only need what your testing site shows you: each match's shared cM (and segments if it's shown), and which of your other matches they share. Initials instead of names are fine.
It's used for your read and nothing else. We don't sell or share it. When your order closes we delete the match names and keep only what we need for a receipt. Details in our privacy note.
Then you get your $49 back. Just reply and say so.
A private page you can read on your phone, with a PDF copy: each match, the family lines, the explanations and the next step, usually a few pages.
DNA gives odds, not certainties, and we'll always say which is which. Often the matches narrow it to one family and one generation, and a single well-chosen test settles it. Sometimes they can't yet, and we'll tell you that plainly, along with what would.
Anything else: desk@familyhistoryresearch.org.
Relationship ranges and odds: the Shared cM Project v4 by Blaine T. Bettinger (CC BY 4.0), with probability groups from Jonny Perl’s interactive version at DNA Painter.