The Hierarchy of Metrics: How Prophit Engineers Diagnose Every Ecommerce Problem

When your brand's contribution margin misses target, where do you start looking? Most brands dive straight into campaign metrics. Prophit Engineers start at the top.

The Hierarchy of Metrics is the diagnostic framework our team uses to go from a business-level problem to a specific campaign-level action. Tony Chopp, VP of Paid Media, and Richard Gaffin walked through it live on the latest episode of the eCommerce Playbook podcast, complete with a real Statlas demo.

Here's the framework and how it works.

The Three Levels of the Hierarchy of Metrics

The hierarchy has three tiers, and you always start at the top.

Level 1: Business Metrics — Revenue, contribution margin, total ad spend, MER. These tell you whether you're hitting your targets. They don't tell you why.

Level 2: Customer Metrics — New customer revenue vs. returning. Paid vs. organic. This layer reveals where the gap is coming from.

Level 3: Channel Metrics — Google iROAS, Meta iROAS, campaign-level performance. This is where you take action, but only after the first two levels have given you context.

The mistake most brands make is starting at Level 3. They see a campaign with a 2.5 ROAS and start optimizing creative. But what if their iROAS target is 2.2? That campaign is actually over-delivering, which means the brand is underspending and leaving growth on the table.

How the Diagnosis Actually Works

Tony walked through a live example in Statlas. Here's what the diagnostic sequence looked like:

Step 1: Contribution margin is 7.5% behind target. Something needs to change.

Step 2: Ad spend is significantly below plan. But MER (overall efficiency) is 16% above target. That combination is a red flag — the brand is too efficient because it's underspending.

Step 3: Customer metrics show paid revenue and new orders are both low. The gap is on the acquisition side.

Step 4: Channel metrics reveal the source. Google Ads spend is 44% behind pace. But Google's iROAS is running at 6.19 against a target of 3.52 — nearly double.

That's not a performance win. That's uncaptured demand.

Step 5: Drill into the specific campaign. A standard shopping non-brand campaign is running at 9x iROAS against a 4.0 target. The fix: open up bids, increase budget, and let the campaign capture the volume it's clearly capable of producing.

Top to bottom. Contribution margin problem identified, diagnosed through customer metrics, traced to a specific Google campaign, and resolved with a budget and bid adjustment. That's the hierarchy at work.

Why Your ROAS Target Might Be Wrong

One of the biggest takeaways from the episode: most brands set ROAS targets based on gut feel.

Tony put it directly: "4 is not necessarily better than 3. If 3 is the right number based on your COGS and incrementality, targeting 4 means you're leaving volume on the table."

The right ROAS target comes from three inputs:

• Your contribution margin requirements (unit economics)
• Measured incrementality for that specific channel
• Customer lifetime value projections

Without those inputs, you're optimizing to a number that feels good but may be actively constraining your growth.

Over-Efficiency Is Underspend in Disguise

This is the pattern the hierarchy catches that most frameworks miss.

When your campaigns consistently exceed ROAS targets, it looks like winning. The dashboard is green. But the math tells a different story: you're only capturing the easiest conversions. The incremental customers who would drive real growth are going unreached.

In the live example, Google was running at nearly 2x its iROAS target. That's not a sign of great performance — it's a sign of constrained budgets. The brand was leaving profitable volume on the table because no one had connected the dots between business metrics (CM miss) and channel metrics (Google over-efficiency).

That connection is exactly what the hierarchy provides.

Good Data In, Good Decisions Out

Tony emphasized that the entire system depends on data quality. The hierarchy only works when:

• Cost data is clean — accurate COGS, delivery costs, and variable expenses feeding into contribution margin
• Audience segmentation is disciplined — campaigns structured to separate new customer acquisition from existing customer reactivation
• Incrementality is measured — geo holdout tests or benchmarks providing real incrementality factors by channel

Without that foundation, every optimization decision downstream is built on assumptions. With it, the path from problem to action is clear.

Frequently Asked Questions

Get This Level of Clarity for Your Brand

The Hierarchy of Metrics is the framework behind every decision Prophit Engineers make for 170+ ecommerce brands. If your brand is ready for daily operating clarity and a system that connects business outcomes to channel-level action, we should talk.

Talk to Us