A simple model for understanding performance

Revenue is down. Is it traffic? Conversion? Average order value? Customer mix? Marketing effectiveness? Seasonality? Competitive pressure? The questions multiply. Dashboards offer dozens of metrics. Analysis paralysis sets in. Hours later, you’re still not sure what’s happening or what to do about it. A simple model cuts through this complexity. Not by ignoring nuance, but by providing structure that makes nuance useful.

The best mental models are simple enough to hold in your head and powerful enough to explain most of what you see. For e-commerce performance, one model does most of the work.

The fundamental equation

E-commerce simplified:

Revenue = Traffic × Conversion Rate × Average Order Value

Every e-commerce business runs on this equation. Traffic is how many potential customers arrive. Conversion rate is what percentage buy. Average order value is how much they spend. Multiply them together, you get revenue.

Why this equation matters

Any revenue change must come from one or more of these three components. There’s no other source. This constraint simplifies diagnosis enormously.

The power of multiplication

Small changes in each component multiply together. 10% more traffic, 10% better conversion, 10% higher AOV doesn’t equal 30% more revenue—it equals 33% more revenue. Multiplication creates leverage.

The vulnerability of multiplication

The reverse is also true. Small declines compound. 10% drops in each component means 27% less revenue. Multiplication amplifies both gains and losses.

Using the model for diagnosis

When something changes:

Revenue up or down?

Start with the outcome. Revenue changed. Now work backward through the equation to find the source.

Check traffic first

Did traffic change? If traffic dropped 20% and revenue dropped 20%, you’ve likely found the cause. Traffic is often the simplest explanation.

Check conversion second

If traffic is stable, did conversion change? Same visitors buying at different rates points to site experience, product appeal, or external factors affecting purchase behavior.

Check AOV third

If traffic and conversion are stable, did order size change? Different product mix, pricing changes, or discount usage affects how much each order is worth.

Often it’s a combination

Revenue changes frequently involve multiple components moving together. Slight traffic drop plus slight conversion drop creates larger revenue drop. The model helps you see the combination.

Going one level deeper

Each component has sub-components:

Traffic breakdown

Traffic comes from sources: organic search, paid ads, direct, social, email, referrals. Traffic changes usually trace to specific source changes. Which source moved?

Conversion breakdown

Conversion varies by traffic source, device, new versus returning, and funnel stage. Overall conversion change might reflect mix shift rather than actual conversion change within segments.

AOV breakdown

AOV reflects product mix, quantity per order, and pricing. Changes might come from different products selling, different quantities, or discount behavior.

When to go deeper

The top-level model identifies which component changed. Sub-components explain why. Go deeper only after identifying which branch to explore.

The model as filter

Reducing noise:

Does it affect the equation?

Many metrics exist that don’t directly connect to revenue. They might be interesting but aren’t essential. The model helps distinguish core metrics from peripheral ones.

Leading versus lagging

Traffic, conversion, and AOV are somewhat leading indicators of revenue. Revenue is the lagging output. Understanding which metrics lead helps predict rather than just report.

What actually moves the needle?

Changes that don’t affect traffic, conversion, or AOV don’t affect revenue. The model clarifies what matters by defining the only paths to revenue change.

Thinking in ranges, not points

Adding realism:

Normal ranges for each component

Traffic might normally range from 800-1200 daily visitors. Conversion from 2.0-2.6%. AOV from $45-$55. Knowing ranges defines what’s normal versus unusual.

Concern thresholds

Values outside normal range warrant attention. Values within normal range are just fluctuation. Ranges prevent overreaction to noise.

The combination effect

All three components at the low end of their ranges produces very different revenue than all three at the high end. Understanding ranges helps understand revenue variance.

Common patterns the model reveals

Typical scenarios:

Traffic drop, everything else stable

Marketing problem or external factor. The business converts fine; fewer people are arriving. Focus on traffic sources.

Conversion drop, traffic stable

Site experience problem, product problem, or traffic quality change. Same number arriving but fewer buying. Focus on why visitors aren’t converting.

AOV drop, traffic and conversion stable

Product mix shift, discount overuse, or pricing issue. People are buying but spending less. Focus on what they’re buying and for how much.

Traffic up, revenue flat

Traffic quality decreased. More visitors but lower conversion or AOV. New traffic isn’t as valuable as existing traffic.

All components slightly down

Broad business softness or seasonal effect. No single cause; general decline across dimensions. May require strategic rather than tactical response.

What the model doesn’t capture

Limitations to acknowledge:

Profitability

Revenue isn’t profit. The model doesn’t account for cost of goods, marketing spend, or operational costs. Revenue can grow while profits shrink.

Customer lifetime value

Single transaction focus. The model doesn’t capture whether customers return or how much they’re worth over time.

Cash flow timing

Revenue recognition versus cash collection. The model tracks revenue, not when money actually arrives.

Sustainability

A promotion might spike all three components temporarily but damage long-term health. The model captures the spike but not the damage.

When to extend the model

As businesses mature, extending toward profit, LTV, and cash flow becomes important. But the basic model remains the foundation that these extensions build upon.

Using the model proactively

Beyond diagnosis:

Setting goals

“We want 20% revenue growth.” The model shows paths: 20% more traffic, or 20% better conversion, or some combination. Goals become actionable through the equation.

Prioritizing initiatives

Which component has most room for improvement? If traffic is strong but conversion is below benchmark, conversion initiatives offer more leverage.

Evaluating opportunities

“This tool claims to improve X.” Does X connect to traffic, conversion, or AOV? If not clearly, the value is questionable.

Forecasting

Expected traffic times expected conversion times expected AOV equals expected revenue. The model structures forecasting by making assumptions explicit.

Teaching the model to teams

Organizational application:

Shared mental framework

When everyone understands the same model, conversations are more productive. “This affects conversion” has shared meaning.

Ownership clarity

Marketing might own traffic. Product might own conversion. Merchandising might own AOV. The model clarifies who owns what.

Cross-functional understanding

The model shows how functions connect. Marketing traffic feeds into product conversion feeds into merchandising AOV. Collaboration becomes obvious.

Simpler reporting

Reports structured around the model are easier to understand. Traffic, conversion, AOV, revenue. The core story before the details.

The daily practice

How to apply regularly:

Morning check-in

Yesterday’s traffic, conversion, AOV, revenue. Compare to typical. Note anything unusual. Two minutes, structured by the model.

Weekly review

Week-over-week comparison of each component. Which moved? Why might that be? Action needed? The model structures weekly analysis.

Monthly assessment

Trends in each component over time. Are we improving traffic? Conversion? AOV? Monthly patterns reveal strategic direction.

When anomalies appear

Walk through the model. Which component changed? One level deeper into that component. Systematic diagnosis versus scattered investigation.

Frequently asked questions

Isn’t this too simple?

Simple doesn’t mean simplistic. The model captures the fundamental structure of e-commerce revenue. Complexity can be added as needed, but the foundation is solid and useful.

What about other important metrics?

Other metrics matter. But most connect to the core equation. Customer acquisition cost relates to traffic. Cart abandonment relates to conversion. Product performance relates to AOV. The model is the trunk; other metrics are branches.

When should I use more sophisticated analysis?

When the simple model doesn’t explain what you’re seeing. If traffic, conversion, and AOV all look normal but revenue is off, something unusual is happening. Go deeper. But start simple.

Does this work for subscription businesses?

The model adapts. Traffic becomes acquisition. Conversion becomes subscription rate. AOV becomes subscription value. Retention adds another dimension. The core structure remains useful.