Quick take
If your reorg is a PowerPoint exercise to make the reduced headcount look intentional, stop. Fix ownership gaps, not org charts. The people who survived the layoffs already know what’s broken – ask them.
I’ve been pulled into three different “help us restructure after layoffs” conversations this year. The pattern is remarkably consistent. Company does layoffs in Q4 2022 or Q1 2023. The remaining engineers are spread thin. Ownership is unclear. Two or three critical systems now belong to nobody. Leadership decides a reorg will fix it.
It usually doesn’t. Not because reorgs are bad, but because they’re solving the wrong problem.
What actually happened
Here’s the real situation at most of these companies. They cut 20-30% of engineering. The layoffs weren’t surgical – they hit across teams, sometimes based on recency bias or cost rather than capability. The survivors are demoralized, overloaded, and wondering if they’re next.
Then someone proposes a reorg. New team names, new reporting lines, maybe a shift from project teams to product teams. The slide deck looks clean. The reality is chaos.
What these orgs actually need isn’t new boxes on a chart. It’s three things: clear ownership of the systems that still matter, explicit decisions about what to stop doing, and honest communication that doesn’t insult people’s intelligence.
When a reorg is actually justified
Sometimes the layoffs genuinely changed the shape of the work enough that the old structure doesn’t make sense. Real signals:
- Two teams were merged but nobody decided which team’s approach wins, so both codebases rot.
- A critical system lost its entire owning team and it’s being maintained by whoever has time, which is nobody.
- The product strategy changed alongside the layoffs, and the old team boundaries don’t align to the new priorities.
- Dependencies between teams have become the primary blocker because the connecting-tissue people were let go.
These are structural problems that structural changes can help with.
When it’s not
And sometimes the reorg is cover. I’ve seen all of these:
- A new VP wants to “put their stamp” on things. The org was fine. The VP needed to justify their hire.
- Conflict between two directors is being “solved” by separating their teams instead of addressing the conflict.
- Leadership wants to signal activity to the board. “We restructured” sounds better than “we’re still figuring it out.”
- Someone read the Team Topologies book on a flight and now every conversation includes the phrase “stream-aligned.”
If the root problem is people, process, or strategy, moving boxes around won’t help. It’ll just add a transition tax to an already exhausted org.
What I actually recommend
Based on what’s worked across these situations, the playbook is unsexy. Which is why it works.
Step 1: Map what’s actually broken. Not theoretically broken. Actually broken. Talk to the ICs. They know which systems are unowned, which dependencies are blocked, and which meetings are pointless. I usually spend the first week just listening. The engineers are relieved someone is asking.
Step 2: Make ownership explicit. Write down, for every critical system: who owns it, who’s on-call for it, and what the plan is. “Shared ownership” is a euphemism for “no ownership.” If a system matters, one team owns it. Full stop.
Step 3: Kill things. This is the hard one. After layoffs, you have fewer people. You can’t do the same amount of work. Something has to stop. The reorg should make it clear what’s being stopped, not just what’s being reorganized. At one company, the single most impactful change was killing two internal tools that each had one user. That freed up enough capacity to actually staff the important systems.
Step 4: Communicate like adults. The worst thing I’ve seen is the “exciting restructuring” email that uses words like “streamlined” and “focused” to describe what everyone knows was a layoff aftermath. Don’t do that. Be direct: “We lost good people. Here’s how we’re adjusting. Here’s what we’re stopping. Here’s what we’re doubling down on. Questions welcome.”
At Dropbyke, we went through a version of this during a pivot. Smaller scale, but the same dynamics. The team that handled it best was the one where the lead said “I don’t have all the answers yet, but here’s what I know and here’s when I’ll know more.” Trust went up. Attrition went down.
The transition matters as much as the design
Even when the new structure is right, the transition can kill it. A few things that consistently matter:
- In-flight work needs a home. Every project that’s mid-stream needs an explicit owner in the new structure. Not “we’ll figure it out” – a name, by the end of the week.
- New teams need a charter. Not a manifesto. A one-pager: what you own, what you don’t, what your first quarter looks like. Teams that don’t get this spend months figuring out their boundaries through conflict.
- Give it six months. The urge to tweak the new structure after four weeks is strong. Resist it. Reorgs take a full quarter to settle and another quarter to show results. Constant adjustment destroys any chance of stability.
How to know if it worked
Short-term (first month): are the ownership questions answered? Do people know who to go to? Has the “who owns this?” Slack question frequency dropped?
Medium-term (one quarter): is delivery predictability improving? Are cross-team dependencies less painful? Are the right things getting built?
Long-term (two quarters): retention. If you’re losing the people you wanted to keep, the reorg failed regardless of what the delivery metrics say. The engineers who stayed through layoffs and a reorg have a very low tolerance for more disruption. Respect that.
The real talk
Reorgs after layoffs are emotional events disguised as structural ones. The people in those boxes are tired, uncertain, and watching closely to see if leadership actually has a plan or is just rearranging furniture.
The best restructuring I’ve seen this year was at a company where the CTO stood up and said: “This is hard. We’re going to get some of it wrong. But here’s the logic, and here’s how we’ll adjust.” No spin. No excitement. Just honesty.
That org stabilized in eight weeks. The one that did the polished all-hands with the aspirational vision? They’re still losing people.