agentleFS
gomarble-aigomarble-ai/marketing-agent/skills/gomarble-skills-ecommerce-brands/SKILL.md

gomarble-skills-ecommerce-brands

Skill15 starsChanged 39 days agoMITView on GitHub

Use for ecommerce business questions beyond a single ad channel: which products to scale or cut, cart abandonment and conversion-rate bottlenecks, customer cohorts and repeat purchase, unit economics and contribution margin, channel mix, promotion impact, device and browser issues, and a weekly ecommerce summary. Works from Shopify and GA4 through GoMarble (read shopify-order-discipline and ga4-source-of-truth first), or from exports.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Minimum data needed: Add-to-cart count, completed purchases, average cart value

    line 38
  2. Distinctive:

    Run Product Performance Ranker

    line 245
  3. Distinctive:

    Check Channel Mix + repurchase by channel

    line 247

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

GoMarble Skills for Ecommerce Brands

9 Core Skills + Weekly Summary for analyzing and optimizing product performance, unit economics, and revenue across all channels.

Read-and-recommend only. These skills never change your accounts. The model reads data and produces recommendations in plain language; you apply them manually.


The 9 Skills

1. Product Performance Ranker

What it does: Ranks SKUs by revenue, ROAS, and profit. Shows your true stars vs dead weight.

How it works:

  • Pareto analysis — top 20% of SKUs drive 80%+ of profit
  • Calculates ROI per product (accounts for COGS, not just ad spend)
  • Identifies high-volume, low-margin vs high-margin, low-volume products
  • Flags seasonal performers vs year-round winners

Minimum data needed: Product-level spend, sales count, revenue, COGS per SKU (last 30 days)


2. Cart Abandonment Debugger

What it does: Finds friction points in checkout. Calculates revenue at risk from abandonment.

How it works:

  • Tracks add-to-cart vs purchase conversion rate by product
  • Identifies high-abandonment products (>50% abandonment rate)
  • Correlates with shipping cost, price tier, and product type
  • Quantifies daily revenue recapturable via recovery email

Minimum data needed: Add-to-cart count, completed purchases, average cart value


3. Customer Cohort Analyzer

What it does: Tracks repeat purchase rate, LTV, and which cohorts are your best payers.

How it works:

  • Segments customers by acquisition month / campaign
  • Calculates repeat purchase rate and LTV by cohort
  • Flags high-acquisition cost cohorts with low repeat rate
  • Identifies best-performing audience segments for acquisition spend

Minimum data needed: Customer creation date, repeat purchase history, customer LTV


4. Unit Economics Calculator

What it does: Isolates profit per order and by product. Shows which channels and products are actually profitable.

How it works:

  • Calculates all-in cost per unit (COGS + fulfillment + payment processing + ad spend)
  • Shows profit margin per SKU
  • Breaks down by channel (email, paid ads, organic)
  • Flags unprofitable products or channels before they bleed cash

Minimum data needed: Sell price, COGS, fulfillment cost, payment fees, ad spend, units sold


5. Channel Mix Optimizer

What it does: Shows which channels (Meta, Google, email, organic) drive the most profitable revenue.

How it works:

  • Calculates ROAS + profit per channel
  • Compares customer LTV by acquisition channel (paid vs organic)
  • Identifies channel overlap (same customer acquired twice?)
  • Recommends budget allocation based on channel efficiency

Minimum data needed: Channel attribution, spend per channel, revenue per channel, customer source


6. Conversion Rate Auditor

What it does: Benchmarks your conversion rate against category norms. Finds the biggest bottleneck.

How it works:

  • Benchmarks site-wide CVR against category average (e.g., apparel vs electronics)
  • Breaks CVR by source (paid, organic, email, direct)
  • Identifies product pages with below-category CVR
  • Flags checkout vs product-page abandonment separately

Minimum data needed: Visits, add-to-cart, purchases (by source and product)


7. Promotional Impact Modeler

What it does: Measures true lift from discounts and sales. Catches discounts that cannibalize rather than expand revenue.

How it works:

  • Compares baseline CVR and AOV to promotion period
  • Calculates incremental revenue (not just total revenue during sale)
  • Identifies customers who would have bought anyway (cannibalization)
  • Measures payback period (how long until discount spending breaks even)

Minimum data needed: Date of promotion, daily revenue before/during/after, customer segment


8. Device & Browser Analyzer

What it does: Shows how mobile vs desktop CVR and AOV differ. Flags bad experiences.

How it works:

  • Compares CVR by device (mobile, tablet, desktop)
  • Tracks AOV by device (mobile users buy less? spend differently?)
  • Flags high-traffic devices with low CVR (performance issue)
  • Identifies browser-specific problems (Safari? Firefox?)

