Product Research · Blog post · source Jun 12, 2026

Launch Tracker in Claude Cowork Live Artifacts

Check daily sales, margins and a new launch without opening Seller Central

The data comes from an Amazon data warehouse he set up on a Mac Mini at home, syncing campaigns, targets, search terms and sales from the Amazon APIs (around 30 tables).

Per source Self-reported by the source, not independently verified.

Curated by Jordache Perozzo, founder of SellerSlice. Reviewed for sources, not personally tested.

Copy-paste promptStarter prompt (written by us)

Build me a one-page launch tracker for ASIN [ASIN], launched on [date]. Day by day since launch, show revenue, units, sessions, conversion rate, ACOS, TACOS, and organic vs PPC units. Read from my connected Amazon data, refresh each time I open it, and flag any day where conversion drops more than 20% from the 7-day average.

The source didn't publish its exact prompt. We wrote this starter so you can try the same workflow. Replace anything in [brackets].

What they did

Mansour Norouzi, a partner at the agency Incrementum Digital who also runs his own supplement brand, does most daily Amazon checks in live artifacts he built with Claude in Cowork: small apps connected to real data that refresh when opened. His folder has a daily sales view pulling orders and units per product from the Seller API, a price-margin explorer that grabs live FBA fees and shows breakeven ACOS at each price point, a competitor price tracker, a promo calendar and a P&L calculator. For a supplement launched two weeks earlier, he had Claude build a launch tracker with revenue, units, sessions, CVR, ACOS, TACOS and the organic vs PPC split, day by day since launch.

How it went

The data comes from an Amazon data warehouse he set up on a Mac Mini at home, syncing campaigns, targets, search terms and sales from the Amazon APIs (around 30 tables). He says each artifact took one conversation to build ('One long conversation tho') and the numbers are there every morning with no exports or spreadsheets.

We only report what the source says. Results are self-reported unless noted.

Worth knowing

The artifacts are only as good as the warehouse behind them, which he set up separately. A commenter warns that Amazon data gets messy once ad, promo and organic signals are mixed in one view.

Original source

Blog postMansour Norouzi (Incrementum Digital) (Mansour Norouzi (LinkedIn))Published Jun 12, 2026 · linkedin.com https://www.linkedin.com/posts/mansournorouzi_claude-live-artifact-activity-7471266357158809601-cQCqOpen the source

Keep going

Related use cases

Product ResearchReported numbers

Daily Rank Check Across 82 SKUs with Claude

Check organic rank for the whole catalog every morning instead of a few bestsellers

82 SKUs“to all 82 SKUs daily” vendor claim

ToolsClaude, Helium 10SourceBlog post · Raeanna Murray (Helium 10)Source dateSep 4, 2026PromptStarter prompt

View promptHide promptStarter prompt

Pull organic rank for every ASIN in my catalog for my tracked keywords. Show, per ASIN and for the whole brand: how many keywords rank 1-10 and 11-20, total indexed keywords, and the combined search volume of keywords in the top 20, each compared with 7 days ago. List the 10 keywords where we lost the most ground.

Product Research

Diagnosing a Hero Product's Sales Drop with Claude and the Helium 10 MCP

Find out why a best seller's sales fell, starting high-level and drilling in ToolsClaude, Helium 10SourcePodcast episode · Ivan Komashinsky (Serious Sellers Podcast #758)Source dateJul 27, 2026PromptStarter prompt

View promptHide promptStarter prompt

Look at [ASIN] over the last two years. How much have sales and profit gone down, and when did the drop start?
Then check sessions and conversion rate in the business reports, keyword rank history and advertising for the same period, and give me the most likely causes in order, with the evidence for each.

Product ResearchReported numbers

Cerebro Export and Keyword Relevance Checks with Claude Cowork

Pull a keyword list and throw out the terms your product doesn't actually match

10-15 minutes“the whole run took 10-15 minutes” vendor-affiliated

ToolsClaude, Helium 10SourcePodcast episode · Andrew Bell and Zoe Lu (AM/PM Podcast)Source dateMay 23, 2026PromptStarter prompt

View promptHide promptStarter prompt

Here is my product truth card: [what the product is, materials, size, power/switch type, style, what's included, what it is NOT].
For each keyword in the attached Cerebro export, mark RELEVANT, PARTIAL or IRRELEVANT against the truth card, with a one-line reason. Flag terms that imply a different product type, a replacement part or a feature we don't have.

Product ResearchOlder source · Apr 2026

Search Query Performance Deep Analysis with Claude Cowork

Turn the Search Query Performance report into keep, grow and cut decisions ToolsClaudeSourceBlog post · Jo LambadjievaSource dateApr 14, 2026PromptStarter prompt

View promptHide promptStarter prompt

Analyze the attached Search Query Performance export (all 33 columns).
1) Split branded vs unbranded queries using my brand terms: [terms].
2) For each query, show my share of impressions, clicks, add-to-carts and purchases vs the market, and the stage where my share drops most.
3) Sort queries into Defend, Growth, Kill and Monitor, with one line of reasoning each.

Product ResearchReported numbers

Sales Database and Buy-Price App Built with Claude and Amazon's Seller Connector

Turn your own sales history into buy prices for sourcing

15 minutes“first dashboard in about 15 minutes of his own time” self-reported

ToolsClaude, Amazon Seller AssistantSourceX post · Wesley ParkerSource dateOct 26, 2026PromptStarter prompt

View promptHide promptStarter prompt

Using the Amazon seller connector, pull my orders, FBA orders and settlements for the last 24 months into a local database with one row per sale.
For each item, show units sold, average and lowest sale price, and the max buy price that leaves 50% profit after Amazon fees and shipping. Read-only: don't change any listing or price.

Product ResearchReported numbers

Multi-Brand Keyword Research via Helium 10 MCP

Run keyword research and rank tracking across several brands without exporting reports

rank 6“from 20+ in its bestseller category to rank 6” vendor claim

ToolsClaude, Helium 10SourceBlog post · Raeanna Murray (Helium 10)Source dateOct 19, 2026PromptStarter prompt

View promptHide promptStarter prompt

For brand [brand], take these root keywords: [list]. Using Helium 10 data, give me each keyword's search volume, my organic rank and the top 3 competitors' ranks, and group keywords into: defend (I'm top 10), push (ranks 11-30, high volume), and ignore. Compare my ranks with last week's and flag every move of 5+ positions.