Categories compared
The same counts for every category, side by side, on one ChatGPT surface at a time.
| Plus | Autos → | Beauty & fitness → | Business → | Finance → | Home & garden → |
|---|---|---|---|---|---|
| How much it searches | |||||
| 100% 20 of 20 | 100% 20 of 20 | 100% 20 of 20 | 100% 20 of 20 | 100% 20 of 20 | |
| 1.1 22 in all | 1.1 22 in all | 1.3 25 in all | 1.4 27 in all | 1.4 28 in all | |
| 5.1 2.7 without the self-report | 5.2 2.3 without the self-report | 5.1 3.6 without the self-report | 4.7 3.9 without the self-report | 6.9 2.8 without the self-report | |
| 80% 16 of 20 | 55% 11 of 20 | 90% 18 of 20 | 85% 17 of 20 | 80% 16 of 20 | |
| Search journeys | |||||
| 70% 14 of 20 | 50% 10 of 20 | 65% 13 of 20 | 55% 11 of 20 | 45% 9 of 20 | |
| 10% 2 of 20 | 5% 1 of 20 | 25% 5 of 20 | 30% 6 of 20 | 35% 7 of 20 | |
| 0% 0 of 20 | 5% 1 of 20 0% without the self-report | 0% 0 of 20 | 0% 0 of 20 | 5% 1 of 20 0% without the self-report | |
| 20% 4 of 20 | 40% 8 of 20 45% without the self-report | 10% 2 of 20 | 15% 3 of 20 | 15% 3 of 20 20% without the self-report | |
| Query kinds | |||||
| 102 queries | 103 queries | 101 queries | 93 queries | 138 queries | |
| 18% 18 of 102 0% without the self-report | 10% 10 of 103 0% without the self-report | 20% 20 of 101 0% without the self-report | 16% 15 of 93 0% without the self-report | 36% 50 of 138 0% without the self-report | |
| What the web can move | |||||
| 50% 64 of 128 | 66% 41 of 62 | 38% 45 of 119 | 48% 56 of 117 | 42% 40 of 95 | |
| 20% 4 of 20 | 30% 6 of 20 25% without the self-report | 30% 6 of 20 | 50% 10 of 20 | 45% 9 of 20 | |
| 40% 48 of 119 | 30% 30 of 99 | 20% 25 of 124 | 31% 46 of 150 | 43% 57 of 132 | |
| 46% 56 of 122 49% without the self-report | 31% 19 of 62 66% without the self-report | 33% 39 of 119 38% without the self-report | 43% 50 of 116 51% without the self-report | 26% 23 of 88 44% without the self-report | |
| 25% 30 of 119 | 9% 9 of 99 | 67% 83 of 124 | 51% 77 of 150 | 20% 27 of 132 | |
| 5% 1 of 22 | 5% 1 of 22 | 12% 3 of 25 | 4% 1 of 27 | 0% 0 of 28 | |
| What reaches the answer | |||||
| 6.0 119 in all | 5.0 99 in all | 6.2 124 in all | 7.5 150 in all | 6.6 132 in all | |
| 8% 71 of 845 | 5% 42 of 869 | 6% 68 of 1182 | 1% 7 of 1114 | 3% 30 of 1127 | |
| 0% 0 of 119 | 1% 1 of 99 | 2% 2 of 124 | 0% 0 of 150 | 0% 0 of 132 | |
| 45% 9 of 20 | 15% 3 of 20 | 85% 17 of 20 | 55% 11 of 20 | 25% 5 of 20 | |
| 2.2 44 in all | 0.6 11 in all | 4.6 92 in all | 2.6 51 in all | 0.8 15 in all | |
| 61% 27 of 44 20 homepages | 91% 10 of 11 6 homepages | 57% 52 of 92 37 homepages | 37% 19 of 51 14 homepages | 80% 12 of 15 9 homepages | |
| 6.4 128 in all | 3.1 62 in all | 6.0 119 in all | 5.9 117 in all | 4.8 95 in all | |
| 0% 0 of 20 | 0% 0 of 20 | 0% 0 of 20 | 0% 0 of 20 | 0% 0 of 20 | |
Each column is the category page's “At a glance” for this surface; the category's prompt cards add up to it. Every prompt is a commercial search typed word for word as US searchers type it, so the numbers hold for prompts like these, not for ChatGPT in general. One run per prompt and surface, so small differences between categories can be chance. Query, journey and query-based web numbers use ChatGPT's own list of its searches (the self-report asked for at the end of every prompt, because the app hides the second-layer fan-outs it runs while reasoning) where it contains every query seen in the stream; the number without it is shown below where it differs.