Final images live 29 Jun 2026 · data through 19 Jul 2026
Verdict after 18 days of clean data
It is underperforming whether the new photography beats the originals.
Against the same dates last year, the 8 updated rugs' click-through rate is
-0% while the unchanged rugs are
+17% over the exact same window.
Recommendation: revert Cream (the highest-volume rug and the only one with a clean individual read) to its original image. The other seven are too low-volume to judge one by one, but the combined signal points the same way.
Year over year: same dates, seasonality removed
2 Jul–19 Jul this year vs the identical dates in 2025. Because both years share the same summer conditions, any gap between the two groups is down to the images, not the weather.
Updated rugs CTR vs last year
1.15% → 1.15%
-0% YoY
Unchanged rugs CTR vs last year
0.74% → 0.86%
+17% YoY
Net gap vs unchanged rugs
-0.13pp
updated minus unchanged, YoY
Updated rugs impressions
22,011 → 21,409
-3% YoY (unchanged +61%)
Weekly click-through rate: test vs control, this year vs last
Item-level Google Shopping CTR by week since the final images went live. Dashed lines are the same rugs over the same weeks last year.
Test rugs (new images)Control rugs (unchanged)Same weeks last year (dashed)
Per-rug detail
Current window (2 Jul–19 Jul) vs the same dates last year. Only Cream has enough volume for a reliable individual read; the rest are directional.
Rug
Impressions
CTR now
Impr last year
CTR last year
YoY
Conv value
Cream
15,062
1.45%
17,649
1.35%
+7%
£1,685
Taupe
3,353
0.57%
2,614
0.31%
+85%
£110
Grey low volume
415
0.00%
139
0.00%
n/a
£0
Caramel low volume
597
0.84%
487
0.00%
n/a
£0
Dark Grey low volume
311
0.96%
63
1.59%
-39%
£0
Champagne low volume
464
0.00%
203
1.48%
-100%
£0
Vole low volume
694
0.00%
847
0.35%
-100%
£0
Chocolate low volume
513
0.39%
9
0.00%
n/a
£0
How this test is read
What changed: the Google Shopping images for 8 single sheepskin rugs were replaced with new studio photography (rug draped over a chair). First versions went live 11 Jun 2026; the final colour-corrected versions replaced them on 29 Jun 2026, and the test is read from 2 Jul onward (72h buffer for Google's image re-review) so only the final photography is measured. Every other sheepskin rug listing (Oyster and White singles, doubles, larges, Tibetans; 204 items) kept its original image and acts as the control group.
Why year-over-year: sheepskin demand is seasonal and this summer has been hot. Comparing the identical calendar dates in 2025 removes seasonality entirely, so every number in this report is a clean like-for-like read.
Click-through rate is the primary metric: the image is what a shopper sees in Google before clicking, so CTR is the cleanest measure of whether the new photography earns attention. Conversion numbers at this volume are directional only (9 conversions on the updated rugs in the current window).
Impression shifts matter too: Google shows listings more when they earn clicks, so a sustained CTR change tends to compound into impression share.
Caramel has almost no traffic in either year, so it contributes nothing to the read either way.