Skip to content

takechanman1228/claude-ecom / skills/ecom

Ecommerce business review for D2C stores from order transaction CSVs. Runs the bundled Python engine (KPI trees, ~30 pass/watch/fail health checks, 30d/90d/365d windows), then interprets the results: either a full narrative business review written to REVIEW.md, or an inline answer to a focused question.

references/impact-formulas.md

View on GitHub
# Ecom Health Check — Revenue Impact Calculation Formulas

## Quick Reference Table

| Check ID | Metric | Formula (per period) | Typical Improvement | Annualised Impact Example ($10M revenue) |
|----------|--------|---------------------|---------------------|------------------------------------------|
| repeat_purchase_rate | Repeat Purchase Rate | `New_Customers * RPR_Improvement_pp/100 * LTV_Repeat` | +3-5 pp | $150K-$500K |
| repeat_customer_revenue_share | Repeat revenue share | `Total_Revenue * Improvement_pp/100 * (1 - Churn_Volatility)` | +3-5 pp | $75K-$200K |
| avg_discount_rate_trend | Avg discount rate | `Discounted_Revenue * Reduction_pp / 100` | -1-2 pp | $50K-$150K |
| large_order_dependency | Large order dependency | Severity-based estimate (no specific formula) | varies | varies |

**Note:** "pp" = percentage points. All examples assume a $10M annual revenue baseline. Actual impact varies by category, AOV, and traffic volume.

---

## repeat_purchase_rate — Repeat Purchase Rate

### What it measures
The share of customers who make a second purchase (repeat purchase rate), and the broader returning customer ratio. Repeat customers drive disproportionate revenue and have dramatically higher conversion rates (60-70%) compared to new prospects (5-20%).

### Calculation Formula

```
Returning_Customer_Ratio = Repeat_Customers / Total_Customers * 100

Repeat_Purchase_Rate = Customers_With_2+_Orders / Total_Customers * 100

LTV_Uplift_Per_1pp_RPR = New_Customers * 0.01 * (LTV_Repeat - LTV_OneTime)

Revenue_Impact = LTV_Uplift_Per_1pp_RPR * (1 + Downstream_Acceleration)
```

**Worked example (per 1 pp repeat purchase rate improvement):**
- Annual new customers: 50,000
- LTV of one-time buyer: $215
- LTV of 2+ purchase buyer: $538 (2.5x multiplier)
- Revenue uplift = 50,000 * 0.01 * ($538 - $215) = **$161,500/year**
- With downstream acceleration (repeat buyers convert to 3rd purchase at 38.8%): effective uplift is higher

### Key LTV Multipliers by Purchase Count

| Purchase Count | Relative LTV | F(n) to F(n+1) Conversion |
|---------------|-------------|--------------------------|
| 1 (one-time) | 1.0x | ~18.8% make a repeat purchase |
| 2 | 2.5x | ~38.8% convert to F3 |
| 3 | 3.8x | ~50%+ convert to F4 |
| 5+ | 7.3x | High retention zone |

### Industry Benchmarks

| Category | Repeat Customer Rate | Repeat Revenue Share |
|----------|---------------------|---------------------|
| Overall ecommerce | 15-30% | 25-45% |
| Consumables (top) | 39-44% | 62-67% |
| Health & supplements | 29% | 45-55% |
| Beauty & cosmetics | 25.9% | 35-45% |
| Fashion & apparel | 15-26% | 13-25% |
| Home & furniture | 14.7% | 12-18% |
| Pet supplies | 30%+ | 45-55% |
| Grocery & food delivery | 30-40%+ | 50-65% |
| Durables / high-ticket | 11-17% | 12-18% |

### Typical Improvement Rates
- Loyalty / VIP tier programme: +5-10 pp repeat rate, 15-25% annual revenue lift
- Post-purchase email sequence (3-5 emails): +3-5 pp repeat purchase rate
- Personalised product recommendations: +2-4 pp
- Subscription / auto-replenish models: +10-15 pp for consumables
- 5% increase in retention -> 25-95% increase in profits (Bain & Company)
- Existing customers spend 67% more after 30+ months (Bain & Company)
- Probability of selling to existing customer: 60-70% vs 5-20% for new

### Sources
- BS&Co — Repeat Purchase Rate Benchmarks: 18.8% across 156K DTC customers (2026)
- BS&Co — LTV by Purchase Count: DTC Benchmarks from 162K customers
- Rivo — VIP Customer Repeat Rate Statistics (27 stats)
- Geckoboard — Repeat Customer Rate KPI (25-30% benchmark)
- Bain & Company — customer retention and LTV research

---

## repeat_customer_revenue_share — Repeat Revenue Share

### What it measures
The proportion of total revenue generated by repeat (returning) customers. Higher repeat revenue share indicates more predictable, stable revenue with lower acquisition cost dependency.

