# Commercial proposition and research

Decision date: September 23, 2026. All demand, pricing and conversion assumptions below are unvalidated. No interviews, private traffic analysis, signed pilots or paying customers support this launch.

## Overview

Kingy Production Benchmarks helps an ecommerce agency production lead choose the workflow for a recurring client job using accepted final outputs, all failed work, generation consumption and active editing time. Initial job: a 15-second, sound-off vertical bottle ad from an approved packshot and approved copy. The buying decision is whether to retain conventional editing, add a generative workflow, or replace a current tool. The target agency hypothesis is a 3–20 person creative/performance shop making repeated product variants, with a production lead accountable for margin and brand approval. These firm-size and volume assumptions describe whom to interview; they are not measured market segmentation.

## Niche choice

Keep product-ad production, narrow to supplied-asset bottle ads. IAB's July 2025 survey reported 86% of buyers used or planned to use GenAI video creative. That supports category interest, not willingness to buy independent testing. Canva already offers conventional product-video assembly, a practical incumbent. Runway explicitly supports image-to-video production and publishes unit pricing. Kingy's existing bottle asset and exploratory trial records reduce setup cost.

Alternative niches considered:

| Job | Why attractive | Why not first |
|---|---|---|
| Product packshot to 15-second ad | Recurring deliverable; exact identity/copy gates; observable editing cost; current Kingy asset and test history | Acceptance is subjective; one SKU cannot establish broad reliability; ad conversion needs a separate live experiment |
| Long video to captioned short | Frequent, easier objective checks and low generation spend | Mature editors already bundle it; differentiation and paid demand still unproven; Kingy lacks a retained comparable cohort |
| Support-answer automation | Clear cost and correctness questions | Requires permissioned customer knowledge and representative support traffic before useful testing; data access and error severity slow a first pilot |
| General coding-agent comparison | Existing Kingy evidence and clear interest | Strong established evaluation competition and less direct reuse of Kingy's video audience; broad tasks dilute a focused commercial offer |

This is an execution-fit judgment, not proof that ads are the largest or strongest market. Change niche if qualified agencies consistently cannot name a costly workflow decision, cannot share assets, or will not pay for a scoped result. Do not respond by widening the product into a generic leaderboard.

## Competitors and buyer alternatives

| Alternative | Verified public offer | Likely buyer appeal (inference) | Kingy opportunity and limit |
|---|---|---|---|
| Artificial Analysis Video Arena | Pairwise video preference voting across video modes | Quick visual model comparison | Add completed-job cost including rejected work and editing; do not claim their preference data are useless |
| Creatify VideoAdAgent Bench | Finished-ad evaluation with agent/model comparisons, structured compliance and costs | Relevant full-ad comparison from an ad-production vendor | Independent preregistration, retained raw attempts and human editing records; Kingy is not the first finished-ad benchmark |
| Creatify product | Plans for marketers/agencies, templates, actor library and models | Buy one integrated workflow instead of researching | A serious future workflow arm when client demand and exact plan access warrant it; absent from first measured shortlist |
| Canva/product editor | Product-video assembly from uploaded assets, text, animation and audio | Familiar, controllable conventional production | Include this type of practical baseline; a good generator clip may still lose on total job cost |
| Agency's own editor or freelancer | Their actual incumbent is learned in intake, not assumed | Uses real assets and knows brand acceptance | Pilot must beat the value of running their own small test; provide transferable logs and stop rules |

Avoid unsupported claims about competitor win rates, independence, commercial rights or human editing cost. Vendor leaderboard claims are not imported as Kingy observations. Kingy's earlier exploratory trials are not equivalent to this preregistered cohort.

## Free and paid offer

Free public offer: the full protocol, current evidence status, three briefs and reusable asset, a labeled template demonstration, historical feasibility notes, cost-scenario calculator, machine-readable results, failure reasons, reproducibility bundle and dated corrections. A completed public comparison will be published only after real trials and human review. Public conclusions are never paywalled behind an email form.

