4,084of 4,246 rows in scope
Overview

Homepage Redesign · Split Test

Raw traffic log of 4,246 sessions collected between Jun 1 – Jun 30, 2026. Every figure below is computed live from the dataset under the active filters.

Experiment running
Sessions in scope
4,084
4,246 raw rows
Unique users
4,084
2 experiment groups
Total conversions
614
Across all variants
Overall conversion rate
15.03%
Pooled across variants
Best performing variant
Variant B
16.23% conversion
Avg. session duration
2m 26s
Mean, outlier sensitive
Median session duration
1m 24s
Robust central tendency
Relative lift
+17.0%
Variant B vs A
Statistical significance
Significant
p = 0.035
Data quality score
91/100
99.61% complete
Experiment verdictSignificant

Variant B wins the split test

Variant B converted 17.0% better than Variant A. With a p-value of 0.0355, the sample provides clear evidence that this difference is real and not random noise.

Recommendation: Ship Variant B. At the observed 17.0% relative lift, rolling out the winning design should raise conversions on the same traffic volume.

Open full statistical analysis
Winning variant
Variant B
Conversion difference
+2.35% pts
Relative lift
+17.0%
Confidence
96.5%
p-value
0.0355
Sample size
4,084

Conversion rate by variant

Wilson 95% interval shown as the whisker

Cumulative conversion rate over time

Running rate per variant — watch for convergence before calling a winner

Generated findings

What the data is telling us

All insights
performance16.23%

Variant B is currently ahead

Variant B converts at 16.23% versus 13.88% for Variant A — a relative lift of 17.0% on the current sample.

statistics96.5%

Result is statistically significant (p = 0.0355)

Variant B converted 17.0% better than Variant A. With a p-value of 0.0355, the sample provides clear evidence that this difference is real and not random noise.

behavior0.042

Longer sessions convert substantially better

Sessions in the 600s+ band convert at 31.43% compared with 12.72% in the 0–30s band (point-biserial r = 0.042).