How AI Search Is Recommending Flight Bookings: Monthly Trends

Executive Summary

The Flight bookings category was quiet in August 2026: no brand recorded a meaningful change in valid recommendation coverage relative to July 2026, and the benchmark classifies every tracked brand as stable on that metric. Flightright is the current coverage leader at 1.8 percent, based on a single qualified recommendation within 55 qualified observations. AirHelp, the July leader at 1.9 percent, recorded 0.0 percent in August; both changes fall inside the range the benchmark treats as ordinary month-to-month variation for this metric.

Raw mention presence moved more than valid recommendation coverage this period. AirHelp's presence rate moved from 40.2 percent in July 2026 to 72.7 percent in August 2026 (up 32.5 points), while AirAdvisor's presence rate moved from 57.9 percent to 29.1 percent over the same period (down 28.8 points). AirHelp is now the most visible brand in the category by raw mention presence; AirAdvisor is second. Neither brand converted its presence into recommendation coverage in August.

Across the two-month series, the category is characterized by high visibility for a few brands and almost no recommendation outcomes for any of them. The benchmark identified 620 prompt-surface observations in July 2026 (476 unique questions) and 706 in August 2026 (531 unique questions). Of those, 157 prompts in July and 97 in August were relevant, yielding 107 and 55 qualified observations respectively. In both months, valid recommendation coverage was concentrated in a single brand at very low rates, with no brand recording a top-three or rank-one recommendation.

Each monthly run begins with 620 prompt-surface observations (476 unique questions) in July 2026 and 706 prompt-surface observations (531 unique questions) in August 2026 across the benchmark's defined AI/search surface universe. Of those, 414 in July and 428 in August mentioned a tracked brand or competitor; 157 in July and 97 in August were relevant, while 257 in July and 331 in August were irrelevant. The public metrics use the 107 qualified observations in July and the 55 qualified observations in August that survive both qualification stages.

Key Findings

Benchmark Context

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set.

The benchmark set contracted from 107 qualified observations in July 2026 to 55 in August 2026, even as the raw collection universe grew from 620 to 706 prompt-surface observations. The larger raw pool produced fewer relevant prompts in August, which is the primary reason the qualified set shrank. All six canonical surface families (ChatGPT, Copilot, Gemini, Perplexity, AI Overviews, and AI Mode) remained active in both months.

Benchmark-Level Metrics

The recommendation-shaped answer share moved from 0.9 percent in July 2026 to 3.6 percent in August 2026, while the valid recommendation shortlist share held nearly flat at 1.8 percent. The gap between these two measures means AI systems produced somewhat more recommendation-shaped answers in August, but the share of those answers that qualify as valid shortlists did not move with it.

AI Recommendation Trend

Recommendation Coverage by Brand

The category is effectively unclaimed. Only one brand holds any recommendation coverage in August 2026, and the prior leader holds none. No brand recorded a top-three placement or a rank-one recommendation in either month.

The benchmark classifies every brand as stable on this metric this month: Flightright's single qualified recommendation in August and AirHelp's loss of its two qualified recommendations from July both fall within the range of normal month-to-month variation for a metric this small.

What Changed This Month

AirHelp

AirHelp's valid recommendation coverage fell from 1.9 percent in July 2026 (2 of 107 qualified observations) to 0.0 percent in August 2026 (0 of 55). The benchmark classifies this as a stable change, within the normal range for this metric.

The larger movement was in visibility. AirHelp's raw mention presence rate moved from 40.2 percent in July 2026 to 72.7 percent in August 2026, up 32.5 points. In absolute terms, the brand appeared in 40 of 55 qualified observations in August, up from 43 of 107 in July.

The distinction the reader should notice: AirHelp appeared far more often across qualified observations in August while recording no qualified recommendations. Presence without recommendation is the defining feature of this brand's August result.

Highest-priority diagnostic: Which prompts mention AirHelp but stop short of recommending it, and which competitor appears in the shortlist when AirHelp is present but not chosen?

