Skip to content
Marketing Analytics LMS

Case 7: Finding Campaign Winners with Multi-Dimensional Filtering

Case 7: Finding Campaign Winners with Multi-Dimensional Filtering

Section titled “Case 7: Finding Campaign Winners with Multi-Dimensional Filtering”

Starting Point

Section titled “Starting Point”

Discovered that /medication-tirzepatide-discount-01 landing page performs at 5.80%

Action

Section titled “Action”

Apply additional filters to find which campaigns/traffic drive the best performance

Filters Applied

Section titled “Filters Applied”
  1. Current URL contains discount-01 (focusing on the landing page)
  2. AND affid equals 1000 (specific affiliate/traffic source)
  3. AND campaign_id equals 120236311971890504 (specific Facebook campaign)

Results with Multi-Dimensional Filtering

Section titled “Results with Multi-Dimensional Filtering”
MetricValue
Step 1: Landing page views423 persons (100%)
Step 2: Achieve screen viewed187 persons (44.21%) 🎯
Drop-off236 persons (55.79%)
Time to convert11m 43s

Performance Comparison

Section titled “Performance Comparison”
SegmentStep 1→2 ConversionPerformance vs Baseline
Overall baseline~9.88%1x (baseline)
Discount-01 landing page only~5.80%0.6x
Discount-01 + affid=1000 + campaign 1202363…44.21%4.5x 🚀

Key Insights

Section titled “Key Insights”

🏆 Winner Identified

Section titled “🏆 Winner Identified”

Specific campaign + affid + landing page combination converts at 44.21%

  • This is 4.5x better than the baseline (~10%)
  • This is 7.6x better than overall average (5.8% for discount-01 alone)
  • Only 55.79% drop-off instead of 90%+

What This Tells You

Section titled “What This Tells You”

  1. Not all traffic to discount-01 page is equal:

    • Without filtering: 5.80% conversion
    • With campaign filters: 44.21% conversion
    • The landing page isn’t the problem - the traffic quality is

  2. Campaign 120236311971890504 is a winner:

    • This specific Facebook campaign drives high-quality traffic
    • Users from this campaign are much more engaged/qualified
    • Likely because of better targeting, creative, or audience match

  3. Affid 1000 indicates successful traffic source:

    • This affiliate/channel ID is bringing quality traffic
    • Should investigate what affid=1000 represents
    • Consider scaling traffic from this source

  4. The combination matters:

    • Landing page alone: 5.80%
    • Landing page + right traffic source + right campaign: 44.21%
    • Success comes from the full stack: creative → targeting → landing page

Actionable Next Steps

Section titled “Actionable Next Steps”

Immediate Actions

Section titled “Immediate Actions”

  1. Scale winners:

    • Increase budget for campaign 120236311971890504
    • Drive more traffic from affid=1000
    • Replicate this campaign’s targeting/creative to other campaigns

  2. Investigate what makes this campaign different:

    • What’s the ad creative?
    • What’s the audience targeting?
    • What’s the messaging/value prop?
    • What UTM terms are being used?

  3. Test replication:

    • Create similar campaigns with same targeting
    • Test the winning creative on other affids
    • Apply learnings to underperforming campaigns

  4. Audit other traffic to discount-01:

    • If overall page is 5.80% but this segment is 44.21%, what’s dragging down the average?
    • Find the low-performing campaigns/affids driving to this page
    • Pause or optimize those campaigns

Further Investigation

Section titled “Further Investigation”

  1. Multi-dimensional winner analysis:

    • Test other combinations: Does affid=1000 work well with other landing pages?
    • Does campaign 1202363… perform well with different affids?
    • Find all winning combinations, not just this one

  2. Cohort through full funnel:

    • Does this 44.21% first-step conversion hold through the entire funnel?
    • Or do these users drop off later?
    • Calculate LTV/conversion to final goal

The Power of Multi-Dimensional Analysis

Section titled “The Power of Multi-Dimensional Analysis”

Single Dimension (landing page only):

  • “discount-01 page converts at 5.80%”
  • Action: Maybe optimize the page or test variations
  • Limited insight

Multi-Dimensional (landing page + affid + campaign):

  • “discount-01 page with affid=1000 and campaign 1202363… converts at 44.21%”
  • Action: Scale this exact combination, investigate what’s working, replicate to other campaigns
  • Actionable, specific, scalable insight

This is how you find winners.

Thinking Flow Demonstrated

Section titled “Thinking Flow Demonstrated”

✓ Step 1: See aggregate (overall ~10% conversion) ✓ Step 5: Segment by dimension #1 (landing page) → Find discount-01 at 5.80% ✓ Step 5 (repeated): Segment by dimension #2 (affid) → Narrow to affid=1000 ✓ Step 5 (repeated): Segment by dimension #3 (campaign_id) → Find 44.21% winner ✓ Step 6: Pattern identified → This specific combo is a massive winner ✓ Step 7: Hypothesis → This campaign has better targeting/creative/audience match ✓ Next: Validate by checking if pattern holds, then scale

Principles Applied

Section titled “Principles Applied”

Tool Features Used

Section titled “Tool Features Used”
  • Multi-condition filtering with AND logic
  • Property-based filters (Current URL, affid, campaign_id)
  • Operators: “contains”, “equals”
  • Real-time recalculation of funnel metrics
  • Person-level data (423 persons, not abstract numbers)
  • Time to convert tracking
  • Visual comparison (100% bar → 44% bar clearly shows strong conversion)
Section titled “Navigation”