Merchant dashboard
What a brand would see: how often shoppers get a confident answer, why photos get refused, and what each session costs.
Aggregate numbers only. No photos, names or addresses are stored, and a random session id is kept only in memory. These numbers describe how the tool behaves. They make no claim about sales or returns.
The cost per session is shown in units. Enter your price per unit to see it in rupees.
Live numbers
Loading the live numbers...
Demo data
Demo data: made up to show the layout. Not from real shoppers. It is here only to show what the dashboard looks like with traffic.
232Sessions that took a photo
51%Reached a confident verdict (118 of 232)
64%Retake-then-success (41 of 64 refused shoppers)
21.7Units per paid session
3652Units spent in total (168 paid sessions)
Verdicts shown
| Verdict | Count | Share of verdicts |
|---|---|---|
| Good match | 61 | 38% |
| Close match | 52 | 33% |
| Poor match | 29 | 18% |
| Can't tell, retake | 18 | 11% |
| Confidence | Count |
|---|---|
| High | 70 |
| Medium | 55 |
| Low | 35 |
Why photos were refused before any paid analysis
| Reason | Photos |
|---|---|
| Face too small | 31 |
| Too dark | 26 |
| Head turned sideways | 14 |
| Light from one side only | 11 |
| Blurred by movement | 9 |
| Shadows on the face | 6 |
| Face cut off by the edge | 5 |
176 photos passed the free on-device check and 96 were refused. 29 of 197 analyses were repeats served from the session cache at no cost.
| Cause | Count |
|---|---|
| Timed out | 2 |
| Service error | 1 |