AirAdvisor

AirAdvisor's valid recommendation coverage held at 0.0 percent, unchanged from July 2026. The brand has not recorded a qualified recommendation in either month of the series.

The larger movement was in visibility. AirAdvisor's raw mention presence rate moved from 57.9 percent in July 2026 to 29.1 percent in August 2026, down 28.8 points. In absolute terms, the brand appeared in 16 of 55 qualified observations in August, down from 62 of 107 in July.

The distinction the reader should notice: AirAdvisor was the most visible brand in July 2026 but ceded that position to AirHelp in August, even as its recommendation coverage remained at zero in both months.

Highest-priority diagnostic: Which surfaces and prompt types account for the presence change, and does the reduced presence correspond to a specific evidence-source change?

Flightright

Flightright's valid recommendation coverage moved from 0.0 percent in July 2026 to 1.8 percent in August 2026, on a single qualified recommendation out of 55 observations. The benchmark classifies this as a stable change, within the normal range for this metric.

The brand also recorded one positive sentiment mention in August, its first of the series, alongside one neutral mention, giving it a net sentiment score of 1.0 on a base of 2 mentions.

The distinction the reader should notice: Flightright is now the coverage leader, but on a base of one recommendation. This reflects the current benchmark snapshot, not an established leadership position.

Highest-priority diagnostic: Which single prompt produced the recommendation, and is that prompt representative of a broader set of high-intent queries?

Skycop

Skycop's valid recommendation coverage held at 0.0 percent in both July 2026 and August 2026 — zero qualified recommendations out of 107 observations in July and zero out of 55 in August. Raw mention presence moved from 3.7 percent to 7.3 percent over the same period, meaning Skycop is appearing somewhat more often in qualified observations, but that increase has not produced a single recommendation to date.

In a market this small, one qualified recommendation was enough to take category leadership this month, and Skycop has not recorded one in either month of the series. AirHelp's presence reached 72.7 percent in August without converting into a recommendation either, so Skycop is not alone in failing to convert visibility — but Skycop's own presence, at 7.3 percent, is a fraction of AirHelp's, which means the gap sits on both sides of the funnel at once: less visibility than the category's most-seen brands, and no recommendation outcome at all. This is a measurable recommendation gap, not a marginal one, and it has persisted for two consecutive months while a competitor moved from zero to category leadership on a single observation.

Highest-priority diagnostic: Which prompts mention Skycop without producing a recommendation, and does the drop-off concentrate in comparison or pricing/value-stage queries — the stages where competing brands are currently capturing the shortlist?

Buyer-Intent Interpretation

In both July 2026 and August 2026, the qualified observations fell entirely into the Brand Recommendation cluster. The Pricing & Value and Multi-Brand Comparison clusters recorded no qualified observations in either month. The public benchmark therefore cannot yet answer questions about how AI systems describe pricing trade-offs, fee structures, or direct brand comparisons in this category.

Brand Opportunity Summary

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations (query, surface, recommendation outcome, rank, sentiment, and citations where exposed). Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns. Source presence is not automatically treated as proof of causation.

About This Benchmark

This report is part of the LLM Authority Index AI Market Discovery research program.

• AI Industry Market Discovery
• AI Industry Market Discovery Methodology
• AI Industry Market Discovery Metrics
• AI Industry Market Discovery Standards

Report-Specific Interpretation Notes

• Small-count movement: the one-point coverage change for Flightright and the two-point loss for AirHelp are single-recommendation movements on sharply different denominators (107 in July, 55 in August). They indicate direction, not established position.
• Qualified denominator vs. raw collection: the qualified set shrank from 107 to 55 observations even as the raw collection grew, which concentrates percentages on a smaller base in August.
• Directional analysis: month-over-month movement identifies changes worth investigating; it does not by itself establish the cause of those changes.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

The aggregate percentages raise questions that sit beneath the surface: which high-intent prompts are won, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. For a category where every brand sits at or near zero recommendation coverage, the question is not just which brand leads, but why the entire category is failing to convert visibility into recommendations.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy.

Request an AI visibility audit