Live · updated hourly

The GEO Index

The publishers and platforms AI search reaches for when it answers — every model aggregated, ranked by citation share, with weekly movement.

Rolling 4 weeks: 221,076 citations · 27,471 model responses · every named domain seen across ≥ 2 independent projects

All insights Claude Perplexity Gemini ChatGPT Grok DeepSeek Google AIO Google AI Mode ChatGPT Web Meta AI AI Source Bias IndexMarkets:United Kingdom United States Colombia Mexico Germany France

This week's movers

Citation-share change, week commencing 2026-08-10 vs 2026-08-17. Minimum 25 citations in both weeks and a 0.2pp move to qualify — domains newly added to monitoring don't count as movers.

Gaining

gov.co 0.1% → 0.5% ▲ +0.3pp

apple.com 0.1% → 0.3% ▲ +0.2pp

facebook.com 0.9% → 1.1% ▲ +0.2pp

tripadvisor.com 0.1% → 0.3% ▲ +0.2pp

expedia.com 0.1% → 0.2% ▲ +0.2pp

tiktok.com 0.1% → 0.4% ▲ +0.2pp

Declining

reddit.com 1.9% → 1.0% ▼ -0.9pp

Most-cited websites, all models

Top 50 domains by share of all citations in the rolling window. The trend column is the weekly share trajectory.

#	Domain	Share	Citations
1	youtube.com	1.9%	4,203
2	reddit.com	1.6%	3,437
3	facebook.com	1.1%	2,399
4	instagram.com	0.8%	1,787
5	google.com	0.7%	1,569
6	linkedin.com	0.4%	975
7	wikipedia.org	0.4%	791
8	booking.com	0.3%	738
9	tripadvisor.com	0.2%	547
10	tiktok.com	0.2%	455
11	gov.co	0.2%	455
12	yahoo.com	0.2%	444
13	expedia.com	0.2%	433
14	forbes.com	0.2%	424
15	apple.com	0.2%	418
16	admin.ch	0.2%	401
17	usnews.com	0.2%	400
18	runrepeat.com	0.2%	391
19	which.co.uk	0.2%	388
20	theguardian.com	0.2%	370
21	trustpilot.com	0.2%	366
22	gov.uk	0.2%	337
23	johnlewis.com	0.2%	333
24	whowhatwear.com	0.1%	314
25	kayak.com	0.1%	306
26	moneysavingexpert.com	0.1%	302
27	mordorintelligence.com	0.1%	288
28	infobae.com	0.1%	270
29	nih.gov	0.1%	266
30	motorpasion.com.mx	0.1%	260
31	hilton.com	0.1%	254
32	gulfnews.com	0.1%	239
33	independent.co.uk	0.1%	227
34	moneysupermarket.com	0.1%	224
35	amazon.com	0.1%	222
36	pages.dev	0.1%	222
37	nyulangone.org	0.1%	221
38	mercadolibre.com.mx	0.1%	221
39	fortunebusinessinsights.com	0.1%	210
40	wise.com	0.1%	210
41	hotels.com	0.1%	207
42	quora.com	0.1%	206
43	english-heritage.org.uk	0.1%	204
44	walmart.com	0.1%	196
45	uswitch.com	0.1%	196
46	nationaltrust.org.uk	0.1%	194
47	outdoorgearlab.com	0.1%	192
48	mountsinai.org	0.1%	192
49	nike.com	0.1%	190
50	larepublica.co	0.1%	187

The race for the top

Where the leading sources sat against each other, week by week. Rank space makes every crossing an event.

Figure 1The race for the top of the index

Rank within the top 6 sources, week by week. Only the biggest mover is coloured — the greys are the rest of the field.

Data table

Series	3 Aug	10 Aug	17 Aug	24 Aug
youtube.com	1	1	1	2
reddit.com	4	2	3	1
facebook.com	2	3	2	4
instagram.com	3	4	4	5
google.com	6	5	5	3
linkedin.com	5	6	6	6

3 places gained — reddit.com, #4 → #1 (24 Aug)

How stable are each model's sources?

Google AI Mode keeps the steadiest source list, retaining a median 30% of its top 50 week to week; Claude rewires fastest (11% retained).

Figure 2Top-50 sources retained, week to week

One brick per domain in each model’s top 50. Solid bricks are the median number retained from one week to the next; hollow ones churn out. A full wall would mean an identical source list every week.

Data table

Series	Retained of 50
Google AI Mode	30%
Perplexity	27%
Google AIO	25%
Gemini	21%
ChatGPT	18%
Claude	11%

How sourced are the answers?

Citations per response over the last 12 complete weeks — whether the models are leaning on the web more or less.

Figure 3Citations per answer, week by week

Each cell is one model in one week; hotter cells mean answers leaned on more sources. Single-hue ramp — the scale runs from the lowest to the highest cell shown.

Data table

Series	1 Jun	8 Jun	15 Jun	22 Jun	29 Jun	6 Jul	13 Jul	20 Jul	27 Jul	3 Aug	10 Aug	17 Aug
Perplexity	4.8	5.9	8	7.9	6.6	6.5	9.7	10.8	16	15.6	14.2	11.2
Gemini	10.7	8.2	9.8	8.2	9	9.4	9.1	10	8.4	9	7.3	6.2
Claude	12.5	10.3	9.8	11.7	13.4	10.3	12.7	9.6	12	9.9	8.7	10.4
Google AIO	3.8	3.1	3	3.9	5.1	6.7	8	7.9	6.7	8.5	8.2	6.5

Methodology

What's counted: every citation/source URL returned by citation-capable AI models across the rolling 4 ISO weeks, aggregated from every analysis run on BrandInsightAI. Citations deduplicate to one registrable domain per model response.
Quality gate: a domain must appear across ≥ 2 distinct projects to be named; shares are computed against ALL citations in scope, so the gate hides names, never distorts percentages.
Movers: week-over-week comparisons use the two most recent COMPLETE ISO weeks (2026-08-10 → 2026-08-17); the in-progress week is excluded because it refreshes hourly, and a domain must clear the citation floor in BOTH weeks — so a newly monitored category entering the dataset never shows up as a gainer.
Freshness: the snapshot behind this page recomputes every hour. All figures are measured, never estimated.