ASICS Corporation led the measured running shoe answers, appearing in 82.9% of 240 returned answers. That appearance rate uses returned answers as its denominator. RunRepeat had a different result: its domain was cited in 70.3% of 239 answers with retrieval evidence.
The split matters for footwear marketers. A brand can be named often without its own site being cited often. A publisher can be cited often while appearing less often as a selection option. These results cover running shoes, running shoes for marathon, and running shoes for flat feet in this measured prompt set.
Study at a glance
running shoes — measured 2026-08-27. Based on 240 answers returned out of 288 attempted across Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), Claude (Anthropic). Appearance figures use answers returned; citation figures use the 239 answers with retrieval evidence.
- AI answers analyzed — 24 prompts × 3 trials × 4 engines: 240
- Top brand: ASICS Corporation — named in 199 of 240 answers: 82.9%
- Engine spread for ASICS Corporation — highest vs lowest engine appearance rate: 95.1% vs 68.3%
- Answers with retrieval evidence — 239 of 288 attempted: 83%
The study-at-a-glance results separate brand appearance from source citation. Those are different measures with different denominators.
Which running shoe brands lead AI answers?
ASICS Corporation ranks first by appearance rate, followed by New Balance, Brooks, Nike, Inc, Hoka, and Saucony. ASICS Corporation was named in 199 returned answers. New Balance was named in 175, while Brooks and Nike, Inc each appeared in 172.
Appearance is not a product recommendation. It records whether an engine named a brand in an answer. Average first-mention position adds context: ASICS Corporation appeared at 3.1 on average, while the other leading brands often appeared later in the answer.
AI visibility leaderboard
| # | Brand | Named in | Appearance rate | 95% interval | Avg position |
|---|---|---|---|---|---|
| 1 | ASICS Corporation | 199/240 | 82.9% | 77.6%–87.2% | 3.1 |
| 2 | New Balance | 175/240 | 72.9% | 67%–78.1% | 5.1 |
| 3 | Brooks | 172/240 | 71.7% | 65.7%–77% | 4.1 |
| 4 | Nike, Inc | 172/240 | 71.7% | 65.7%–77% | 5.2 |
| 5 | Hoka | 168/240 | 70% | 63.9%–75.4% | 5.2 |
| 6 | Saucony | 164/240 | 68.3% | 62.2%–73.9% | 5.4 |
| 7 | On | 113/240 | 47.1% | 40.9%–53.4% | 9.6 |
| 8 | Puma | 96/240 | 40% | 34%–46.3% | 7.8 |
| 9 | Mizuno Corporation | 79/240 | 32.9% | 27.3%–39.1% | 9.7 |
| 10 | Adidas | 63/240 | 26.3% | 21.1%–32.2% | 4.9 |
| 11 | Altra | 60/240 | 25% | 19.9%–30.8% | 7.3 |
| 12 | RunRepeat | 31/240 | 12.9% | 9.3%–17.8% | 2.9 |
| 13 | Topo Athletic | 30/240 | 12.5% | 8.9%–17.3% | 7.4 |
| 14 | Sources | 19/240 | 7.9% | 5.1%–12% | 13.3 |
| 15 | Fleet Feet | 16/240 | 6.7% | 4.1%–10.6% | 3.6 |
ASICS Corporation leads at 82.9% (199 of 240 answers, 95% interval 77.6%–87.2%). Where two intervals overlap, this sample does not establish that one brand is more visible than the other.
The rank order needs restraint. ASICS Corporation's 95% interval of 77.6% to 87.2% overlaps New Balance's interval of 67% to 78.1%. New Balance, Brooks, Nike, Inc, Hoka, and Saucony also have overlapping intervals across much of that group. Within this sample, those differences are within the margin of the sample where intervals overlap. Brooks and Nike, Inc have the same 71.7% appearance rate and the same 65.7% to 77% interval.
How did the leading brands appear in AI answers?
ASICS Corporation — 82.9%
ASICS Corporation ranks first, with an 82.9% appearance rate and a 77.6% to 87.2% interval among 240 returned answers. Perplexity named ASICS Corporation in 58 of 61 answers, or 95.1%, and ChatGPT named it in 56 of 59, or 94.9%. Gemini named it in 41 of 60 answers, or 68.3%, while Claude named it in 44 of 60, or 73.3%. Its average first-mention position was 3.1. ASICS Corporation's domain was cited in 1.7% of 239 answers with retrieval evidence.
