AI Search Visibility for Trail Running Shoes: 2026 Brand Study

HOKA has the highest measured appearance rate in trail running shoe answers: 82.1% of 134 returned answers named the brand. Salomon followed at 76.1%, and Altra at 70.9%. These results cover the measured trail running shoe questions, not all consumer searches or demand.

The finding separates brand mention from source citation. HOKA was named often, but its owned domain had a 0% citation rate among 134 answers with retrieval evidence. Publisher sites had a different role. iRunFar appeared as a cited source in 74.6% of answers with retrieval evidence.

For category marketers, a strong name-recognition position does not by itself mean the brand's own pages are supplying the evidence behind an answer.

Which trail running shoe brands appear most often?

HOKA holds the top point estimate, followed by Salomon, Altra, Brooks, Saucony, La Sportiva, Topo Athletic, and New Balance. The table uses returned answers as its denominator. It measures whether an answer named a brand, not whether an engine cited the brand's website.

Trail running shoe appearance leaderboard

Appearance rate is the share of the 134 returned AI answers in which each brand was named.

The ranking is useful as a sample snapshot, but the leading positions are close. Every adjacent pair from HOKA through New Balance has overlapping 95% appearance intervals. The sample therefore does not establish that one of those adjacent brands is more visible than the other. It does establish that each appeared at the measured rate shown in the leaderboard.

The top brands are present across all four systems, but their rates and answer positions differ. Appearance rates below use 134 returned answers. Owned-domain citation rates use 134 answers with retrieval evidence.

Top brand profiles

Where do the AI engines disagree?

The four engines do not produce the same brand set. Perplexity returned 35 answers, ChatGPT 34, Claude 33, and Gemini 32. Their different models and retrieval tools can produce a gap that reflects system design rather than a brand's category-wide visibility.

Brand appearances by engine

Each cell uses that engine's own answer count as its denominator, so engines can be compared row by row.

Claude is the clearest separator among the top brands. It named HOKA in 19 of 33 answers, Salomon in 12, Saucony in 11, Altra in 8, La Sportiva and Topo Athletic in 5 each, Brooks and New Balance in 3 each, and Nike in 2. Perplexity named HOKA, Salomon, and Brooks in 34 of its 35 answers. ChatGPT named Altra in 32 of 34 answers, while Perplexity did so in 32 of 35.

Several brands are concentrated in one system. OutdoorGearLab appeared in 16 of Gemini's 32 answers and in none of the other engines' returned answers. RunRepeat appeared in 8 Gemini answers and none elsewhere. Altra Lone Peak and Fresh Foam X Hierro each appeared in 12 and 9 Perplexity answers respectively, with no appearances in the other three engines. NNormal appeared in 11 Perplexity answers, none in ChatGPT, 8 in Gemini, and 2 in Claude.

New Balance shows another large split: 29 of 35 Perplexity answers versus 4 of 32 Gemini answers and 3 of 33 Claude answers. Merrell, Inc appeared in 23 of 34 ChatGPT answers but 4 of 32 Gemini answers and 1 of 33 Claude answers. The North Face, Inc appeared in 16 of 35 Perplexity answers and none of Claude's 33 answers. These patterns mean a single-engine audit can conceal material variation.

Which brands lead each trail running shoe question type?

The demand-cluster view separates general trail running shoe questions from ultramarathon and wide-foot questions. It is the appropriate view for deciding where a brand's appearance is concentrated, rather than treating the overall leaderboard as a product-fit recommendation.

Demand-cluster appearances

The same answers regrouped by the demand cluster each prompt was sampled for; cluster rates use that cluster's own answer count.

Read each cluster as a measured question set. A brand can have a strong overall appearance rate while its visibility is concentrated in only one of the covered question types. The overall ranking does not establish product performance, fit, or suitability for an individual runner.

Which sites supply citations in trail running answers?

Publisher sites supply a large share of the retrieval evidence. iRunFar was cited in 100 answers, or 74.6% of the 134 answers with retrieval evidence. OutdoorGearLab followed with 82 citations and RunRepeat with 77.

Owned-domain citation rates

Share of the 134 answers with retrieval evidence that cited a URL on the brand's own website.

