luvv to helpDiscover the Best Free Online Tools
Topic 6 of 8

Funnel Stage Metrics By Channel

Learn Funnel Stage Metrics By Channel for free with explanations, exercises, and a quick test (for Marketing Analyst).

Published: December 22, 2025 | Updated: December 22, 2025

Table of Contents

Why this matters

Marketing analysts are asked to answer questions like: Which channels actually create qualified pipeline? Where are we leaking leads? What should we scale or fix this week? Funnel stage metrics by channel let you identify bottlenecks across awareness, acquisition, qualification, and conversion so you can allocate budget and run focused experiments.

  • Who this is for: marketing analysts, growth marketers, performance marketers, founders who need clear channel ROI.
  • Prerequisites: comfort with spreadsheets; basic understanding of UTMs/channels; optional SQL for joining events to leads/opportunities.
  • Learning path: get UTM hygiene right β†’ define funnel stages β†’ compute per-stage rates and costs β†’ compare channels β†’ prioritize experiments and budget.

Concept explained simply

Think of each channel as a series of gates. At every gate, some people move forward and some drop off. Your job is to measure, per channel, how many enter each gate, how many pass, how long it takes, and what it costs.

Mental model: leaky pipes. Each channel has sections (stages). Fix the leakiest section first; then add more water (budget) to the best pipes.

Common funnel stages and how to map them
  • Awareness: impressions or reach (ads), email sends/opens, social views.
  • Traffic: clicks and website sessions/visits.
  • Lead: form fills, signups, trials (unique leads).
  • MQL: leads that meet fit/intent thresholds.
  • SQL: accepted by sales (meetings booked, qualified opportunities).
  • Opportunity: opportunity or deal created.
  • Customer: closed won.

Map events to channels using UTMs, referrers, or email campaign IDs. Decide your attribution rule before computing the metrics.

Key formulas (per channel)
  • Conversion rate stage A β†’ B = B count / A count
  • Drop-off rate A at next gate = 1 βˆ’ (B / A)
  • CPL (cost per lead) = Channel spend / Leads
  • CAC (cost per customer) = Channel spend / Customers
  • Qualified rate (lead β†’ MQL, or lead β†’ SQL) = Qualified / Leads
  • Lead-to-close rate = Customers / Leads
  • Stage velocity (days) = median(date B βˆ’ date A)
  • Channel share at stage X = Channel X count / Total count at stage X
Data checklist before you compute
  • Unique lead IDs and deduping across forms.
  • Consistent channel taxonomy (e.g., paid_search, paid_social, organic, email, referral, direct).
  • Attribution choice documented (first click for awareness; last non-direct for lower-funnel is a common pattern).
  • Time window alignment (same period for counts and spend).
  • Exclude internal traffic and test purchases.
  • Minimum volume guardrails (avoid big decisions on tiny samples).

Worked examples

Assume a one-month window and last non-direct attribution for leads and customers.

Example 1: Paid Search

  • Impressions 100,000 β†’ Clicks 2,800 β†’ Visits 2,600 β†’ Leads 260 β†’ MQL 156 β†’ Customers 18; Spend: $12,000
  • CTR = 2,800 / 100,000 = 2.8%
  • Visit β†’ Lead = 260 / 2,600 = 10%
  • Lead β†’ MQL = 156 / 260 = 60%
  • MQL β†’ Customer = 18 / 156 β‰ˆ 11.5%
  • CPL = $12,000 / 260 β‰ˆ $46.15; CAC = $12,000 / 18 β‰ˆ $666.67
  • Velocity: lead β†’ MQL median 2 days; MQL β†’ customer 40 days

Example 2: Paid Social

  • Impressions 200,000 β†’ Clicks 3,000 β†’ Visits 2,700 β†’ Leads 81 β†’ MQL 49 β†’ Customers 10; Spend: $9,000
  • CTR = 3,000 / 200,000 = 1.5%
  • Visit β†’ Lead = 81 / 2,700 = 3%
  • Lead β†’ MQL β‰ˆ 49 / 81 = 60.5%
  • MQL β†’ Customer = 10 / 49 β‰ˆ 20.4%
  • CPL β‰ˆ $9,000 / 81 β‰ˆ $111.11; CAC = $9,000 / 10 = $900
  • Velocity: lead β†’ MQL 3 days; MQL β†’ customer 45 days

Example 3: Email

  • Sent 20,000 β†’ Opens 4,000 β†’ Clicks 1,200 β†’ Visits 1,100 β†’ Leads 330 β†’ MQL 198 β†’ Customers 25; Cost: $500
  • Open rate = 4,000 / 20,000 = 20%
  • Visit β†’ Lead = 330 / 1,100 = 30%
  • Lead β†’ MQL = 198 / 330 = 60%
  • MQL β†’ Customer β‰ˆ 25 / 198 β‰ˆ 12.6%
  • CPL β‰ˆ $500 / 330 β‰ˆ $1.52; CAC = $500 / 25 = $20
  • Velocity: lead β†’ MQL 1 day; MQL β†’ customer 35 days

Interpretation: Paid social leaks at visit β†’ lead. Email converts visits to leads very well and is extremely cost-efficient here. Paid search is balanced but more costly than email per customer.