New Balance — 72.9%
New Balance ranks second, with a 72.9% appearance rate and a 67% to 78.1% interval among returned answers. Perplexity named New Balance in 58 of 61 answers, or 95.1%, and ChatGPT named it in 56 of 59, or 94.9%. Gemini named it in 42 of 60, or 70%, but Claude named it in 19 of 60, or 31.7%. New Balance appeared at an average first-mention position of 5.1. Its domain was cited in 0.4% of answers with retrieval evidence.
Brooks — 71.7%
Brooks ranks third, with a 71.7% appearance rate and a 65.7% to 77% interval among returned answers. Perplexity named Brooks in 58 of 61 answers, or 95.1%, while ChatGPT named it in 57 of 59, or 96.6%. Gemini named Brooks in 37 of 60 answers, or 61.7%, and Claude named it in 20 of 60, or 33.3%. Brooks' average first-mention position was 4.1. Its domain was cited in 8.8% of answers with retrieval evidence.
Nike, Inc — 71.7%
Nike, Inc ranks fourth, with a 71.7% appearance rate and a 65.7% to 77% interval among returned answers. Perplexity named Nike, Inc in 54 of 61 answers, or 88.5%, and ChatGPT named it in 56 of 59, or 94.9%. Gemini named Nike, Inc in 39 of 60 answers, or 65%, while Claude named it in 23 of 60, or 38.3%. Nike, Inc appeared at an average first-mention position of 5.2. Its domain was cited in 1.3% of answers with retrieval evidence.
Hoka — 70%
Hoka ranks fifth, with a 70% appearance rate and a 63.9% to 75.4% interval among returned answers. Perplexity named Hoka in 55 of 61 answers, or 90.2%, and ChatGPT named it in 55 of 59, or 93.2%. Gemini named Hoka in 35 of 60 answers, or 58.3%, while Claude named it in 23 of 60, or 38.3%. Hoka's average first-mention position was 5.2. Its domain was cited in 0% of answers with retrieval evidence.
Saucony — 68.3%
Saucony ranks sixth, with a 68.3% appearance rate and a 62.2% to 73.9% interval among returned answers. Perplexity named Saucony in 58 of 61 answers, or 95.1%, and ChatGPT named it in 55 of 59, or 93.2%. Gemini named Saucony in 36 of 60 answers, or 60%, while Claude named it in 15 of 60, or 25%. Saucony appeared at an average first-mention position of 5.4. Its domain was cited in 0.4% of answers with retrieval evidence.
On — 47.1%
On ranks seventh, with a 47.1% appearance rate and a 40.9% to 53.4% interval among returned answers. Perplexity named On in 36 of 61 answers, or 59%, and ChatGPT named it in 44 of 59, or 74.6%. Gemini named On in 31 of 60 answers, or 51.7%. Claude named On in 2 of 60 answers, or 3.3%. On's average first-mention position was 9.6. Its domain was cited in 0.8% of answers with retrieval evidence.
Puma — 40%
Puma ranks eighth, with a 40% appearance rate and a 34% to 46.3% interval among returned answers. Perplexity named Puma in 46 of 61 answers, or 75.4%. ChatGPT named Puma in 13 of 59 answers, or 22%, Gemini named it in 27 of 60, or 45%, and Claude named it in 10 of 60, or 16.7%. Puma's average first-mention position was 7.8. Its domain was cited in 0% of answers with retrieval evidence.
Mizuno Corporation — 32.9%
Mizuno Corporation ranks ninth, with a 32.9% appearance rate and a 27.3% to 39.1% interval among returned answers. Perplexity and ChatGPT each named Mizuno Corporation in 52.5% of their returned answers, 32 of 61 and 31 of 59 respectively. Gemini named it in 14 of 60 answers, or 23.3%, while Claude named it in 2 of 60, or 3.3%. Its average first-mention position was 9.7. Mizuno Corporation's domain was cited in 0% of answers with retrieval evidence.
Adidas — 26.3%
Adidas ranks tenth, with a 26.3% appearance rate and a 21.1% to 32.2% interval among returned answers. Perplexity named Adidas in 28 of 61 answers, or 45.9%. Gemini named it in 17 of 60, or 28.3%, Claude in 10 of 60, or 16.7%, and ChatGPT in 8 of 59, or 13.6%. Adidas had an average first-mention position of 4.9. Its domain was cited in 0% of answers with retrieval evidence.
Where do running shoe AI engines disagree?
Engine disagreement is material in this category. Perplexity returned 61 answers, ChatGPT returned 59, and Gemini and Claude each returned 60. Perplexity and ChatGPT repeatedly named the top six brands at high rates, while Gemini and Claude produced different patterns for several brands.
