Your Surname Now Has a To-Do List
 |
| WikiTree+ Surname Collaboration Score word cloud* |
For years now, whenever someone has asked me how the Peasley One Name Study is coming along, I've given some version of a shrug. I could tell them which branches I'd worked on lately and which ones were still waiting for me, but I couldn't tell them whether the study as a whole was in better shape than it was last year.
So when Surname Collaboration Scores showed up on WikiTree+, naturally my curiosity was piqued! Finally, a way to answer that question. And the best part is that it isn't just my answer — everyone working the same surname can see it too.
The bigger project
These scores aren't a standalone gadget. They're one piece of a larger 2026 push from the WikiTree Team: improved surname pages, weekly surname email updates, connecting FamilyTreeDNA surname group projects to the tree, and more support for coordinators through the One Name Studies Project. Aleš Trtnik built the first draft of the scores.
They're still in development, so if you have thoughts, the G2G discussion is where the Team wants to hear them.
Five numbers, added up and divided by five
The math is refreshingly plain. No weighting, no black box. Here are the five pieces:
- WikiTree+ Errors — the percentage of profiles with no reported Errors. Warnings and Hints don't count against you.
- Connected — the percentage connected to the one big, free tree.
- Merged — the percentage with no pending merges or unmerged matches.
- Sourced — the percentage of profiles that actually have sources. This one is changing, and I'll come back to it.
- Content Rank — the average Content Rank across open profiles with that Last Name at Birth.
Four of those five are things you can just work down a list and fix, and each one has a report behind it! If your Connected number is low, you go to the unconnected profiles report for your surname and start adding relatives until they link up. There's no mystery about what to do next, which is exactly what I want from a tool.
A few things worth knowing before you go looking:
- Scores are only calculated for surnames with 250 or more open profiles at that LNAB.
- Private profiles aren't included.
- Last Name at Birth only — current and other last names don't count.
- Scores are recalculated regularly. The help page describes a weekly cycle timed to the surname email updates, though the exact cadence is still settling.
The breakdown is where the gold is
The overall percentage is the headline, but click on it, and you get the components, and that's the real gift.
I watch two names. Peasley is my study. Paisley is a variant registered under my study at the Guild of One-Name Studies, though the WikiTree study there isn't mine to lead — I'm just a member.
Paisley shows 99 unconnected profiles and 95 with errors. Peasley shows 14 and 39.
Now, 99 unconnected profiles could feel discouraging, but it isn't! It's a Saturday afternoon with a number attached, and I know exactly where to start. That's so much better than the vague feeling that there's work out there somewhere.
The variant detail that surprised me
Here's the part I didn't expect, and it matters for anyone running a one-name study.
The standard surname report scores each spelling separately. Smyth is not Smith. Peaslee is not Peasley. Pasley is not Paisley. If you think of your study as covering half a dozen spellings — and mine certainly does — the default view chops it into pieces.
But the One Name Studies report combines registered variants into one score!
On the standard report, Paisley by itself is 869 profiles. On the ONS report, Paisley with Pasley and Pacley is 1,262 profiles at 88.48%. Peasley alone is 584; Peasley with Peisley and Peaslee is 1,333 at 94.59%.
So there's the argument for registering your study, right there in the numbers. Register with the One Name Studies Project, and your variants get counted together, the way you already think about them. Don't, and your surname gets split into pieces that might not even reach the 250-profile threshold on their own.
One thing to be deliberate about: different registries draw the variant boundaries differently, so the WikiTree grouping and my Guild registration don't line up exactly.
Bio Check is becoming the sourcing test
Now back to that fourth component.
Look at the WikiTree+ report, and you'll see a column labelled bioCheck Sourced sitting next to the plain Sourced one. That isn't a stray extra — it's where that component is heading, and I think it's a real improvement.
The old measure counts profiles that don't carry the {{Unsourced}} research note box, which really means profiles nobody has flagged yet.
Bio Check, an app by Kay (Sands) Knight, actually reads the biography. It sorts profiles into Sourced (there's something there that tells you where the information came from), Marked (it carries the Unsourced box), or ? / Maybe — nothing in the biography identifies where any of it came from, and nobody has flagged it.
That last group is the one I care about. Those are the profiles hiding in plain sight, and I know I have some.
The gap between the two columns is exactly that population. Paisley: 97.01% by the old measure, 91.14% by Bio Check. Peasley: 99.49% against 98.97%, so barely a gap there. Smith: 97.79% against 89.22%, which across 306,000 profiles works out to something like 26,000 people whose profiles look fine and aren't.