Step-by-step workflow

  1. Define stages and attribution. Write down exact events and the attribution rule you will use per stage.
  2. Map channels. Normalize UTMs and referrers into a clean channel dimension.
  3. Aggregate counts. For each channel and stage, count unique entities (visits, leads, customers) in the same time window.
  4. Join spend. Add ad spend or attributed costs per channel for the period.
  5. Compute metrics. Per-stage conversion rates, drop-offs, CPL, CAC, and stage velocities.
  6. Compare channels. Identify the weakest stage per channel and the best-in-class benchmark for each stage.
  7. Decide actions. Fix the weakest stage with targeted experiments; scale channels with strong CAC and healthy through-funnel rates.
Attribution choices and when to use them
  • First click: useful for awareness stage comparisons.
  • Last non-direct: useful for lead and customer credit in simple funnels.
  • Position-based or data-driven: better if you have long, multi-touch journeys.

Pick one and be consistent within your analysis window. You can run a sensitivity check later.

Exercises

Note: Anyone can do the exercises and the quick test. Only logged-in users will have progress saved.

Exercise 1 β€” Compute stage metrics and CAC

Use the dataset below (one-month window, last non-direct attribution):

Channel,Visits,Leads,MQL,Customers,Spend
Paid Search,5000,400,220,30,18000
Paid Social,8000,240,120,20,12000
Organic Search,6500,455,182,28,4000
  • Task A: For each channel, calculate Visit β†’ Lead, Lead β†’ MQL, and MQL β†’ Customer conversion rates (percent).
  • Task B: Compute CAC per channel.
  • Task C: Identify the biggest bottleneck for Paid Social and name one fix.
  • Task D: Which channel has the highest Visit β†’ Lead rate? Which has the lowest CAC?
Hints
  • Conversion rate = next stage / previous stage.
  • CAC = Spend / Customers.
  • Compare rates across channels to spot relative weaknesses.

Exercise 2 β€” Diagnose and recommend

Scenario: Your CFO asks why CAC went up despite similar customer counts. Paid Social spend increased while Visit β†’ Lead stayed at 3%. Email volume is flat but high quality. Provide 3 specific experiments for Paid Social to improve Visit β†’ Lead, and 2 guardrails to keep CAC healthy over the next month.

Hints
  • Targeting, creative-message-offer alignment, and landing-page friction are common levers.
  • Think about capping bids, shifting budget, and stopping underperformers quickly.
  • Self-check checklist:
    • You computed all rates correctly to one decimal place.
    • Your CAC math matches spend divided by customers.
    • Your recommendations directly address the weakest stage.
    • You included at least one budget guardrail or stop-loss rule.

Common mistakes and self-check

  • Mixing attribution models between stages. Self-check: confirm the same rule was used across channels in your table.
  • Double-counting leads across forms. Self-check: verify unique lead IDs and dedup rules.
  • Comparing channels with tiny samples. Self-check: mark cells with counts below your minimum threshold (e.g., under 20 events).
  • Ignoring time-to-convert differences. Self-check: review stage velocity; recent leads may not have had time to close.
  • Optimizing only for CPL while lead quality is poor. Self-check: track lead β†’ MQL and lead β†’ customer alongside CPL.

Practical projects

  • Build a funnel-by-channel dashboard in a spreadsheet: include counts, rates, drop-offs, CAC, and velocity with traffic-light coloring.
  • Run a one-week rapid test: choose one weak stage for one channel; implement 2 variants; report lift and confidence band.
  • Budget shift simulation: move 20% spend from the weakest CAC channel to the best; project impact using current stage rates.

Mini challenge

Given this snapshot for the last 2 weeks:

Channel,Visits,Leads,Customers,Spend
Display,7000,140,8,6000
Referral,1200,120,18,500
  • Which channel would you scale next week and why?
  • Which stage needs the most work for Display?
  • Name one quick experiment for Display and one for Referral.

Next steps

  • Recalculate your own funnel by channel for the last complete month using a single attribution rule.
  • Pick one channel and one weakest stage; queue 2 experiments that ship within a week.
  • Review learning path topics: channel tagging hygiene, attribution basics, LTV and payback, and cohort analysis to validate long-term quality.

Practice Exercises

2 exercises to complete

Instructions

Use the dataset (one-month window, last non-direct attribution):

Channel,Visits,Leads,MQL,Customers,Spend
Paid Search,5000,400,220,30,18000
Paid Social,8000,240,120,20,12000
Organic Search,6500,455,182,28,4000
  • Compute Visit β†’ Lead, Lead β†’ MQL, MQL β†’ Customer for each channel.
  • Compute CAC for each channel.
  • Identify the biggest bottleneck for Paid Social and propose one fix.
  • Which channel leads on Visit β†’ Lead? Which has the lowest CAC?
Expected Output
Percent rates for 3 stages per channel; CAC per channel in dollars; a short bottleneck diagnosis for Paid Social; channel picks for top Visit→Lead and lowest CAC.

Funnel Stage Metrics By Channel β€” Quick Test

Test your knowledge with 7 questions. Pass with 70% or higher.

7 questions70% to pass
Reach Frequency And CPM QualityBrand Versus Performance Split

Have questions about Funnel Stage Metrics By Channel?

AI Assistant

Ask questions about this tool