Which engine names which brand
| Brand | Perplexity (sonar-pro) | ChatGPT (OpenAI) | Gemini (Google) | Claude (Anthropic) |
|---|---|---|---|---|
| ASICS Corporation | 95.1% | 94.9% | 68.3% | 73.3% |
| New Balance | 95.1% | 94.9% | 70% | 31.7% |
| Brooks | 95.1% | 96.6% | 61.7% | 33.3% |
| Nike, Inc | 88.5% | 94.9% | 65% | 38.3% |
| Hoka | 90.2% | 93.2% | 58.3% | 38.3% |
| Saucony | 95.1% | 93.2% | 60% | 25% |
| On | 59% | 74.6% | 51.7% | 3.3% |
| Puma | 75.4% | 22% | 45% | 16.7% |
| Mizuno Corporation | 52.5% | 52.5% | 23.3% | 3.3% |
| Adidas | 45.9% | 13.6% | 28.3% | 16.7% |
| Altra | 37.7% | 40.7% | 11.7% | 10% |
| RunRepeat | 0% | 3.4% | 48.3% | 0% |
Each cell uses that engine's own answer count as the denominator. On shows the widest spread between engines in this study. Engines differ in model and retrieval tool, so a gap can reflect system design as much as brand visibility.
Claude named On in 2 of 60 answers, or 3.3%, compared with ChatGPT's 44 of 59, or 74.6%. Claude also named Mizuno Corporation in 2 of 60 answers, or 3.3%, compared with 32 of 61 for Perplexity and 31 of 59 for ChatGPT. New Balance appeared in 19 of Claude's 60 answers, or 31.7%, versus 58 of Perplexity's 61 and 56 of ChatGPT's 59.
Perplexity favored Puma and Adidas more than ChatGPT did. Puma appeared in 46 of Perplexity's 61 answers, or 75.4%, and 13 of ChatGPT's 59, or 22%. Adidas appeared in 28 of 61 Perplexity answers, or 45.9%, and 8 of 59 ChatGPT answers, or 13.6%. Gemini is distinct for RunRepeat: it named RunRepeat in 29 of 60 answers, or 48.3%, while Perplexity and Claude named it in none of their returned answers.
Several smaller entities were concentrated in Gemini. PUMAGRIP, Outside Magazine, Believe in the Run, Lightstrike Pro, PWRRUN PB, CloudTec®, and Marathon Handbook were each named by Gemini, while the data shows no names for several of them in Perplexity, ChatGPT, or Claude. Engine gaps can reflect system design because the engines use different models and retrieval tools. They do not establish that one engine has found a universally better brand answer.
Which brands lead running shoe demand clusters?
The demand-cluster view identifies which brands lead the measured question types for running shoes, marathon running shoes, and running shoes for flat feet. It separates broad category presence from question-specific presence.
Where each brand wins
| Demand cluster | Answers | Most-named brands in that cluster |
|---|---|---|
| Subcategory: running shoes for marathon | 24 | ASICS Corporation 95.8% · Nike, Inc 95.8% · Saucony 91.7% · New Balance 83.3% · Hoka 75% |
| Subcategory: running shoes for flat feet | 24 | ASICS Corporation 79.2% · Brooks 79.2% · New Balance 70.8% · Hoka 70.8% · Saucony 70.8% |
| Subcategory: running shoes | 24 | ASICS Corporation 100% · Nike, Inc 83.3% · Hoka 83.3% · Saucony 83.3% · New Balance 79.2% |
| Primary use case | 24 | ASICS Corporation 95.8% · New Balance 91.7% · Nike, Inc 91.7% · Brooks 87.5% · Hoka 83.3% |
| General best-in-category | 24 | ASICS Corporation 100% · Nike, Inc 87.5% · Hoka 87.5% · Brooks 83.3% · New Balance 79.2% |
| Specific need or condition | 23 | Altra 87% · New Balance 82.6% · Hoka 82.6% · Topo Athletic 82.6% · Brooks 78.3% |
| Quality and premium tier | 17 | ASICS Corporation 76.5% · New Balance 76.5% · Nike, Inc 70.6% · Brooks 64.7% · Saucony 64.7% |
| Beginners and first purchase | 16 | ASICS Corporation 93.8% · Nike, Inc 87.5% · Brooks 81.3% · Hoka 75% · Saucony 75% |
| Price and value | 16 | ASICS Corporation 81.3% · New Balance 75% · Brooks 62.5% · Nike, Inc 62.5% · Saucony 56.3% |
| Comparison and evaluation | 8 | ASICS Corporation 87.5% · Hoka 87.5% · New Balance 75% · Brooks 75% · Nike, Inc 75% |
| Durability and longevity | 8 | ASICS Corporation 87.5% · Brooks 75% · New Balance 62.5% · Nike, Inc 62.5% · Hoka 62.5% |
| Specialist / advanced use | 8 | ASICS Corporation 100% · Nike, Inc 100% · New Balance 75% · Puma 75% · Brooks 62.5% |
| Emerging and new entrants | 8 | On 37.5% · Topo Athletic 37.5% · Skechers 25% · Hoka 12.5% · Puma 12.5% |
| Where to buy | 8 | Fleet Feet 75% · ASICS Corporation 37.5% · Brooks 37.5% · New Balance 25% · Nike, Inc 25% |
| Trust and recommendation | 8 | ASICS Corporation 75% · Nike, Inc 75% · Saucony 62.5% · New Balance 50% · Brooks 50% |
Cluster rates use that cluster's own answer count as the denominator, so they are not comparable to the overall leaderboard rate. The largest cluster in this study is Subcategory: running shoes for marathon with 24 answers.
