Collab Posts

Co-authored posts that distribute across both accounts' follower graphs.

2.7x engagement lift for 10K-50K brands. 3.4x for XS brands.
Adoption was only 4.1% by Q4 2024 — most brands are not doing this yet.[1]
Source: Emplifi 2025, 4,823 brands with collabs out of 83,965 total

Evidence

How to Execute

  1. Pick adjacent-niche partners, not same-niche. Shared buyer, different offer, shared aesthetic. Low audience overlap = more net-new reach.[12]
  2. Prioritize same-size or slightly larger partners. Smaller but high-trust niche partners can outperform larger accounts on ROI.[10]
  3. Aim for 2-4 meaningful collabs per month (~25-33% of feed). Over-collaborating dilutes brand coherence.[1]
  4. Collab Reels for discovery. Styling demos, before/after, product-in-use, BTS, room reveals, material close-ups.[5]
  5. Collab Carousels for saves. "3 ways to style this," material education, room recipes, project credits.[6]
  6. Chain for conversions: Collab Reel/carousel → both reshare in Stories → Stories carry link sticker, DM keyword, or product tag.[13]
  7. Build the flywheel: Test many → score → re-engage winners → accumulate UGC → amplify best content.[14]
  8. Decide who publishes. Only the publisher controls comment deletion and sees profile visit/follower data. Decide before the brief is sent.[27]

Outreach Workflow

Shortlist first, then outreach. Modash quotes an operator with a 90% response rate because she does a real pre-check before messaging — even 5-10 minutes reviewing profile, recent posts, and fit.[25]

Message Pattern

Personalized proof you actually looked + one-sentence brand intro + one concrete opportunity + easy yes/no CTA.[25]

Follow-Up Sequence

DayAction
Day 0First touch with one concrete angle
Day 4Short follow-up referencing a recent post, re-attaching materials
Day 10Deadline nudge: "we're finalizing partners in 5 days"
Day 14Final nudge — ask which way they're leaning, say you'll stop

Most marketers follow up 1-2 times. A Reddit audit of ~50 pitch emails supports this 4-step cadence.[25][28]

What Gets Ignored

Generic flattery, vague asks, hiding details behind links, no stated deliverables/timeline/payment terms. Put campaign name, deliverables, deadlines, usage, exclusivity, and payment in the message itself.[28]

Deal Structure

Four Buckets

TypeWhen to Use
No-strings giftingSend product, see who is genuinely enthusiastic. No obligation.
BarterProduct in exchange for pre-agreed deliverables.
Straight paidFlat fee for defined package. Micro-influencer range: $200-$2,000/post for 10K-50K.[33]
HybridFlat fee + affiliate commission, or product + commission. Most flexible.[30]

Affiliate benchmarks: three-fifths of brands pay 10%-14% per sale.[30]

Deal Memo Template

One-page deal memo covering:[31][32]

FieldDetail
Publisher accountWho posts — only they see profile visits and can manage comments
DeliverablesExact format: Reel, carousel, Story set, or true collab post
Draft due dateFull campaign timeline in writing
RevisionsHow many rounds included
Go-live windowDate range + coordination plan
Usage rightsOrganic only or paid too; which channels; how long. Adds 25-100% to base fee.[31]
Exclusivity24hr blackout around launch + ~30-day category exclusivity[32]
CompensationFlat fee, product value, commission %, bonus triggers
TrackingCode, link, or none
Payment terms50/50 or net-15 preferred. Net-30+ can be a dealbreaker.[31]
Kill feeMissed deadline / termination terms

Visual Example Dissections

National Geographic x Hugh Jackman: Dual-IP Reel

Opens on the animal wolverine, uses Ryan Reynolds narration, then folds in Hugh Jackman's Wolverine character. 233,000+ likes. Hook: word-association bait where "wolverine" means two things.[26]

Structure to steal: (1) open on the unexpected subject → (2) introduce the voice/personality bridge → (3) reveal the second brand's world → (4) end with both brand assets only after the story lands.

Nike Football x Erling Haaland: Athlete + Product + Scarcity

"Manchester on Point" — Haaland "taking over the city" in the new Phantom GX. 290,000+ likes. Product story tied to his record-breaking season, not a generic promo.[26]

Shot logic to copy: (1) hero identity immediately → (2) city/context shots that scale the moment → (3) close product shots with personal details → (4) scarcity/drop cue.