### Calculation Formula

```
Repeat_Revenue_Share = Revenue_From_Repeat_Customers / Total_Revenue * 100

Revenue_Stability_Score = Repeat_Revenue_Share * (1 - Monthly_Revenue_CV)
  where CV = coefficient of variation of monthly revenue

Impact_Per_1pp_Improvement = Total_Revenue * 0.01 * (1 - Blended_CAC_Rate)
  (repeat revenue has near-zero marginal CAC vs 15-30% for new customer revenue)
```

**Worked example (per 1 pp shift from new to repeat revenue):**
- Annual revenue: $10M
- Current repeat revenue share: 35%
- CAC as % of new customer revenue: 25%
- Effective savings = $10M * 0.01 * 0.25 = **$25,000/year in CAC savings**
- Plus: reduced revenue volatility and improved forecasting accuracy

### Industry Benchmarks

| Category | Repeat Revenue Share |
|----------|---------------------|
| Overall ecommerce | 25-45% |
| Consumables (top performers) | 62-67% |
| Consumables (mid-tier) | 55-64% |
| Fashion & apparel | 13-25% |
| Durables / high-ticket | 12-18% |
| Subscription-based | 65-80% |

### Key Insight: Over-indexing
- Consumable brands: repeat customers generate 1.5x their share of revenue (e.g., 44% of customers drive 67% of revenue)
- Fashion / durables: repeat customers generate ~1:1 their share (no over-index)
- Implication: repeat revenue improvement strategies have highest ROI for consumable categories

### Typical Improvement Rates
- Loyalty programme implementation: +5-10 pp repeat revenue share over 12 months
- Subscription model introduction (consumables): +15-25 pp
- Email lifecycle marketing (post-purchase flows): +3-5 pp
- Personalisation engine: +2-4 pp
- Healthy target: 40%+ repeat revenue share for most ecommerce

### Sources
- BS&Co — Repeat Purchase Rate Benchmarks (2026): 156K DTC customer analysis
- Finaloop — Ecommerce Profit Benchmarks (2024-2025 P&L data)
- Rivo — VIP Customer Repeat Rate Statistics

---

## avg_discount_rate_trend — Average Discount Rate (Margin Erosion)

### What it measures
The average discount percentage applied across all discounted transactions. avg_discount_rate_trend measures how deep discounts are when they are applied.

### Calculation Formula

```
Avg_Discount_Rate = Total_Discount_Amount / Gross_Revenue_Before_Discounts * 100

Margin_Recovery_Per_1pp = Discounted_Revenue * 0.01

Effective_Margin_Impact = Margin_Recovery / (Revenue * Gross_Margin_Rate)
  (percentage of margin recovered relative to total margin pool)
```

**Worked example (per 1 pp reduction in average discount depth):**
- Annual gross revenue (before discounts): $12M
- Discounted portion: $4M (33% of revenue is discounted)
- Current average discount: 20%
- Margin recovered per 1 pp = $4M * 0.01 = **$40,000/year**
- If gross margin is 55%: this represents 0.6% of total margin pool recovered

### Benchmarks

| Discount Depth Tier | Range | Typical Context |
|---------------------|-------|-----------------|
| Light discounting | 5-10% | Loyalty perks, first-purchase incentive |
| Moderate discounting | 10-20% | Standard promotions, seasonal sales |
| Heavy discounting | 20-30% | Clearance, competitive pricing |
| Aggressive discounting | 30%+ | End-of-life, Black Friday, distressed inventory |

### Margin Erosion Math
| Gross Margin | 10% Discount Impact | 20% Discount Impact | 30% Discount Impact |
|-------------|---------------------|---------------------|---------------------|
| 30% | Gives away 33% of profit | Gives away 67% of profit | Sells at cost |
| 40% | Gives away 25% of profit | Gives away 50% of profit | Gives away 75% |
| 50% | Gives away 20% of profit | Gives away 40% of profit | Gives away 60% |
| 60% | Gives away 17% of profit | Gives away 33% of profit | Gives away 50% |

### Typical Improvement Rates
- Replacing flat discounts with tiered spend thresholds: -2 to -3 pp depth
- Gift-with-purchase instead of % off: equivalent conversion lift at 0% discount depth
- Personalised discount depth (ML-based willingness-to-pay): -3 to -5 pp
- Reducing discount frequency (fewer sale events): -1 to -2 pp average depth

### Sources
- Opensend — Average Discount Rate Statistics for eCommerce (2025)
- SalesSo — Discounting Statistics in Sales (2025)
- Finaloop — Ecommerce Profit Benchmarks: P&L metrics (2024-2025)

