Every marketplace provides dozens of metrics. Amazon alone offers Brand Analytics, Business Reports, Advertising Console, Inventory Performance Index, Account Health Dashboard, and more — each with its own set of numbers, charts, and trends. The result is information overload that paralyzes decision-making.
The brands that outperform on marketplaces are not the ones tracking the most metrics. They are the ones tracking the right metrics — the ones that directly connect to actionable decisions. After managing analytics across 50+ brands in 18 countries, we have distilled marketplace analytics down to the metrics that actually matter.
12 | Core metrics that drive 90% of decisions
The Vanity Metric Problem
A vanity metric is any number that looks good in a report but does not inform a specific action. Pageviews, total sessions, social media followers, total review count — these metrics make founders feel good but do not tell you what to change.
The distinction between vanity and actionable metrics is simple: if a metric changes, do you know what to do differently? If the answer is no, it is a vanity metric.
| Vanity Metric | Why It Is Misleading | Actionable Alternative |
|---|---|---|
| Total sessions | High sessions with low conversion = wasted traffic | Session-to-conversion rate |
| Total revenue | Revenue without margin context is meaningless | Contribution margin per unit |
| Total reviews | More reviews ≠ better conversion after a threshold | Review velocity (new reviews/week) |
| Best Seller Rank (BSR) | BSR is relative and category-dependent | Organic unit sales per day |
| Impressions | Impressions without clicks = irrelevant visibility | Click-through rate (CTR) |
| Social media followers | Followers rarely correlate with marketplace sales | Attributed revenue from social |
Before adding any metric to your dashboard, ask: "If this number drops by 20%, what specific action would we take?" If you cannot answer in one sentence, the metric does not belong on your primary dashboard. Reserve it for deep-dive analysis, not daily monitoring.
The 12 Metrics That Matter
We organize marketplace metrics into four categories: Revenue Efficiency, Customer Acquisition, Inventory Health, and Customer Satisfaction. Each category contains three core metrics.
Revenue Efficiency
1. Contribution Margin per Unit
Contribution margin = Selling price − COGS − Marketplace fees − Fulfillment − Advertising cost per unit
This is the single most important metric in marketplace selling. It tells you exactly how much profit each unit sold generates after all variable costs. Track it by SKU, by marketplace, and by month. If contribution margin is negative or declining, everything else is noise.
2. Total Advertising Cost of Sale (TACoS)
TACoS = Total ad spend ÷ Total revenue (organic + paid)
While ACoS measures advertising efficiency in isolation, TACoS reveals whether your advertising is building organic visibility. Healthy TACoS trends downward over time as organic sales grow relative to paid. A stable or increasing TACoS means your advertising is not translating into lasting ranking gains.
3. Revenue per Session
Revenue per session = Total revenue ÷ Total sessions
This composite metric captures both conversion rate and average order value in a single number. It is more useful than tracking conversion rate and AOV separately because it reflects the total economic value of each listing visit.
| Metric | How to Calculate | Good | Warning | Critical |
|---|---|---|---|---|
| Contribution Margin | Revenue − all variable costs | >25% | 10–25% | <10% |
| TACoS | Ad spend ÷ total revenue | <10% | 10–18% | >18% |
| Revenue per Session | Revenue ÷ sessions | >$4.00 | $2–$4 | <$2.00 |
Customer Acquisition
4. Click-Through Rate (CTR)
CTR = Clicks ÷ Impressions
CTR measures the effectiveness of your search result presentation — your main image, title, price, and review stars. Low CTR means shoppers see your product but choose not to click. This is a listing quality signal, not an advertising metric.
5. Cost per Acquisition (CPA)
CPA = Total ad spend ÷ Total orders from ads
CPA tells you the actual dollar cost to acquire each customer through advertising. Unlike ACoS, which is a percentage, CPA is an absolute dollar figure that is easy to compare against contribution margin. If your CPA exceeds your contribution margin, every ad-driven sale loses money.
6. Organic Sales Percentage
Organic % = Organic units ÷ Total units
The percentage of your sales that come without advertising spend. This metric reveals your brand's true marketplace strength. A mature product should generate 50–70% of sales organically. Below 40% organic suggests over-dependence on advertising and weak organic ranking.
