Evidence-based default: not yet. One disciplined account at 3-5 posts/week usually beats two half-fed accounts on total growth.
| Finding | Evidence | Confidence |
|---|---|---|
| One account at 3-5 posts/week usually beats two half-fed accounts on total growth | Buffer 2025 (2.1M posts): growth doubles from 0.12% to 0.26% going 1-2 to 3-5/week[1] | Strong |
| At 16K, the account is NOT maxed out | 10K-50K band averaged 17.20% audience growth in 2025[2] | Strong |
| Instagram engagement down ~24% YoY across all brand sizes in 2025 | Socialinsider 2026 (35M posts, 447K pages)[2] | Strong |
| Home Decor saw ~30% engagement decline specifically | Rival IQ 2025 (4M+ posts)[3] | Strong |
| No algorithmic sibling-account penalty exists | Meta transparency docs[4] | Strong |
| Real cannibalization risk is human attention, not algorithm | Inferred from ranking logic: overlapping accounts mean followers pick one[4] | Moderate |
| Split works ONLY when child has different buyer, content engine, or business model | Pattern across R'Kan, PODIUM, BINST, Studio SPONG, Studio Wolf[5][6] | Moderate |
R'Kan Edition won because it had a sharper story: "functional art" vs the parent's broad "design company" positioning.[6] The market found this easier to understand and reshare.[7]
But R'Kan Edition is the outlier. Most small-scale splits stay small:
| Child Account | % of Parent Size | Assessment |
|---|---|---|
| R'Kan Edition | 180% | Winner — sharper story than parent[7] |
| VALRYGG Objects | 24% | Exists but parent is real destination[8] |
| HoBaan (Studio SPONG) | 12% | Conceptually clear, commercially tiny[9] |
| REIKA (VALRYGG) | 5% | Distinct offer but almost no traction[8] |
| SW Bespoke (Studio Wolf) | 2% | Warm-start from small parent does not magically create traction[10] |
Usual failure mode: satellites never escape parent gravity. They read as sub-brands or design offshoots, not independently followable worlds.[6]
Child brands massively outgrow parent: Boca do Lobo (566K, 5.4x parent), LUXXU (487K, 4.6x).[6] Content volume is immense — Boca do Lobo has 21,550 posts, LUXXU 19,100.[6]
Covet is 500+ employees.[11] The content production does not scale down to 2-10 people. Small studios can copy the architecture, not the production intensity.[6]
| Requirement | Detail |
|---|---|
| Publishing floor | ~3 quality feed posts/week + consistent Stories per account[1] |
| Team reality | 2-3 person team: realistic for one main + one child IF shared production backbone[6] |
| Content split | At 10 posts/week total: split 7/3 or 6/4, parent gets more[6] |
| Pre-seed | 9-15 posts before launch[5] |
| Follower transfer | Expect low-single-digit transfer (~5% at best) from parent[5] |
| Decision point | 30/60/90 day sequence with keep/kill decision at day 90[5] |
Rule: if you cannot sustain ~3 quality feed posts/week + Stories on child WITHOUT weakening parent, do not launch.[5]
Narrow accounts built around one sharp aesthetic often grow more efficiently than broader umbrellas. But the best are not "small niches" — they are "compressed worlds": recognizable in one frame, rich enough to keep unfolding, symbolic enough to signal identity.[6]
| Account | Followers | Focus |
|---|---|---|
| deVOL Kitchens | ~880K | Kitchens only[6] |
| House of Hackney | ~426K | Maximalist botanical[6] |
| APPARATUS | ~295K | Sculptural lighting[6] |
| Selkie | ~901K | Fantasy silhouette[6] |
Breadth wins when the broad account is a publisher/discovery engine (AD ~11M, Design Milk ~4M).[6] For product brands, compression beats sprawl.
| If Primitif is... | Then... |
|---|---|
| A compressed world with its own followable obsession | Separate account has a case — but must clear the minimum viable bar above |
| An aesthetic variation within the same design philosophy | Keep as a content pillar on The Medium's account — use Highlights, pinned posts, and format discipline[5] |
Split works when the child gives people a different reason to follow every week — not when it duplicates the parent's content with a different mood.[6] Successful splits have: different buyer or buying occasion, different functional proposition, own repeatable content engine, and a team that can sustain real cadence on both.[5]