Paid agency pilot hypothesis, revised at Curtis's request: **US$500 maximum total**, one agency, one product and one agreed 15-second ad brief, three fixed workflows including its practical incumbent, three production sessions per workflow (nine slots). Deliver all retained attempts subject to provider/client rights, accepted exports where any pass, a private quality/cost scorecard, retry and edit-time ledger, a workflow recommendation at the agency's labor rate, reusable prompts/edit recipe, and a 30-minute readout. One written factual-correction round. Target delivery five business days after assets, access, reviewer availability and acceptance gates are agreed. Keep the same acceptance gates, retry limits, 30-minute active-work limit per session and two independent human reviewers. One brief gives a narrow answer for that job and does not establish general workflow reliability.

The $500 ceiling includes up to US$25 generation consumption and any applicable customer taxes or fees; no add-on charges apply to the agreed pilot. No guarantee of nine usable ads, sales lift or production availability. No agency is charged or booked by the inquiry form. Written scope, rights, review staffing and commercial terms come before accepting payment. The separate free public cohort retains its three briefs and 27 sessions. This customer price is not authorization to spend on generation or hire reviewers.

Planning economics, not observed margin: at most 4.5 operator hours for nine 30-minute sessions, plus a planning allowance of 2.5 hours for intake, independent review, report and readout. Seven total hours × $50 = $350; generation allowance $25; total $375, leaving $125 at a $500 customer total before taxes, acquisition, payment fees and overhead. At ten total hours × $50 + $25, cost is $525 before those items. Track actual hours and keep the customer total at or below $500. If the scope cannot be delivered within that ceiling, narrow it transparently before booking or decline the engagement. Never label this a proven profitable service.

## Guidance and demand test

Value hypothesis: a production lead will pay to avoid tool commitments and unbillable retries, and to hand their team an accepted repeatable recipe. The private pilot is worth paying for only if the agency has a consequential upcoming workflow decision and transferable learnings exceed the fee.

Qualify using current monthly deliverables, acceptance owner, present retry/edit burden, next decision date, a representative asset pack, and the dollar value of the decision. Ask for the incumbent workflow and a real rejected example. Do not use hypothetical purchase intent as a sale.

Before broader promotion, seek five qualified agency conversations through individually approved outreach. Internal decision rules, not industry benchmarks: at least three identify a concrete costly decision; at least two request written scope; at least one pays the proposed fee or a clearly scoped paid diagnostic. Ten qualified conversations and no paid commitment justify revisiting the offer/niche. Visits, form tests, downloads and compliments are not commercial validation. Curtis owns outreach approval; drafts remain unsent.

## Sources

All checked September 23, 2026. Public primary sources and local Kingy evidence were used. CRM, meeting transcripts and private customer evidence were not available for this proposition and were not needed to launch an explicitly unvalidated pilot offer.

- https://www.iab.com/news/nearly-90-of-advertisers-will-use-gen-ai-to-build-video-ads/ — category adoption signal, July 15, 2025; intention is not paid demand.
- https://artificialanalysis.ai/video/arena — video preference comparison interface.
- https://d35ghwdno3nak3.cloudfront.net/videoadagent_bench/v1/leaderboard.html — vendor finished-ad benchmark methodology and scope; not independent Kingy performance evidence.
- https://creatify.ai/pricing — competing integrated product and agency offer; dynamic plan prices are not used in Kingy's cost calculations.
- https://www.canva.com/create/product-videos/ — conventional product-video production alternative.
- https://docs.dev.runwayml.com/guides/pricing/ — $0.01/API credit; Gen-4.5 12 credits/sec; Veo 3.1 Fast without audio 10 credits/sec at check date. These are estimates until reconciled to usage receipts.
- https://docs.dev.runwayml.com/api/ — exact endpoint, model IDs, native duration, portrait settings and audio control.
- https://ai.google.dev/gemini-api/docs/video — Google recommends Gemini Omni Flash as default; route inclusion is a future cohort decision rather than a silent replacement.
- Kingy prior local `kingy-field-bottle-20usd-v1` ledger, September 5, 2026 — five exploratory requests, missing raw videos and human review. New public use is authorized by this build-and-launch request; prior run spending permissions do not authorize new spending.