Minimum data needed: Traffic and conversion by device + browser


9. Weekly Ecommerce Summary

What it does: Top products, profit alerts, channel performance. One snapshot, all insights.

Includes:

  • Top 5 products by profit this week
  • Cart abandonment rate + revenue at risk
  • Repeat purchase rate for this week's cohort
  • Profitable vs unprofitable channels
  • Conversion rate vs last week
  • AOV trend + high-AOV products
  • Recommended actions (pause, scale, test, fix)

Minimum data needed: Full Shopify/WooCommerce export (products, orders, customers, events)


How to Use These Skills

  1. Identify the question — Which products to scale? Where's the CVR bottleneck? Is this cohort profitable?
  2. Gather the data — Product performance, customer data, channel data
  3. Run the relevant skill — Get diagnosis + financial impact
  4. Take action — Scale profitable products, pause unprofitable ones, fix high-abandonment

Data Requirements by Source

SkillShopify APICSV ExportGA4Manual
Product Performance Ranker✅✅⚠️ (need revenue)✅
Cart Abandonment Debugger✅✅✅⚠️
Customer Cohort Analyzer✅✅✅✅
Unit Economics Calculator✅✅❌✅ (need COGS)
Channel Mix Optimizer✅ (w/ UTM)✅✅⚠️
Conversion Rate Auditor✅✅✅✅
Promotional Impact Modeler✅✅✅✅
Device & Browser Analyzer❌⚠️✅⚠️
Weekly Ecommerce Summary✅✅✅❌

✅ = works perfectly | ⚠️ = requires manual input or separate export | ❌ = not possible from this source


Key Metrics Reference

MetricDefinitionGood Benchmark
CVR (Conversion Rate)Purchases ÷ Sessions1–3% (varies by category)
AOV (Average Order Value)Total Revenue ÷ OrdersVaries by industry
ROASRevenue ÷ Ad Spend≥3.0 for profitability
LTV (Lifetime Value)Total customer revenue over time≥3x CAC
Repeat Purchase RateRepeat customers ÷ Total customers≥20% is strong
Cart AbandonmentAbandons ÷ (Abandons + Purchases)50–80% is normal
Profit Margin(Revenue − COGS − Fulfillment) ÷ Revenue≥30% healthy
CAC (Customer Acquisition Cost)Ad Spend ÷ New Customers÷ LTV should be ≥3

Output Format for Recommendations

### [Skill Name]: [Product/Channel/Cohort Name]

Finding:
[1–2 sentences describing the issue or opportunity]

Financial Impact:
| Metric | Current | Benchmark | Gap | Annual Impact |
|--------|---------|-----------|-----|---------------|
| ROAS | X | Y | $Z | $ABC |

Recommendation:
1. [Specific action with financial rationale]
2. [Specific action with financial rationale]

When to Run Each Skill

SituationUse SkillTimeline
Deciding which products to advertiseProduct Performance RankerWeekly
Revenue is downConversion Rate AuditorImmediately
Scaling one productCart Abandonment DebuggerBefore scaling
New cohort acquiredCustomer Cohort AnalyzerEvery acquisition campaign
Want to discountPromotional Impact ModelerBefore promotion
Checking profitabilityUnit Economics CalculatorWeekly/Monthly
Comparing channelsChannel Mix OptimizerMonthly
Mobile traffic is highDevice & Browser AnalyzerMonthly
Planning sprintWeekly Ecommerce SummaryEvery Monday

Common Guardrails

Before scaling a product:

  • ✅ Confirm profit margin is positive (not just positive ROAS)
  • ✅ Confirm inventory is sufficient
  • ✅ Confirm repeat purchase rate (scaling one-time buyers is unprofitable)

Before discounting:

  • ✅ Confirm baseline (non-discount) revenue and margin
  • ✅ Calculate breakeven point (how many extra orders to justify discount?)
  • ✅ Segment discount only to high-LTV cohorts

Before pausing a channel:

  • ✅ Confirm it's truly unprofitable (not just high-LTV payback period)
  • ✅ Confirm brand impact (organic traffic may suffer if you pause paid)

Questions to Ask Yourself

  • What's my most profitable product? Am I spending enough on it?
  • Which cohort has the highest LTV? Can I find more customers like them?
  • Is my CVR dropping? Is it a device issue or a general site problem?
  • Are customers abandoning because of price, shipping cost, or something else?
  • Which channel gives me the most repeat customers (not just first order)?

Next Steps

  1. Pull product-level data for last 30 days
  2. Run Product Performance Ranker
  3. Audit Unit Economics for top 5 products
  4. Check Channel Mix + repurchase by channel
  5. Fix highest-impact issue first (usually CVR or product selection)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.