Among citation-eligible entities, OutdoorGearLab had a 61.2% owned-domain citation rate, RunRepeat had 57.5%, and REI Co-op had 47%. Altra had 2.2%, Nike had 1.5%, and Salomon, La Sportiva, Merrell, Inc, Inov-8, NNormal, and Prodigio each had 0.7%. HOKA, Brooks, Saucony, Topo Athletic, New Balance, The North Face, Inc, On, Adidas, ASICS, Dynafit, Cascadia, and Fresh Foam X Hierro had 0%.

Cited publisher domains

Publishers cited as evidence, counted once per answer rather than once per link.

The source chart shows a publisher-heavy evidence mix. Trail and Kale appeared in 69 answers with retrieval evidence, REI Co-op in 63, and Runner's World in 58. Reddit appeared in 46. These figures measure citation occurrence in answers with retrieval evidence, not traffic, readership, product quality, or editorial endorsement.

What does this mean for trail running shoe brands?

Interpretation: brands need to track two separate outcomes. Brand appearances show whether a name enters an answer. Owned-domain citations show whether the brand's own site is part of the evidence used in answers with retrieval evidence.

HOKA illustrates the separation. It has the highest appearance rate but no owned-domain citations in this sample. OutdoorGearLab and RunRepeat show the reverse pattern: low appearance rates as named entities, paired with high citation rates for their domains. A brand team can use that contrast to audit product pages, fit guidance, and material that independent publishers cite.

Interpretation: engine-specific review matters. Perplexity, ChatGPT, Gemini, and Claude produced different patterns for New Balance, Merrell, Inc, Brooks, OutdoorGearLab, RunRepeat, NNormal, and several product names. Measure the engines where the brand expects to be found, then compare brand mentions with the cited domains behind those answers.

How was the trail running shoe study measured?

This flagship study measured 24 stratified prompts three times per engine. It used Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), and Claude (Anthropic). The study attempted 288 answers and received 134 returned answers.

Study method

How this study was measured

• Sampling frame fixed first: 24 prompts allocated across the demand clusters before anything was asked.
• Same questions put to every engine: Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), Claude (Anthropic), 3 trial(s) each.
• 134 answers returned of 288 attempted, 134 with retrieval evidence.
• Product lines rolled up to parent brands using a registry fixed before scoring.
• Appearance rate reported with a 95% Wilson interval; citation metrics counted only on answers with evidence.

Appearance rates use returned answers as the denominator. Citation and source rates use answers with retrieval evidence as the denominator. The study recorded 134 answers with retrieval evidence, a 46.5% retrieval rate, and 154 answers without retrieval evidence. Results are limited to these trail running shoe prompts on the measurement date. The study cannot measure consumer demand, sales, or product quality.

Methodology appendix

This appendix is generated directly from the measurement record for this study. It is included so any brand named here can reproduce or challenge the result.

What was measured

• Category: Trail Running Shoes
• Measurement date: 2026-08-27
• Depth: flagship — 24 prompts, 3 trial(s) per prompt per engine, 4 engines
• Answers attempted: 288
• Answers returned (appearance denominator): 134
• Answers with retrieval evidence (citation denominator): 134 (46.5% of attempted)
• Answers that failed outright: 153
• Analysis version: study-methodology-v2 · entity registry: entity-registry-2026-08-27

Claim limit: Flagship depth: 24 stratified prompts measured three times per engine, so both between-prompt and repeat-generation variation are observable.

How the prompts were sampled

Fixed sampling frame: client-named subcategories are allocated a guaranteed cell each, then a documented set of generic commercial demand clusters (general best, primary use case, specific need, quality tier, price/value, beginner, comparison, durability, specialist, emerging, retail channel, trust) receives the remaining allocation round-robin. The prompt writer supplies wording inside these cells only and does not choose the category's demand structure.

Results are not strictly comparable across engines: each engine uses a different model, a different retrieval tool and a different answer format, so a gap between two engines can reflect system design as much as brand visibility.

