Does the brand with the most citations rank first?

Across 103,679 eligible AI answers, the brand attached to more distinct cited pages than every other ranked brand was the top recommendation in 42,552 answers, or 41.0421%. It ranked lower in 61,127 answers, or 58.9579%. We call that brand the cited-page leader.

Semrush's ghost-citation study established that citation and brand mention are different outcomes. This result refines that boundary: citation volume and recommendation rank also answer different questions inside one answer. A source audit should record both.

The broader cut contained 145,062 answers where at least one ranked brand had a cited page. The top-ranked brand was tied or clear for the most cited pages in 71,839 answers, or 49.5230%. Another 41,383 answers had multiple brands tied for the most cited pages, so the main result excludes them.

The conclusion does not depend on assigning tied answers to a winner. Even the inclusive definition leaves the top recommendation below 50%.

On ChatGPT Search, the cited-page leader ranked first in 14,962 of 33,639 eligible answers, or 44.4781%. On Google AI Mode, it ranked first in 27,590 of 70,040 answers, or 39.3918%. The gap was 5.0863 percentage points.

A single combined rate hides a measurable engine difference. Compare citation support with rank on each engine instead of treating one engine as a proxy for the other.

The cited-page leader ranked second in 18,167 of 103,679 answers, or 17.5224%, and third in 13,665, or 13.1801%. Together, second and third place accounted for 31,832 answers, or 30.7025%. Another 29,295 cited-page leaders ranked fourth or lower.

Evidence can cluster around a close alternative rather than the winner. Audit the whole ranked shortlist so a well-supported competitor at position two or three is not missed.

The shortlist is the set of brands one answer ranked. When an answer ranked two or three brands, the cited-page leader ranked first in 16,715 of 30,778 answers, or 54.3083%. The rate fell to 38.3129% across 52,588 answers with four to six brands and 28.0067% across 20,313 answers with seven or more brands.

Citation count becomes a weaker shortcut for first place as the answer considers more brands. Record shortlist size when comparing citation support across prompts.

With a one-page lead over the next brand, the cited-page leader ranked first in 16,096 of 41,540 answers, or 38.7482%. The rate was 41.7972% for a two- to three-page lead and 43.3811% for a lead of four or more pages. The largest group was only 4.6329 percentage points above the one-page group.

A wider citation lead does not make citation count equivalent to recommendation rank. This result is descriptive and does not identify why an engine placed a brand first.

Among 21 industries with at least 1,000 eligible answers, Sports had the highest rate at 600 of 1,144, or 52.4476%. Sales and Marketing had the lowest at 545 of 1,622, or 33.6005%. Software reached 42.6919%, and Artificial Intelligence reached 41.9155%.

Use an industry baseline before treating a brand's rate as unusual. The spread identifies where citation support and recommendation rank differ most. It does not show that industry caused the difference.

Among brands that were the clear cited-page leader in at least 100 answers, AiDash ranked first in 351 of 401, or 87.5312%. Bright Data reached 85.1393%, while Wise reached 82.1839%. At the other end, Notion reached 17.7083%, Ahrefs reached 11.0345%, and Monday.com reached 7.9787%.

These rates identify brand-specific audit priorities, not brand quality. Prompt mix differs by brand, so Compare each brand with its own prompts and recurring competitors before comparing rates across brands.

The window contained 252,422 answers with ranked brands. Of those, 227,755 ranked at least two brands and 201,784 had one clear first recommendation. The method then excluded 56,722 answers with no cited ranked brand and 41,383 answers with tied cited-page leaders, leaving 103,679 answers with one clear leader.

Each cited page counted once per brand in an answer. Different names for the same brand were merged, and displayed names were checked so that product names were not shown as brands. Requiring at least two cited pages for the leader produced a 40.1258% rate. Requiring a lead of at least two pages produced 42.5755%. The study describes an observed cut, not recommendation quality, factual support, causation, user demand, or search rank.

The cited-page leader ranked first in 41.0421% of 103,679 eligible answers. The rate was 44.4781% on ChatGPT Search and 39.3918% on Google AI Mode. It fell from 54.3083% on two- or three-brand answers to 28.0067% on answers with seven or more brands.

For each priority prompt, record the first recommendation, the full ranked shortlist, and the cited pages attached to every brand. Compare results by engine and shortlist size. Investigate recurring competitors with more attached evidence even when they rank lower. Rerun the fixed-window study next quarter before treating these observed rates as permanent.