N

Never 86'd · research

Public operator evidence desk

— August 14–20, 2026

Thirteen signals in seven days. Evidence of language—not market size.

A focused public scan found recurring restaurant-side discussion of delivery economics, refunds, dispute controls, payout holds, promotions, and reconciliation. The sample is documented and reproducible enough to support problem discovery. It is not a census, incidence rate, or proof of recoverable dollars.

Qualifying signals

13

Named operator cases

7

Facebook

4

TikTok

3

LinkedIn

6

— Included

  • Publicly visible during the window
  • Restaurant-side third-party delivery economics or control
  • Unique after deduplication
  • Source attribution preserved

— Excluded

  • Consumer-only or driver-only posts
  • Stale material outside the window
  • Generic restaurant-margin content without a 3P link
  • Never86'd's own posts
  • Duplicate reshares
  • Unsupported legal or recovery claims

— Source ledger

Every item keeps its limit attached.

Download public JSON ↗

D1

2026-08-14

Facebook · Direct operator voice

Refund and dispute evidence

Open public source ↗

Evidence limit

Public restaurant-origin post; statement and bank outcome were not independently audited.

D2

2026-08-15

Facebook · Direct operator voice

Refund ownership and direct-order workflow

Open public source ↗

Evidence limit

First-person operator comment; no platform statement or contract was reviewed.

D3

2026-08-14

Facebook · Direct operator voice

Delivery-channel contribution economics

Open public source ↗

Evidence limit

First-person operator account; percentages and restaurant identity were not independently audited.

D4

2026-08-17

Facebook · Direct operator voice

Channel exit and direct-order economics

Open public source ↗

Evidence limit

Named restaurant-origin source page; the attributed cost figure was not independently audited.

C1

2026-08-15

LinkedIn · Carried operator case

Dispute and account-control workflow

Open public source ↗

Evidence limit

Public operator-side case; underlying account history and statement were not supplied.

C2

2026-08-14

TikTok · Carried operator case

Payout hold and verification

Open public source ↗

Evidence limit

Media-carried operator case; no independent conclusion about residual disputed amounts.

C3

2026-08-18

TikTok · Carried operator case

Channel-switch economics

Open public source ↗

Evidence limit

Vendor-carried restaurant case; useful demand signal, not an independent savings study.

M1

2026-08-14

TikTok · Corroborating mechanism

POS-to-statement-to-deposit reconciliation

Open public source ↗

Evidence limit

Restaurant-finance specialist mechanism evidence, not an operator complaint.

M2

2026-08-14

LinkedIn · Corroborating mechanism

Fee structure and margin pressure

Open public source ↗

Evidence limit

Advisory-market signal, not a restaurant statement or audited loss case.

M3

2026-08-16

LinkedIn · Corroborating mechanism

Net-deposit accounting and payout reconciliation

Open public source ↗

Evidence limit

Finance-specialist explanation; not an allegation against a specific marketplace.

M4

2026-08-14

LinkedIn · Corroborating mechanism

Marketplace cost-stack economics

Open public source ↗

Evidence limit

Platform-side perspective supporting the multi-line economics mechanism.

M5

2026-08-14

LinkedIn · Corroborating mechanism

Payout matching and marketplace availability drift

Open public source ↗

Evidence limit

Automation-vendor workflow thesis, not an operator-origin loss case.

M6

2026-08-18

LinkedIn · Corroborating mechanism

Promotion control and payout workflow

Open public source ↗

Evidence limit

Competitive vendor claim included only as market-demand evidence.

— What the study permits

We can say the documented sample contained 13 qualifying signals, including seven direct or carried operator cases, and that the same control path appeared across economics, disputes, promotions, payout holds, and reconciliation. We cannot say only 13 people discussed 3Ps, that all 13 were operator complaints, that any charge was wrong, or that the sample proves prevalence, market size, or paid demand.

Reddit and YouTube produced no qualifying result under the focused in-window method; that is not a claim of platform-wide silence. X received limited coverage and is excluded from the denominator. Private conversations were excluded from the public count.

Public posts are linked, summarized, and treated as attributed claims—not copied as testimonials or presented as audited findings. Read the evidence, privacy, and corrections standard.