How each metric is defined

• Appearance rate — the share of returned answers that named the brand. Measured on every answer the engine returned, whether or not it searched the web, so an engine's retrieval behaviour does not silently change the denominator.
• Macro appearance rate — the mean of the per-prompt appearance rates, so each prompt counts once regardless of how many engines answered it.
• 95% interval — Wilson score interval on the appearance rate. Where two brands' intervals overlap, this study does not support a claim that one is more visible than the other.
• Owned-domain citation rate — the share of answers with retrieval evidence that cited a URL on the brand's own domain. This is a narrow measure of the brand's own site being used as a source; it is not a measure of total mentions.
• Share of all citations — the brand's own domain as a share of all attributed publisher citations in the study.
• Not measured — the brand's official domain could not be established, so no citation result exists. This is never reported as 0%.
• 0% — the metric could be evaluated and the event did not occur.
• Average position — mean position of the brand's first mention in the answer's ordered list of named brands.
• Cross-engine consistency — the share of engines that named the brand at least once.
• Repeat consistency — the share of repeated prompt/engine cells where the brand appeared in every trial. Lower values mean the engine's answer is unstable rather than the brand being absent.

Evidence rules

• An answer counts as having retrieval evidence only when the engine returned a retrieval receipt (search annotations or grounding records). A URL the model merely typed into its prose is not treated as evidence that it searched.
• Sources are attributed to the publisher behind each retrieval receipt. Intermediary and redirect hosts are never reported as publishers.
• Publisher identity could not be established for 0 retrieval receipt(s); those are disclosed here rather than attributed or dropped.
• Source counts are answer-level: a publisher counts once per answer in which it was cited, not once per URL.

Entity handling

Product lines and model names roll up to their parent brand for brand-level metrics, using a registry fixed before scoring. Nothing is merged on the basis of this study's own counts alone.

• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• HOKA Bondi 9 counted under HOKA Bondi (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Lone Peak counted under Altra (product_line)
• Altra Lone Peak 8/9 counted under Altra Lone Peak (model)
• Kjerag 2.0 counted under Kjerag (model)
• Peregrine 16 counted under Peregrine (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Norda 005 counted under Norda (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Peregrine 16 counted under Peregrine (model)
• Peregrine 16 counted under Peregrine (model)
• Lone Peak 9 counted under Altra (model)
• Peregrine 16 counted under Peregrine (model)
• Lone Peak 9 counted under Altra (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)
• Peregrine 16 counted under Peregrine (model)
• Lone Peak 9 counted under Altra (model)
• Peregrine 15 counted under Peregrine (model)
• Vista counted under UPPAbaby (product_line)
• Lone Peak 9 counted under Altra (model)
• Altra Lone Peak 9 counted under Altra Lone Peak (model)

Held for human classification before the next study: La Sportiva, Saucony Peregrine 16, Saucony Xodus Ultra 4, HOKA Speedgoat 7, Nike ACG Pegasus Trail, Nike Pegasus Trail 5, Brooks Cascadia 19, Brooks Caldera 7, Altra Timp 6, La Sportiva Prodigio Pro, La Sportiva Bushido III, Topo Athletic Ultraventure 4, Topo Athletic Traverse, Salomon Speedcross 6, Salomon Pulsar, Salomon Aero Glide 4 GRVL, Adidas Terrex Agravic 4, Adidas Terrex Agravic Speed Ultra 2, Salomon Ultra Glide 4, Inov-8, Switchback Travel, La Sportiva S.p.A, Outside Online, Tom’s Guide, Re-rank these by a specific priority, La Sportiva Prodigio 2, Altra Experience Wild 3, Brooks Cascadia 20, The North Face Vectiv Enduris 4, The North Face Altamesa 500 V2, Adidas Agravic Trail Ultra 2, Drop, Trail Scrambling, Drop Alternative, Reputation, REI Co-op Expert Advice, La Sportiva Akasha II, Brooks Cascadia Elite, Kailas, Hoka Speedgoat 6.

Limits of these findings

• This is a point-in-time measurement of what the named systems returned for these exact prompts on 2026-08-27. Engine behaviour changes without notice.
• The prompt set is a stratified sample of category demand, not a census. Results describe these prompts and should not be read as representative of all searches in the category.
• Where intervals overlap, differences between brands are not statistically distinguishable at this sample size.
• A brand not listed here was not measured as absent — it simply was not named in these answers.
• Appearance is a visibility measure, not an endorsement, a quality judgement or a market-share estimate.
• 1 of 25 ranked brand(s) had no official domain established before scoring, so they carry appearance results only and are excluded from citation metrics and source tables rather than being reported as zero.