Notion x Mylene Mae: Utility Reel, Not Ad Reel

Shows how her "side quests" become ideas, projects, and lists inside Notion. 489K+ views, 33.2K likes, 220+ comments. Doesn't start by describing Notion — starts with a relatable behavior and uses the product as the organizing mechanism.[34]

Structure to steal: (1) problem or relatable behavior → (2) creator's personal framing → (3) live walkthrough inside their routine → (4) subtle payoff instead of hard CTA.

Key insight from Away's playbook: the highest-performing brand contribution is "product + prop + permission." The creator contribution is "format + voice + audience expectation." Don't over-script — treat creators as creative partners, not script readers.[35]

Expected Outcomes

Benchmarks by Account Size (10K-50K)

MetricReelsCarousels
Avg views~2,460~4,275
Avg comments~12~10
Avg saves~7~8
Engagement rate~0.52%~0.55%

Source: Socialinsider 2026.[6] These are solo-post averages. A collab should beat these.

Planning Model

Practical bar: "Better than your own 90-day median on either reach or saves."[6]

BucketViewsSavesMeaning
Weak collab~2K-5K (baseline)Single-digitNo meaningful lift — review partner fit
Solid collab5K-15K+ (above baseline)Double-digitWorking — re-engage this partner
Breakout collab15K-50K+ (above ceiling)Follower lift felt over following week

These are inferences from public benchmarks, not a collab-specific dataset. Only the publisher sees the actual profile visit and follower data.[27]

Failure Cases

PatternWhat Goes Wrong
Wrong partner selectionChoosing for follower count instead of audience fit. Most common mistake.[27]
Vague outreachPitch reads like blast email, hides details behind links, never states deliverables or payment.[36]
Workflow friction after the yesToo many approvals, vague briefs, changing deliverables, late payment, unclear usage rights.[31]
Over-scriptingTreating creators as script readers instead of creative partners. Best posts fit the creator's native format.[35]
Publishing chaosPost goes live whether or not invite is accepted. If collaborator declines, you can't re-add them. Coordinate before publishing.[26]
Bad incentive designCommission-only scares strong creators (they shoulder all risk). Flat-fee-only misaligns incentives. Hybrids usually work better.[37]

Examples

BrandSizeModelResults
Ondecor[14] ~25K 6-month creator program. Free wallpaper + 10% affiliate. Mostly nano/micro, re-engaged high performers. 270+ collabs, 1.3M+ impressions, $37.6K EMV, 360+ UGC assets, 35 repeat creators
Beau Bottles[17] ~40K Gifting + affiliate + UGC-for-ads. 15K+ follower priority, usage rights up front. 233 creators, $82K+ tracked sales, 34K+ clicks, ~2M impressions in 3 months
Not Just Analytics[27] Repeated collab posting between Upgrade and Will. High audience overlap, consistent execution. 6,800 followers in 2 months with very high engagement rate
@sonne.bali[18] ~4.5K "Layers of Living" with Heveya + Euodia Home via Jia CURATED. Event-based collab pattern with cross-disciplinary partners
@kevalaceramics[18] ~46K Hospitality adjacency: bespoke work for Pan Pacific Hotel. Project-based collabs as portfolio + social proof
Relevance to The Medium

Bali has a dense, cross-disciplinary design ecosystem with more natural high-fit partners than most markets — interior designers, boutique villas, ceramic studios, textile makers, design events like Jia CURATED and Seminyak Design Week.[18][19] Nobody in the Bali competitor set uses systematic co-authored collab posts.[20] Five priority partner buckets: (1) interior designers/architects, (2) villas/hotels/restaurants using your pieces, (3) ceramic/tableware brands, (4) design events/markets/media, (5) builders/material partners.

Open Questions
[2] Instagram Help — How Collab posts work
[3] Instagram — Ranking explained
[4] Hootsuite — Collab Post guide
[5] Buffer 2025 — Content format data
[6] Socialinsider 2026 — 35M posts benchmark study
[7] HypeAuditor 2025 — Engagement rate benchmarks
[8] J. Retailing & Consumer Services 2025 — Influencer-brand fit
[9] J. Academy of Marketing Science 2025 — Meta-analysis: fit effects
[10] Modash 2025 — Influencer trends
[12] Dash Social — Collab vs duplicate content
[13] Luminous Insights — Stories persuasion study
[14] Influencer Hero — Ondecor case study
[18] Bali collab examples — Jia CURATED; Bali Competitor Landscape
[19] Seminyak Design Week — seminyakdesignweek.com
[20] Bali Competitor Landscape — no competitor uses systematic co-authored collabs
[37] Sprout Social — Affiliate influencer marketing