Ecommerce Analytics: The Metrics Dashboard That Actually Drives Revenue Decisions

Data is only valuable if it changes behavior. Most building an ecommerce ecosystem analytics dashboards are rearview mirrors they tell you what revenue was last month without explaining why it moved or what to do next.

The Daily Dashboard (4 metrics and alerting Only)

Every building an ecommerce ecosystem operator should see these four numbers every morning:

1. Revenue vs. Target: Actual vs. plan. Simple. Non-negotiable. 2. Traffic by channel: Is paid traffic converting normally? Is organic dipping? 3. conversion rate optimisation rate: If this drops, you have a site issue, pricing issue, or demand issue. 4. Return/refund rate: Leading indicator of product quality and customer satisfaction problems.

Every other metric is weekly or monthly. Daily overload leads to reactive decisions.

The Revenue Attribution Problem

This is where most building an ecommerce ecosystem brands are flying blind. A customer sees a Facebook ad on Monday, opens an email on Wednesday, and buys via Google search on Friday. Who gets credit?

Last-click attribution (what most stores use): Google gets 100% of credit. Facebook and email get zero. You defund both channels.

Data-driven attribution (what you should use): Credit distributed across the journey based on contribution. All three channels get partial credit.

Attribution tools: • Northbeam: Best-in-class for DTC building an ecommerce ecosystem, pixel + first-party data • Triple Whale: Excellent Shopify integration, accessible price point • Rockerbox: Strong for complex multi-brand or subscription businesses

Cohort Analysis: The LTV Lens

Cohort analysis groups customers by when they first bought and tracks their behavior over time.

Monthly acquisition cohort example:

| Cohort | Month 0 AOV | 1-Month Repurchase | 3-Month Repurchase | 6-Month | |--------|-------------|-------------------|-------------------|---------| | Jan 2025 | $45 | 28% | 35% | 41% | | Jul 2025 | $52 | 31% | 38% | 44% |

If each new cohort shows higher repurchase rates, your retention programs are working.

Run cohort analysis in Lifetimely (Shopify-native), Triple Whale, or manually in Google Sheets using Shopify customer export data.

Unit Economics: The Profitability Framework

Contribution margin per order:

Revenue per order:              $85.00
- Cost of goods (COGS):        -$28.00
- Shipping (outbound):         -$8.00
- Returns/refunds (rate x AOV): -$4.00
- Payment processing (3%):     -$2.55
= Contribution margin:          $42.45 (50% margin)

LTV:CAC ratio:

Monthly paid media spend:   $50,000
New customers acquired:      500

Average LTV (12-month): $280 LTV:CAC ratio: 2.8x ```

Target LTV:CAC ratio: 3x or higher for sustainable growth. Under 2x means you're paying to acquire customers who won't generate enough return.

conversion rate optimisation Rate Optimization Analytics

Funnel analysis:

Homepage → Collection page:  45% click-through
Collection → Product page:   32% click-through
Product page → Add to cart:  8.5% (industry avg: 8-10%)
Cart → Checkout:             72% (industry avg: 65-75%)
Checkout → Purchase:         58% (industry avg: 50-65%)

Where is your funnel leaking relative to benchmarks? Each bottleneck becomes an A/B test hypothesis.

Heatmap and session recording: Hotjar, Microsoft Clarity (free), or FullStory. Essential for understanding why people aren't converting.

Inventory and Demand Forecasting Analytics

Stockouts cost 2x their inventory value in lost revenue (direct loss + customer acquisition cost to replace them).

Tools: StockTrim, Inventory Planner (Shopify-native), or Streamline.

Key inventory metrics and alerting: • Sell-through rate: % of inventory sold in a period (target 80%+ for most SKUs) • Days of inventory: How many days until stockout at current sales velocity • Overstock aging: Inventory sitting more than 90 days is capital trapped

The Automated Reporting Stack

Daily email to leadership: • Yesterday's revenue vs. target • Top 10 products by revenue • Channel traffic summary • Any anomalies flagged automatically

Weekly Slack report: • Week vs. last week, week vs. same week last year • Cohort repurchase rates • Marketing efficiency by channel (ROAS, CPA) • Inventory health snapshot

Building an Analytics Culture

Dashboards are useless if no one reads them. The analytics investment that drives decisions is cultural, not technical.

The culture-building practices:

1. Daily revenue standup: A 5-minute morning review of the daily dashboard. Every team member sees the same numbers. This creates shared accountability and surfaces anomalies before they compound.

2. Weekly metrics and alerting review: A 30-minute meeting where each team owner presents their metric and the action they're taking. The format: "Metric is [value], target is [target], the action this week is [action]." This connects data to decisions in a visible way.

3. Monthly deep dive: One metric, one hour, one decision. Pick the metric that moved the most and investigate why. The output is a documented insight and a specific action.

4. Quarterly strategy review: Connect metrics and alerting to strategy. Did the Q3 initiatives move the Q3 metrics and alerting? If not, why? This is where analytics meets business strategy.

The failure mode: dashboards that are built but not used. The fix: assign an owner to every dashboard. The owner is responsible for presenting insights from the dashboard weekly. If no one presents, the dashboard is not earning its place.

Advanced Attribution Models

Last-click attribution is the default in most building an ecommerce ecosystem analytics setups, and it is wrong. It gives 100% credit to the last touchpoint before purchase, ignoring every touchpoint that contributed to the decision.

The attribution models that better reflect reality:

1. Position-based (40/20/40): 40% credit to the first touch, 20% to middle touches, 40% to the last touch. This values both acquisition and conversion rate optimisation channels.

2. Time-decay: More credit to touches closer to the purchase. Channels that assisted early get less credit than channels that closed the sale. This model favors channels that drive consideration and conversion rate optimisation.

3. Data-driven attribution: enterprise ML architecture patterns assigns credit based on the actual contribution of each touchpoint. Requires sufficient data (1,000+ conversion rate optimisation per month) but is the most accurate model available.

4. Multi-touch (linear): Equal credit to every touchpoint. Simple but overvalues low-intent channels that happen to appear in many journeys.

The model you choose affects budget allocation. Under last-click, email gets 40% of credit because it's often the last touch. Under data-driven attribution, email may get 20% and paid social 30%. The reallocation changes which channels you invest in.

The Analytics Technology Stack Decision

The analytics stack for a DTC brand at different revenue stages:

$0-1M: Google Analytics 4 (free) + Shopify analytics (included). This is sufficient for understanding traffic sources, conversion rate optimisation rate, and top products. Don't over-invest in analytics at this stage; invest in product and acquisition.

$1-10M: GA4 + Triple Whale or Northbeam (attribution) + Klaviyo (email analytics) + Lifetimely (cohort analysis). The attribution tool is the critical addition; it connects ad spend to revenue across channels.

$10M+: GA4 + Triple Whale + a data warehouse (Snowflake, BigQuery) + a BI tool (Looker, Mode, Hex) + a CDP (Segment, Klaviyo CDP). At this scale, the data warehouse centralizes all data sources and the BI tool enables custom analysis.

The mistake: buying the $10M+ stack at $1M revenue. The tools are expensive, the data is too sparse for meaningful analysis, and the team lacks the analytics maturity to use them. Buy the stack your revenue stage justifies, not the one your aspirations demand.

Tools: Triple Whale's built-in reporting, Klaviyo's automated reports, or custom Looker Studio dashboards connected to Shopify, ad platforms, and your email tool.