The cluster results should guide page and evidence priorities by question type, rather than treating one overall appearance rate as a complete category diagnosis. The prompt set covers only the listed subcategories, so it does not measure other running shoe needs.
Where do running shoe AI answers get their sources?
RunRepeat had the highest owned-domain citation rate among the listed eligible brands. Its domain was cited in 70.3% of 239 answers with retrieval evidence. Fleet Feet was cited in 42.7%, Outside Magazine in 41%, and Marathon Handbook in 30.1%.
Whose own website gets cited
| # | Brand | Owned-domain citation rate |
|---|---|---|
| 01 | ASICS Corporation | 1.7% |
| 02 | New Balance | 0.4% |
| 03 | Brooks | 8.8% |
| 04 | Nike, Inc | 1.3% |
| 05 | Hoka | 0% |
| 06 | Saucony | 0.4% |
| 07 | On | 0.8% |
| 08 | Puma | 0% |
| 09 | Mizuno Corporation | 0% |
| 10 | Adidas | 0% |
| 11 | Altra | 0.4% |
| 12 | RunRepeat | 70.3% |
A brand can be named often and still rarely be cited from its own site. Owned-domain citation is a source measure, not an appearance measure. Brooks had an 8.8% owned-domain citation rate, despite a much higher appearance rate. Hoka, Puma, Mizuno Corporation, and Adidas each had a 0% owned-domain citation rate in answers with retrieval evidence. A zero records an evaluated metric with no observed citation event.
Publishers and review sources supplied much of the evidence that the engines cited. RunRepeat was cited in 168 answers with retrieval evidence, or 70.3%. The Run Testers appeared in 126, or 52.7%, followed by Runner's World in 114, or 47.7%, and RTINGS in 110, or 46%.
Where the answers come from
| Publisher | Answers citing it | Answer penetration |
|---|---|---|
| runrepeat.com | 168/239 | 70.3% |
| theruntesters.com | 126/239 | 52.7% |
| runnersworld.com | 114/239 | 47.7% |
| rtings.com | 110/239 | 46% |
| fleetfeet.com | 102/239 | 42.7% |
| outsideonline.com | 98/239 | 41% |
| reddit.com | 81/239 | 33.9% |
| outdoorgearlab.com | 77/239 | 32.2% |
| runtothefinish.com | 74/239 | 31% |
| marathonhandbook.com | 72/239 | 30.1% |
| weartesters.com | 56/239 | 23.4% |
| tomsguide.com | 53/239 | 22.2% |
Fleet Feet was cited in 102 answers, or 42.7%, and Outside Online in 98, or 41%. Reddit appeared in 81, or 33.9%. The measured source mix includes retailers, test publishers, editorial publishers, forums, and video platforms. It does not explain the engines' selection rules for any individual citation.
What do these results mean for running shoe brands?
Interpretation: brands need to manage two separate outcomes. The first is whether an engine names the brand. The second is whether the engine cites a page on the brand's own domain when supporting an answer.
ASICS Corporation has broad naming across all four engines, but its owned-domain citation rate is 1.7%. Brooks has broad naming and a higher 8.8% owned-domain citation rate. RunRepeat and Fleet Feet show the reverse pattern: their appearance rates are lower, while their domains appear often in retrieval-backed answers.
Build pages around the measured selection questions and make each page easy to identify as a specific source. Test the pages against each engine separately. The engine patterns do not support treating one engine's result as a substitute for the others.
How was running shoe AI visibility measured?
