Nobody Owns the Seams
Managing at the seams
The HRBP Lab · Issue 005 · May 30, 2026
Hard to believe it is almost June. Lots of HR and AI in the news but some topics, like this one, are evergreen so here are my thoughts this week from the HRBP Lab. Thanks for following along!
A few weeks ago, a peer in our weekly sync started venting. She had spent two days chasing an ER case that should have closed in October and somehow hadn't. The system said closed. The tracker said open. The team thought they'd handed it off three months ago.
She wasn't asking for help fixing the case. She was asking why this kept happening.
I caught myself saying the same sentence I've said to almost every HRBP I've ever worked with: managing at the seams, where global meets regional partners — I've seen this in every organization that has a mixed global and functional structure. The phrase came out before I'd thought about it. That happens when something has been true for fifteen years.
That same day, on a totally separate three-way call, the same pattern surfaced. Different case, different team, different geography. Same shape. An action that should have happened between two scopes never had an owner. Two failure modes in one day, same seam.
This issue is what I've been trying to say since the same pattern surfaced three weeks running. Every HR mess I've worked has had the same fingerprint, and it isn't an ER mess. It's a seam mess. The HRBP move isn't more trackers. It's naming the seam and forcing somebody to own it.
The seam is a category, not a team problem
Most of what gets diagnosed as an ER failure, a comp failure, or a recruiting failure is something else underneath. The case didn't fall apart because ER dropped it. It fell apart because the work that should have happened between two scopes (ER and HRBP, COE and HRBP, peer HRBPs across a regional cut) never had an owner. Nobody dropped it. There was nothing to drop.
Once you can see the shape, you can't unsee it. The same fingerprint shows up under almost every "why didn't this close?" — and almost none of those messes are best fixed by adding tracking. They're fixed by naming what lives in the gap.
This isn't an HR pathology. Last week Linus Torvalds wrote that the Linux kernel security mailing list has become "almost entirely unmanageable" — not because real bugs are up, but because AI-generated bug reports are, and maintainers are eating the gap. The list was designed assuming reporters would do real work before filing. AI flipped that assumption, and nobody named the seam before the volume showed up.
Same fingerprint. The work between two scopes had an owner by convention, not by design. When the volume changed, the convention broke.
HRBPs sit at the seam by structural position
The HRBP role is the only one in the system that already sits at the intersection of leader strategy, COE function, and regional reality. We see the case the leader sees, the policy the COE is enforcing, and the people on the ground all in the same week. Nobody else in the org has all three vantage points at once.
That means HRBPs are the natural people to see seams. We're also, structurally, the people most tempted to absorb them. The work that lives between two scopes ends up on the desk of whoever notices it first. Usually the HRBP. And the cost is high. Within a year, every HRBP I've seen do this becomes the de-facto seam-tax collector for the whole org, drowning in coordination work that was never theirs.
The HRBP job is to name seams. Not to fill them. Those are different moves.
Name the seams, don't fill them
The trap is thinking the answer to "this work fell through the cracks" is "I'll take it." It isn't. The answer is to name the work that lives in the gap, point at it out loud, and force a three-option decision: you, them, or built into the process. Three options, one has to be picked.
Here's an abstracted version. A case I worked recently: an ER incident from October that didn't get followed up properly. The case closed quietly in the system. By May, the person at the center of it had moved through a different process for a different reason, and ended up on a performance plan instead of being exited the way October — properly followed through — would have led to. The misdesignation cost time, money, and a complicated negotiated exit. None of it needed to happen if the work between the October close and the May designation had had an owner. Name that gap before the next October incident, and the work either gets owned, gets handed off explicitly, or gets built into the process.
Five questions for a 30-minute session
Take these into a 30-minute 1:1 with a leader you support, or with the peer HRBP whose scope sits next to yours. Same structural pattern as Issues 002, 003, 004 — five questions, one session:
Where does your scope end and your ER partner's scope begin?
Where does your scope end and your COE partner's — comp, talent, L&D — scope begin
Where does your scope end and your peer HRBP's, in the next geo or function, begin?
Of those three seams, which one has produced a wrong designation, a stalled action, or a late escalation in the last 90 days?
Who owns the fix — you, them, or the process?
The session ends with one named seam, one named recent failure, and one named owner. That's the smallest version of the move. It does not require a re-org. It requires thirty minutes and a willingness to say the gap out loud.
Sidebar — when something just broke
The body framework is for planning. This one is for triage. When something has just gone wrong that shouldn't have, ask three questions:
What just happened that shouldn't have?
Whose work was supposed to prevent it?
Whose work was actually supposed to prevent it?
The gap between Q2 and Q3 is the seam. Name it before the next ten things break the same way.
Before you do anything
Pick one seam this week — between you and the scope right next to yours. Don't fix it yet. Name it out loud in your next 1:1 with the person on the other side, and ask whose work it is.
That's the whole move. The seam isn't yours to fill. It's yours to name.
-Josh
The prompt I'm using this week
This pairs with the five questions above. Use it before or during a 1:1 with a peer HRBP or a leader you support.
"I'm helping [PEER NAME / LEADER NAME] map the seams between our scopes — the work that lives between us and isn't formally owned by either side. We want to leave the conversation with one named seam, one named recent failure that came out of it, and one named owner. Help me run a short discovery on the five questions below. For each one, suggest a follow-up I can ask if their answer is short or vague. After the conversation I'll come back with their answers and we'll draft the explicit owner decision together.
Where does your scope end and your ER partner's scope begin?
Where does your scope end and your COE partner's — comp, talent, L&D — scope begin?
Where does your scope end and your peer HRBP's, in the next geo or function, begin?
Of those three seams, which one has produced a wrong designation, a stalled action, or a late escalation in the last 90 days?
Who owns the fix — you, them, or the process?"
Why it works: Most of the time, the people on either side of a seam have never said out loud where their work ends. The conversation is uncomfortable, which is the point. The prompt makes Claude do the structural work — pulling answers deeper, watching for hand-waving — so you can stay focused on the person across from you. The "we'll draft the owner decision together" framing turns the 1:1 into the front end of a real artifact, not the front end of a strategy doc.
If you try it, tag it #buildtogether and show me what you built. I'll feature the best ones in a future issue.
A footnote for the analytical types: if you want a more forensic version, swap the five questions for a "seam audit" prompt — paste in your scope description plus one recent miss, and ask Claude to surface the likely seam location based on the failure pattern. Less conversational, more diagnostic. I'll write that one up properly in a future issue if there's appetite.
From the Archives:
The Extra Time Problem — Issue 001, 2026-04-25 (the deliberate-subtract setup)
Feeling like it's Groundhog Day? AI can help. — Issue 002, 2026-05-09 (the deliberate-codify middle move)
What only a human can do — even when AI could — Issue 003, 2026-05-16 (the deliberate-protect move)
Where do tomorrow's HRBPs come from? — Issue 004, 2026-05-23 (the deliberate-invest close)
Torvalds on the Linux security list — The Register, May 18, 2026 (external cross-signal — same seam, different field)
If someone you know is navigating the same question, forward this their way. New here? Subscribe at thehrbplab.ai to get the next one. Want to catch up on past issues? Browse the archive.
Subscribe now