Inventory Health
7. Days of Supply
Days of supply = Current inventory ÷ Average daily sales
This metric tells you how many days your current inventory will last at the present sell-through rate. Optimal days of supply depends on your lead time — typically 1.5x to 2x your lead time. Too high means excess capital and storage costs. Too low means stockout risk.
8. Stockout Rate
Stockout rate = Days out of stock ÷ Total days in period
Every day out of stock is a day of lost sales, lost organic ranking, and potentially lost customers to competitors. Industry leaders maintain stockout rates below 2%. The average e-commerce brand runs at 8–15% stockout rates. Each percentage point of stockout rate costs approximately 1–1.5% of potential annual revenue.
9. Inventory Turnover
Inventory turnover = COGS ÷ Average inventory value
Turnover measures how efficiently you convert inventory investment into sales. Higher turnover means less capital tied up in inventory. For most e-commerce categories, 6–12 turns per year is healthy. Below 4 turns indicates overstock or slow-moving inventory.
A stockout does not just cost you the lost sales during the out-of-stock period. On Amazon, a stockout of 7+ days typically drops your organic ranking by 30–50%, which reduces sales for 2–4 weeks after you are back in stock. The true cost of a 7-day stockout is approximately 3–4 weeks of reduced sales volume. For a product selling 50 units per day at $25, that is $26,000–$35,000 in lost revenue.
Customer Satisfaction
10. Return Rate
Return rate = Returned units ÷ Shipped units
Returns destroy margins. Each return costs the referral fee (partially), the fulfillment fee (fully), and often the product value if it cannot be resold as new. Track return rate by SKU and by reason code. A rising return rate signals product quality issues, listing inaccuracy, or packaging problems that must be addressed immediately.
11. Review Rating Trend
Track the rolling 30-day average review rating rather than the lifetime average. The recent trend matters more than the cumulative score. A product with a 4.5 lifetime rating but a 3.8 rating over the last 30 days has a quality or customer experience problem that requires investigation.
12. Customer Feedback Score (Seller Rating)
Your seller feedback rating affects Buy Box eligibility, account health, and customer trust. Maintain above 95% positive feedback. Below 90% puts you at risk of account-level consequences.
| Metric | How to Calculate | Target | Review Frequency |
|---|---|---|---|
| Days of Supply | Inventory ÷ daily sales | 45–90 days | Daily |
| Stockout Rate | OOS days ÷ total days | <2% | Weekly |
| Inventory Turnover | COGS ÷ avg inventory | 6–12x/year | Monthly |
| Return Rate | Returns ÷ shipments | <5% (category-dependent) | Weekly |
Dashboard Design Principles
Knowing which metrics to track is half the battle. Presenting them in a way that enables fast, accurate decision-making is the other half.
The Three-Layer Dashboard Architecture
Layer 1: Executive Summary (1 screen). Show the 12 core metrics with trend indicators (up/down/stable) and color coding (green/yellow/red against targets). This layer answers "Is the business healthy?" in under 30 seconds.
Layer 2: Diagnostic View (per metric). When a metric in Layer 1 triggers a warning, Layer 2 provides drill-down detail. If contribution margin drops, Layer 2 shows margin by SKU, by marketplace, and by cost component to identify the source.
Layer 3: Deep Analysis (on demand). Full exploratory analytics for specific investigations — cohort analysis, market basket analysis, competitive benchmarking, seasonal decomposition. This layer is used weekly or monthly, not daily.
The biggest mistake in analytics dashboard design is putting everything on one screen. When everything is visible, nothing is visible. A well-designed executive dashboard should contain no more than 12–15 data points. Every additional metric reduces the attention given to the metrics that matter.
Set a fixed weekly analytics review — same day, same time, same team. Spend the first 15 minutes on Layer 1 (executive summary) to identify any metrics trending toward warning or critical. Spend the next 30 minutes on Layer 2 (diagnostic drill-down) for any flagged metrics. Assign specific actions with owners and deadlines for any issue identified. This 45-minute weekly ritual replaces hours of ad-hoc dashboard browsing and produces consistently better outcomes.