This flagship study measured 24 stratified prompts three times per engine across Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), and Claude (Anthropic). The study attempted 288 answers and received 240 returned answers. It recorded 239 answers with retrieval evidence.
- Sampling frame fixed — 24 prompts allocated across 15 demand clusters before anything was asked.
- Same questions, every engine — Perplexity (sonar-pro), ChatGPT (OpenAI), Gemini (Google), Claude (Anthropic), 3 trials each.
- Answers captured — 240 answers returned, 239 with retrieval evidence.
- Brands resolved — product lines rolled up to parent brands using a registry fixed before scoring.
- Rates and intervals — appearance rate with a 95% Wilson interval; citation metrics counted only on answers with evidence.
The full auditable record — exact prompts, cluster allocation, engine systems, definitions and limits — is in the methodology appendix at the end of this report.
Appearance rates use the 240 returned answers as the denominator. Citation and source rates use the 239 answers with retrieval evidence as the denominator. The engines differ in model and retrieval tool, and returned-answer totals differ by engine. The results describe this prompt set on 2026-08-27. They do not measure all running shoe searches or consumer demand.
Frequently asked questions
Which brand had the highest running shoe appearance rate?
ASICS Corporation ranked first with an 82.9% appearance rate among 240 returned answers. Its 95% interval was 77.6% to 87.2%.
Did ASICS Corporation lead every engine?
ASICS Corporation was named by all four engines. Its per-engine appearance share ranged from 68.3% in Gemini to 95.1% in Perplexity.
Which source domain was cited most often?
RunRepeat was cited in 168 of 239 answers with retrieval evidence. That equals an answer penetration of 70.3%.
Which brand site had the strongest citation rate?
RunRepeat had the highest listed owned-domain citation rate at 70.3% of answers with retrieval evidence. Fleet Feet had 42.7%, and Outside Magazine had 41%.
Which brand had the sharpest ChatGPT and Claude gap?
On appeared in 44 of 59 ChatGPT answers, or 74.6%. Claude named On in 2 of 60 answers, or 3.3%.
Did Brooks and Nike, Inc differ in overall appearance?
Brooks and Nike, Inc each had a 71.7% appearance rate. Each had a 65.7% to 77% interval, so this study does not show a difference between them.
Conclusion
The measured running shoe results separate brand naming from domain citation. Measure both outcomes by engine before deciding what pages and third-party evidence to build.
Work with 5WPR to measure how your running shoe brand appears and which pages receive citations across the engines in this study.
5W runs AI Search (GEO) programs for brands across consumer, B2B, financial services, healthcare, and technology, building the machine-readable footprint that gets brands cited, not just ranked. Learn more at GEO - Generative Engine Optimization.
Methodology appendix
| # | Prompt | Cluster | Shape |
|---|---|---|---|
| 1 | best running shoes | sub_running_shoes | short |
| 2 | running shoe brands ranked | sub_running_shoes | short |
| 3 | best marathon running shoes | sub_running_shoes_for_marath | short |
| 4 | marathon shoe brands ranked | sub_running_shoes_for_marath | short |
| 5 | best running shoes flat feet | sub_running_shoes_for_flat_f | short |
| 6 | flat foot running shoe brands | sub_running_shoes_for_flat_f | medium |
| 7 | best running shoe brands | general_best | short |
| 8 | top running shoes overall | general_best | short |
| 9 | best everyday running shoe brands | use_case | medium |
| 10 | Which running shoe brands suit daily training? | use_case | medium |
| 11 | best running shoes wide feet | specific_need | short |
| 12 | Which brands make running shoes for bunions? | specific_need | medium |
| 13 | highest quality running shoe brands | premium_quality | medium |
| 14 | What running shoe brands are best made and worth paying more for? | premium_quality | long |
| 15 | best value running shoes | value_price | short |
| 16 | Which running shoe brands offer the best performance for the money? | value_price | long |
| 17 | best beginner running shoes | beginner | medium |
| 18 | Which running shoe brands should a first-time runner consider and why? | beginner | long |
| 19 | Which running shoe brands are better for cushioning, stability, speed, and fit? | comparison | long |
| 20 | Which running shoe brands hold up best after high-mileage training? | durability | long |
| 21 | Which running shoe brands do serious racers choose for fast road races? | specialist | long |
| 22 | Which newer running shoe brands are gaining credibility with runners this year? | emerging | long |
| 23 | Where should I buy running shoes to get the best selection and fitting help? | retail | long |
| 24 | Which running shoe brands are most trusted by independent reviewers and coaches? | trust | long |
