The Retail Media Desk
Skills · Subagent · Command · PDF
.claude/the-retail-media-desk/zip
skills/06
├ incrementality-test-designer
├ onsite-offsite-allocator
├ post-campaign-review
├ retailer-brief
├ retailer-scorecard
├ rmn-proposal-evaluator
agents/01
├ retail-media-analyst
commands/01
├ /rmn-evaluate
FIELD-GUIDE.pdf8pp
English
One payment · No account · No subscription
- ✓ Download link in your inbox in seconds
- ✓ Secure card checkout · VAT handled
- ✓ 14-day money back, no questions
- ✓ Every future revision included
- ✓ Runs in Claude Code and claude.ai
Built for Claude first, our recommended agent. Skills follow the open Agent Skills standard (SKILL.md), so they also run in any agent that reads it; subagents and slash commands are Claude Code formats.
The Retail Media Desk
Buy retail media without the vendor gloss
six skills·one subagent·one command·8-page guide
Your agent scores a retail media proposal before you sign it, designs the holdout that would prove the placement is incremental, and reads the network's wrap report as a claim.
Executive summary
- The problem
- The network that sells the placement also measures it and writes the report, so attributed return arrives looking like proof. Renewals get argued from that report, and sales the brand would have made anyway are billed back to it as results.
- What it does
- Six skills and a proposal-reading analyst subagent score each proposal on a weighted scorecard, design the holdout or geo test before the flight, read the wrap report as a claim, and keep a per-network record of promised, delivered and proven.
- Why take it
- You answer for the budget, not the network. This lets you keep, fix or stop each line item on evidence the network did not grade itself.
For brand and category teams buying from networks that sell, measure and report the same media.
In the first week you will
- 01Run /rmn-evaluate on the proposal on your desk and get the weighted scorecard and the questions to send back
- 02Design the holdout or geo test with incrementality-test-designer and read iROAS rather than attributed ROAS
- 03Put the network's wrap report through post-campaign-review and get a keep, fix or stop call per line item
Contents of the Kit
- 01Six installable agent skills: rmn-proposal-evaluator, incrementality-test-designer, onsite-offsite-allocator, retailer-brief, post-campaign-review, retailer-scorecard
- 02The retail-media-analyst subagent: reads a rate card or proposal and extracts the terms into a structured sheet
- 03An /rmn-evaluate slash command for scoring any proposal in one pass
- 04A weighted proposal scorecard that flags vendor-gloss phrases and asks the questions that expose them
- 05An 8-page Field Guide on buying from networks that grade their own homework
- 06Plugin manifest included: install as a Claude Code plugin or copy the folders
One skill, as it ships
SKILL.md
rmn-proposal-evaluator
- Use when
- Use when a retailer, marketplace, pharmacy or DIY chain sends a media proposal, joint business plan annex, rate card or partnership deck and the brand has to decide whether and how much to buy.
- Requires
- The proposal
- Will not
- Never invent a score for a dimension the proposal does not address · Never convert the network's ROAS, lift or reach claims into brand-side forecasts · Never compare against industry averages or benchmarks the user has not supplied
- Returns
- Proposal scorecard: {network label} / {proposal title or date}
One of six skills in the kit · All of them install the same way
The contents, in full · six skills, one subagent, one command
- 01rmn-proposal-evaluatorScores a network's proposal on audience claims, measurement method (closed-loop versus modeled), minimum spends, inventory transparency, brand safety, data access and reporting cadence. Flags the vendor-gloss phrases and produces the questions that make them concrete.
- 02incrementality-test-designerDesigns a holdout or geo test: control selection, minimum detectable effect as a labeled assumption, duration, contamination risks, and what to read at the end (iROAS, not ROAS). No invented benchmarks.
- 03onsite-offsite-allocatorSplits a budget between sponsored products and display on the retailer's property and retailer-data audiences off it, by objective: share of shelf, new-to-brand, category launch, promo support. Every allocation carries its assumption.
- 04retailer-briefThe brief a brand sends to a network: objective, SKUs, flight, budget, audiences, measurement demands, reporting demands, assets, approvals. Written so the network cannot later say it was not asked.
- 05post-campaign-reviewTurns the retailer's report into the brand's review: what was measured versus what was inferred, attribution windows checked, what can be computed from the supplied numbers, and a keep, fix or stop call per line item.
- 06retailer-scorecardThe brand's record of a network across a period of flights, which is what a renewal is actually argued from: promised, delivered and proven in three columns that never merge, the register of incremental against attributed reads per placement type, every change of reporting basis mid-period, and the questions the network has left open across flights.
- 07retail-media-analyst (subagent)Reads the rate card or proposal document and extracts the structured terms the evaluator needs.
- 08/rmn-evaluateThe slash command that scores any proposal or rate card in one pass and produces the questions to send back.
Retail media has one structural problem the manual names plainly: the network sells the media, measures the media and reports the media. The Retail Media Desk gives the brand side its own instruments. No benchmark is invented to fill a gap, every allocation and every minimum detectable effect carries its assumption on the face of the output, and what the network measured is reported apart from what it inferred.
Pairs with
The Third Wave of Digital Marketing maps the two fault lines of retail media. This kit is what the brand team runs across them.
Before you buy
01Do I need Claude Code?
No. Every skill uploads to claude.ai as a zip. Subagents and slash commands are the Claude Code extras.
02I am not a programmer.
Install is copying folders into a .claude directory. The Field Guide in each kit assumes nothing beyond that.
03Will it work in my agent?
Skills follow the open Agent Skills standard (SKILL.md), so any agent that reads it can run them. Built for Claude first, our recommended agent.
04Do the tools invent numbers?
Never. Every figure is supplied by you, confirmed by you, or labeled as an assumption in the output.
05How does it arrive?
Instantly. Pay by card and the download link is in your inbox within seconds, and on the confirmation screen.
06Is this a subscription?
No. One payment, lifetime access to your download, every future revision of what you bought included.
07What if it is not useful to me?
Write to info@conclusionmp.com within 14 days of purchase and you get a full refund. No questions, no forms.
Ready when you are
The Retail Media Desk, $12.
Instant download, one payment, every future revision included. Or take The Retail Shelf, four items for $29, and this kit works out at $7.25.