Your score may well drop when this lands, and that's alright! It's the number getting honest, and it rewards real sourcing rather than tidy templates. For a one-name study, it's better information, because it tells you where to look instead of where somebody already looked.
Two practical things from the Bio Check help that surprised me:
Where it looks. Bio Check reads sources inside <ref></ref> tags in the Biography section and anything after the == Sources == heading. The Research Notes section is ignored completely, and so is any line beginning with See Also:. So if your only citation for someone is sitting under See Also, Bio Check won't count it. Worth checking your own habits there — I had to check mine!
Inline citations are worth double. Bio Check also produces its own Bio Score, where a valid source inside a <ref> earns +2 and one listed under == Sources == earns +1.
Coming soon: a weekly email for your surname
Honestly, the piece I'm most looking forward to isn't a score at all.
WikiTree is building weekly email updates, one for each surname. Follow a surname and you'll get:
- Activity in the G2G forum for that surname
- Activity on profiles with that surname over the past seven days
- >New DNA test-takers with the surname, and their Y-DNA haplogroups
- That week's Surname Collaboration Score
- Contact information for the One Name Study coordinator, if there is one
- Links to the ONS home page, the DNA collaboration page, the main G2G post, and the surname's FamilyTreeDNA group project
- Who's newly following the surname, and the top contributors
Think about what that replaces! Right now, keeping up with a surname means checking the forum, the profiles, the DNA project and the coordinator's inbox, and remembering to do all of it. This one just arrives.
For a one-name study that's a real shift — the difference between a study you have to go check on and one that reports back to you.
Three scores, three jobs
These are easy to mix up, so let me lay them out plainly:
- Surname Collaboration Score is surname-level. How is this name doing as a collective project?
- Content Rank is profile-level, and it's about findability. Will a searching cousin land here?
- BioCheck Score is profile-level, and it's about sourcing and style. Is this biography built properly?
Content Rank deserves a note of its own. The Team said they considered leaving it out because, unlike the other four, it isn't a simple percentage you can work down a list. It has dozens of components and folds the other four in as inputs. WikiTree is clear that it is not a measure of quality.
But look at what it's doing to the scores. Across every surname I've checked, the four straightforward components sit up in the high 80s and 90s while Content Rank drags along behind: Smith around 60%, Paisley around 66%, Peasley around 86%. That's exactly why it belongs. Without it, the score would have a ceiling you could reach and then stop.
And here's the good news: almost any improvement to a profile improves its Content Rank. You don't have to reverse-engineer anything. You just need to improve the profile.
Joseph Peasley is my working example. His Content Rank breakdown covers biography and sources, names, vital events, relatives, images, DNA test connections, categories, profile boxes, and external links. Notice that categories and DNA connections are in there — the one-name study work feeds this directly!
Four things you can do this week
- Look up every spelling you work on the surname report, then check the One Name Studies report to see whether your variants are being counted together.
- Click the score and read the components. Pick the lowest one that isn't Content Rank — that's your highest-leverage work, and it comes with a ready-made list.
- Run Bio Check against your surname and sort by the Sourced column. The profiles marked
?are the ones nobody knows about yet. It's also built into the WikiTree Browser Extension if you'd rather have it while you edit. - Check whether your surname already has a study. The index of one-name studies is alphabetical, and there's a separate list of studies needing a coordinator. If yours isn't there, email the Project Leaders and they'll do the setup work for you. Then answer the G2G welcome post to join and get your badge, and add ONE_NAME_STUDIES to your followed tags.
Come talk surnames with me
If any of this is your kind of problem, I'm presenting on it this weekend.
- Host: The Surname Society
- Date: Saturday, 15 August 2026
- Time: 8am PST/4pm BST
I'll walk a real one-name study from the surname genealogy page all the way down to a single ancestor, go deeper on these scores than a blog post allows, and show how name studies, place studies, and DNA group projects feed each other. There will be time for questions, and I'd much rather spend it on your surnames than on mine!
Are you running a one-name study, or thinking about starting one? Come and tell me which surname — I'd love to hear what you're working on.
Hope to see you around the branches!
*Word cloud image created by Azure Rae Robinson from WikiTree+ Surname Collaboration Score data, August 2026, using Claude (Anthropic). Text size = profile count; text weight = collaboration score.
Very helpful explanations, thank you! Will think about these possible new features.
ReplyDelete