Real-Time vs. Batch Reporting
Not every metric needs real-time updates. Implementing real-time dashboards is expensive (in engineering cost, API costs, and cognitive load), and most marketplace metrics do not change meaningfully hour to hour.
| Metric | Recommended Frequency | Why |
|---|---|---|
| Inventory levels / days of supply | Real-time or daily | Stockout prevention requires current data |
| Advertising spend and ACoS | Daily | Budget pacing and bid adjustments |
| Contribution margin | Weekly | Fee and cost changes are not instantaneous |
| TACoS and organic % | Weekly | Trends matter more than daily fluctuations |
| Return rate | Weekly | Pattern identification needs multi-day aggregation |
| Inventory turnover | Monthly | Meaningless at shorter intervals |
| Customer satisfaction scores | Weekly | Review sentiment needs aggregation |
Common Analytics Mistakes
Mistake 1: Comparing metrics across categories. A 15% conversion rate in supplements is not comparable to a 15% conversion rate in consumer electronics. Benchmarks must be category-specific. Always compare your metrics against category averages, not cross-category benchmarks.
Mistake 2: Reacting to daily noise. Daily sales fluctuations of ±20% are normal statistical variation, not actionable signals. Set alert thresholds based on rolling 7-day averages to filter noise from genuine trends.
Mistake 3: Tracking metrics without targets. A metric without a target is just a number. Set specific, time-bound targets for each core metric based on historical performance and business goals. Review and adjust targets quarterly.
Mistake 4: Measuring activity instead of outcomes. "We tested 15 new keywords this week" is an activity. "Our organic click share increased by 3 percentage points this week" is an outcome. Build your analytics around outcomes, not activities.
FAQ
What analytics tools should I use for Amazon selling?
For most sellers, a combination of Amazon's native tools and one third-party analytics platform provides complete coverage. Amazon Brand Analytics, Business Reports, and the Advertising Console are free and essential. For third-party analytics, tools like Helium 10 ($97–$397/month), Jungle Scout ($49–$129/month), or Sellerboard ($19–$79/month) provide profitability tracking, keyword analytics, and competitor monitoring. For larger operations managing 100+ SKUs across multiple marketplaces, we recommend building custom dashboards in tools like Looker, Tableau, or Power BI that pull data from Amazon's SP-API. This provides the flexibility to create the three-layer dashboard architecture described above with metrics tailored to your specific business.
How do I calculate true profitability per SKU on Amazon?
True SKU-level profitability requires accounting for all cost layers: revenue minus referral fee, FBA fulfillment fee, storage fee (allocated per unit), advertising cost per unit, return cost per unit (return rate × average return cost), inbound shipping cost per unit, COGS (landed cost including freight and duties), currency conversion fees (for cross-border), and VAT/compliance costs (allocated per unit). Amazon's settlement reports provide most of the fee data, but you need to supplement with your own COGS data, allocated overhead, and compliance costs. Tools like Sellerboard automate much of this calculation by integrating settlement data with manually entered COGS and overhead figures.
What is a good conversion rate on Amazon?
Amazon average conversion rate across all categories is approximately 10–15% for organic traffic and 8–12% for advertising traffic. However, category variation is enormous. Consumables and essentials often convert at 15–25% because the purchase is habitual. High-consideration electronics may convert at 5–10%. Apparel converts at 3–8% due to fit uncertainty. Rather than benchmarking against a platform average, compare your conversion rate against your category average (available in Brand Analytics) and your own historical performance. A conversion rate consistently below your category average indicates listing quality issues that need addressing.
How often should I review my marketplace analytics?
We recommend a daily 5-minute check (inventory levels and advertising spend pacing), a weekly 45-minute structured review (all 12 core metrics), and a monthly 2-hour deep dive (competitive analysis, trend identification, strategic planning). Quarterly, conduct a comprehensive business review that examines all metrics in the context of your annual goals and adjusts targets and strategies accordingly. The most important discipline is consistency — a structured weekly review produces far better outcomes than sporadic, reactive data checking driven by anxiety rather than process.
Should I hire a data analyst for marketplace analytics?
If your marketplace revenue exceeds $2 million annually, a dedicated analytics resource (full-time or fractional) typically pays for itself through improved inventory management, advertising efficiency, and pricing optimization. Below $2 million, the core analytics described in this guide can be managed by an operations manager using Amazon's native tools and one third-party platform. The key is whether analytics drives regular decision-making in your business. If your team reviews data weekly and makes specific changes based on what they find, you are extracting value from analytics. If dashboards exist but nobody looks at them regularly, a data analyst will not help — the problem is process, not talent.