---

## Cross-Metric Impact Chains

Many of these metrics are interconnected. Improving one often creates cascading effects on others.

```
Repeat Purchase Rate Improvement (repeat_purchase_rate)
  --> Higher repeat revenue share (repeat_customer_revenue_share)
       --> Lower blended CAC
            --> Revenue stability + margin improvement

Discount Reduction (avg_discount_rate_trend)
  --> Direct margin recovery
       --> Better brand perception
            --> Higher full-price conversion
```

### Priority Matrix: Effort vs Impact

| Priority | Check | Typical Effort | Typical Impact | Time to Realise |
|----------|-------|---------------|----------------|-----------------|
| 1 | repeat_purchase_rate | Medium | Very High (LTV) | 4-12 weeks |
| 2 | avg_discount_rate_trend | Low | Medium-High | Immediate |
| 3 | repeat_customer_revenue_share | Medium | Medium | 8-24 weeks |

---

## Per-Check Impact Models

Formal models using the $10M store baseline.

### repeat_purchase_rate — Repeat Purchase Rate (Formal)

\(\Delta LTV_{12} = (p_{RPR} \cdot u)(O_{rep} - 1) \cdot AOV_{rep}\), \(\Delta R_{annual} = N_{new} \cdot \Delta LTV_{12}\)

```python
def ltv_uplift_from_rpr(p_rpr, uplift_rel, o_rep, aov_repeat):
    return p_rpr * uplift_rel * (o_rep - 1.0) * aov_repeat

def annual_revenue_uplift_from_rpr(new_customers, p_rpr, uplift_rel, o_rep, aov_repeat):
    return new_customers * ltv_uplift_from_rpr(p_rpr, uplift_rel, o_rep, aov_repeat)
```

**Example.** \(N_{new}=60K\), \(p_{RPR}=25\%\), 5% relative lift, \(O_{rep}=3\), \(AOV_{rep}=\$110\): \(\Delta LTV_{12}=\$2.75\), annual = **$165K**. **Confidence: Medium-Low.** Caveat: +5pp absolute vs +5% relative yields ~4x difference.

### repeat_customer_revenue_share — Repeat Revenue Share (Formal)

\(\Delta R_{shift} = R(s_{tgt}-s_{cur})\), \(\text{Net savings} = (\Delta R_{shift}/AOV_{new}) \cdot (CAC - \text{retention\_cost})\)

```python
def repeat_share_shift_savings(annual_revenue, s_cur, s_tgt, aov_new, cac_new, retention_cost=0.0):
    delta_rev = annual_revenue * (s_tgt - s_cur)
    fewer_new = delta_rev / aov_new
    return {"delta_repeat_revenue": delta_rev, "fewer_new_orders": fewer_new,
            "net_savings": fewer_new * cac_new - fewer_new * retention_cost}
```

**Example.** 35% to 44%, CAC=$40, retention=$5: **$315K net savings**. **Confidence: Medium.** Caveat: constant total revenue is simplifying.

### avg_discount_rate_trend — Discount Reduction (Formal)

\(P'=P_0(1-d+\Delta d)\), \(Q' \approx Q(1+\epsilon \cdot \Delta d/(1-d))\), \(\Delta GP = (P'-c)Q' - (P-c)Q\). Revenue quality: \(\Delta R_{\text{const vol}} = R/(1-d) \cdot \Delta d\).

```python
def discount_margin_recovery(list_price, cogs, d_current, d_reduction_pp, elasticity, annual_orders):
    p, p_new = list_price*(1-d_current), list_price*(1-d_current+d_reduction_pp)
    q_new = annual_orders * (1 + elasticity * ((p_new/p) - 1.0))
    return {"delta_revenue": p_new*q_new - p*annual_orders,
            "delta_gross_profit": (p_new-cogs)*q_new - (p-cogs)*annual_orders}
```

**Example.** \(P_0=\$125\), d=20%, c=$55, Q=100K, 1pp reduction, \(\epsilon=-1.5\): **GP +$38K** (revenue -$65K). Constant-volume: $125K. **Confidence: Medium.** Caveat: elasticity is the key unknown.
deprecation-and-migrationaddyosmani/agent-skillsManages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another. Use when migrating a database schema in production, such as renaming or dropping a column without downtime (expand/contract). Use when deciding whether to maintain or sunset existing code.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
performance-optimizationaddyosmani/agent-skillsOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall
agent-cost-reportthedotmack/claude-memBelievable agent cost report for any period, default the last 7 full days PT, not counting today. Measured tokens from Claude Code transcripts priced at OpenRouter list prices (ESTIMATED), measured provider spend when a sanctioned source exists, note-taker cost separate, Timing-style HTML/PDF plus report.json, line-items.csv, evidence.json.Stars: 118kUpdated Claude CodeCodexVerifiedInstall