The split/ergo cohort grew up
Split and ergonomic boards spent a decade as a hobbyist sub-cohort orbiting a stagger-majority. By 2026 the Voyager line, the Glove80, and a steady trickle of r/mk transition posts have pushed split/ergo across a visibility threshold — still a minority, but now a reference category instead of a footnote.
Split and ergonomic boards have been around long enough that the elder statesmen — the Atreus, the ErgoDox EZ, the Kinesis Advantage lineage — predate most of the hobby's current vocabulary. For most of that history the cohort sat off to the side: present, defended, but small enough that a stagger-board reader could spend years in keyboard discourse without engaging with it. That side position is what's changed. The cohort isn't bigger than stagger in 2026, and the Trends Tracker doesn't claim it is. What it has become is a category that turns up unprompted in transition posts, configurator pages, and developer newsletters — a default option to consider, not a curiosity to explain.
A decade as a sub-cohort
The lineage is real and worth respecting. The original Atreus shipped in 2014 as a 40-key split-adjacent ortho with a small, devoted following. The ErgoDox EZ took the open-source ErgoDox PCB and turned it into a pre-built that someone could buy without a soldering iron, which is what most of the early growth needed. The Kinesis Advantage stayed in its own corner of the typist-ergonomic world — concave key wells, contoured thumb clusters, a fixed geometry that worked for the people it worked for. None of them were mass-market. All of them sustained.
What they didn't do was cross into the stagger-board reader's default consideration set. A reader picking up a custom in 2018 weighed a 60% against a 65% against a TKL. Splits were a separate hobby they might investigate later. That's the framing that has shifted.
The Voyager line did the work
The clearest single product story is ZSA's Voyager. A low-profile split with choc-spaced switches, hot-swap sockets, and a thin aluminum body, the Voyager has been ZSA's flagship since 2022 and has shipped enough units to anchor a real product line — replacement plates, alternate switches, a configurator that gets updates rather than retirement notices. ZSA's Voyager product page reads like a settled SKU now, not a launch.
What the Voyager did for the cohort is give it a recognizable silhouette at a coherent price band. A reader who'd seen Moonlander photos but bounced off the size now had a low-profile alternative from the same maker, and the Moonlander itself stayed available as the chunkier sibling. That's the supply pattern that turns a single-board curiosity into a product family — and a product family is what gets cited in roundups.
Glove80 in dev newsletters
The other product that keeps showing up is MoErgo's Glove80. Column-staggered, sculpted palm rest, concave key wells that owe a clear debt to the Kinesis Advantage — but priced and marketed as a contemporary product rather than a clinical-ergonomic one. The Glove80 turns up in developer newsletters and r/mk transition posts at a rate that the Advantage 2 never did, partly because MoErgo leaned hard into the wireless story and partly because the form factor photographs well.
The Glove80 isn't outselling stagger boards. It's showing up in the discourse in a way that says the cohort has its own anchor at the column-staggered, palm-rested end of the spectrum — and between Voyager at the low-profile flat end and Glove80 at the sculpted-well end, the category has the two reference points it needs.
Why now and not five years ago
The same reader could have bought a Kinesis Advantage 2 in 2019 and gotten most of the ergonomic payoff. Why is 2026 the year the cohort crossed?
Three forces line up. The first is firmware: ZMK's wireless story caught up to QMK's feature set, which made wireless splits stop feeling like a science project. A reader weighing QK75 for a wireless mid-tier 75 in 2026 can also weigh a wireless split at a comparable price band without the battery-life caveat that defined splits in 2022. The second is keycaps: choc-spaced sets and sculpted MT3 alternatives both shipped enough volume to make a split build look like a finished build instead of a parts pile. The third is supply: pre-built, hot-swap, in-stock split SKUs are now a thing, which is the prerequisite for any cohort to grow past the people willing to solder.
None of those forces existed at saturation in 2021. All three are sitting at usable in 2026.
What we're watching
Split/ergo isn't replacing stagger. The Bakeneko65 still owns the gateway tier for a first custom; the live Trends Tracker for 2026-W19 has the split/ergo row at score +32 and direction up, which is meaningful but doesn't dethrone the stagger majority. What 2026 will tell us is whether the cohort stops at "reference category" or grows into a second axis of default — the way 75% has converged on the layout axis, splits may be converging on the ergonomic axis, and the two convergences aren't competing for the same readers.
Two specific watches. First, whether a sub-$300 pre-built split from a well-distributed maker lands in the next two quarters; the ladder still has a missing rung between the entry kits and the Voyager band. Second, whether ZMK Studio's configurator UX gets close enough to QMK/VIA's to remove the last firmware-friction excuse. If both happen, the cohort doesn't just stay visible — it becomes a serious option for a reader whose first custom isn't their last.
Build sheet
Keep reading
- Trends
When customs became vendor-first
From 2018 through 2022 the typical custom shipped through one host vendor and a fan of regional proxies; by 2026 the baseline custom is a configurator on a vendor's own storefront. Boards moved. Keycap group buys mostly didn't.
thock5 min read - Trends
How 'thock' became an acoustic spec, not an accident
A decade ago a thocky board was the win condition of a $300 custom — the payoff for the right switches, the right lube job, the right patience. In 2026 the same sound is table-stakes for a $150 prebuilt: factory-lubed switches, silicone dampening, gasket-mounted plates, PBT doubleshot. The stack that used to be enthusiast-only is now what mid-tier vendors design toward from day one.
thock7 min read - Trends
How 75 percent became the default custom layout
Two years ago a custom-board recommendation defaulted to 65% with a polite note about 75% as the next step up. The Sonnet 2026 refresh, the QK75, and a crowded mid-tier have flipped that order. 75% is now the layout customs reach for first.
thock5 min read - Trends
How Hall-effect quietly became the new mid-premium default
The polling-rate hype is loud and easy to wave off — 8K, 32K, numbers chasing each other up the spec sheet. The quieter story is that the magnetic-switch tier has stopped being a Wooting-shaped niche and has settled into the $100–$230 prebuilt slot the scene actually buys from. The volume has moved, not just the marketing.
thock7 min read