Lesson 6: Reading Clusters & Breaking Brick Walls
About 20 minutes · What you'll have at the end: at least one cluster labeled with its family branch, and a method you can repeat on every cluster you have.
Yesterday you made clusters. Today you'll learn what genealogists actually do with them. The core insight:
A cluster is a question with a findable answer. Every cluster says: “these people all descend from one ancestral couple — which one?” Answer it once, and you've explained every match in the group.
The workflow: anchor → label → hypothesize
- Anchor. Look for any match in the cluster you can already place — a known cousin, or someone whose tree obviously overlaps yours. That person “anchors” the cluster to a branch of your family.
- Label. Whatever couple the anchor descends from, the cluster is (as a strong working hypothesis) that couple's descendants. Label it: “Miller/Schmidt side,” “maternal grandfather's line.” Do this for every cluster with an anchor and you've mapped most of your match list to your tree in an afternoon.
- Hypothesize. Now the interesting clusters: the ones with no anchor. Those are DNA from a branch you can't yet name — which usually means a brick wall, an unknown ancestor, or (for adoptees and unknown-parentage searches) exactly the family you're looking for. Open the trees of that cluster's members and look for the surnames and places that repeat — the recurring couple in their trees is your candidate ancestor. That's the brick-wall method, and it has solved cases that decades of paper research couldn't.
Practical moves in the Client
- Use Matches to open a cluster member and review their shared matches and tree details side by side.
- Since you gathered trees in Lesson 4, they're searchable locally — scan cluster members' trees for repeated surnames without opening a website. (The People section searches across all of them; the Gold tools in Lesson 8 automate the comparison.)
- Keep notes as you go — even a simple spreadsheet of “cluster → hypothesis → evidence” pays off fast. The Tips & Tactics page has more workflow guidance, and the FAQ answers the common “why is this match in two clusters?”-type questions.
✅ Do this now
- Pick one cluster with a known cousin and label it with the ancestral couple it points to
- Pick one cluster full of strangers, open 3–4 members' trees, and list surnames that appear more than once
- Write down one hypothesis (“this cluster is probably the ____ line”) — you'll test it as more data arrives
You now have the complete core loop: gather → cluster → anchor → hypothesize. Everything from here is more data and better tools — which is exactly what Lessons 7 and 8 are about.