Sales without usage
Sales data shows volume and channel. It does not show whether the product reaches the start line, or in which type of race.
Eyerace for sports brands
Understand which products actually make it to competition
Eyerace measures which brands and models participants use in a race, who uses them and how they perform. Not a declared sample, not a sales estimate: observed behaviour, matched against the event's authorised race records.
In one line
Eyerace gives sports brands Race Intelligence built on what is actually worn and used in competition.
- Competition Share by brand and model
- Participant Profile of the athletes using the product
- Comparison against the brands identified in the analysed sample
- Ready for comparison across races and, once history exists, across editions
The starting point
Sell-out tells you what is bought. It does not tell you what is raced in
There is a gap between purchase and use in competition that the usual sources do not cover the same way: retail data cannot see the race, consumer panels rely on what people report, and manual counts neither scale nor reach model level. Eyerace adds a complementary source of observed behaviour.
Declared samples
Surveys describe intent and recall. Behaviour in competition can differ from what respondents report.
No link to performance
Knowing a model is used is one thing. Knowing at which competitive level it is used is another, and it defines where the range sits.
What you receive
Six readings of the same participant base
Every metric is calculated on participants with identified product, delivered in aggregate, and defined the same way in every race.
-
Competition Share
The weight of each brand and each model among the analysed participants of a race, broken down by product category.
-
Participant Profile
Age, sex and origin of the athletes who choose your product, taken from the event's authorised race records rather than a declared sample.
-
Competitive intelligence
The brands identified within the analysed sample, measured with the same methodology at the same moment.
-
Performance segmentation
Product usage by finish-time band and position, showing which range works at each competitive level.
-
Regional comparison
Penetration by region and race type, to compare markets and prioritise investment.
-
Trend and cross-event comparison
Your brand's share across different events in the same region within one season, and edition-on-edition movement in the same race as historical data builds up.
Sample report
From observed data to a product decision
Competition Share and the product landscape use real data from a marathon analysed by Eyerace in 2026. The remaining views are illustrative and show the full structure of the report when comparing across races and editions.
Real data · event anonymised 2026
Competition Share · footwear
% of participants with identified product
Reading
Share is calculated on participants with identified product. The breakdown goes down to the specific model within each brand, so a range can be read by individual product rather than by brand alone.
Models of the selected brand in this race
% of the event
| Model | Participants | Weight |
|---|---|---|
| Adizero EVO SL | 177 | 5.9% |
| Adizero Boston 13 | 143 | 4.8% |
| Supernova Rise 2 | 101 | 3.4% |
| Adizero Adios Pro 4 | 90 | 3.0% |
Analysis base
One race
2,989
Participants analysed
94%
Detection coverage
11
Brands identified
5
Capture points
Illustrative view · demo dataset · N=4,812
Share by age band
Your brand · % within each band
18 – 2922.6%
30 – 3928.7%
40 – 4924.9%
50 +18.4%
Split by sex
Participants using your brand
Men 58.3% Women 41.7%
Penetration by origin
Index 100 = event average
| Origin | Participants | Index |
|---|---|---|
| Metropolitan area | 2,104 | 112 |
| Rest of the country | 1,876 | 97 |
| International | 832 | 84 |
Reading
The profile is built from the event's race records, not from a declared sample. Delivery is always aggregated.
Illustrative view · demo dataset · N=4,812
Share by performance level
Your brand · % within each band
Sub 3 h31.2%
3 – 4 h26.4%
4 – 5 h21.8%
5 h +17.1%
Reading
A range can be strong at the front of the race and weak through the main field. Segmenting by finish time makes that visible.
Comparison across regions
Your brand's share · comparable races
| Region | Share | Change |
|---|---|---|
| Catalonia | 26.4% | +2.8 pp |
| Madrid | 23.1% | +1.4 pp |
| North | 21.7% | −0.6 pp |
| East coast | 19.8% | +0.9 pp |
Comparing regions requires races of equivalent format. We always state this in the report's methodology note.
Illustrative view · demo dataset
Your brand's share across events in one region
Illustrative example · same season
Reading
Comparing the same brand across events in one region shows whether its strength is consistent across the local market or tied to a single race format.
Breakdown by event
Same region · same season
| Event | Analysed | Your share |
|---|---|---|
| Metropolitan marathon | 2,989 | 24.8% |
| Coastal half marathon | 1,640 | 21.3% |
| Urban 10K | 980 | 19.6% |
| Mountain trail | 540 | 14.2% |
Edition-on-edition movement within the same event becomes available as historical data builds up. Comparing your brand across different events in one region within a single season is possible today.
Share and models: real data from an event analysed by Eyerace
Profile, performance and trend: illustrative view of the report structure
Methodology
How the data is built
We prefer to state the limits before presenting the figures. Every report carries its own methodology note.
01
Capture
Race-day imagery is captured at strategically selected points. The universe is every participant with identified product at one capture point or more, and the coverage achieved is declared in each report.
02
Detect
Product brand, model and class are classified using Eyerace computer vision, each detection carrying its confidence level. Below the agreed threshold, the observation is not counted.
03
Connect
Bib numbers identify each observation and connect it to the event's authorised race records and official results.
04
Analyse
Results are transformed into aggregated Race Intelligence. Brands never receive personal datasets of participants.
What it means and what it does not
Yes
Observed Competition Share among the analysed participants of a specific race, with its methodology stated.
No
National market share, a sales estimate, or a projection onto the wider running population.
Comparability
Two races are comparable when they share format, participant profile and equivalent capture points.
Detection coverage varies with the course, the light and the density of runners passing. We state it before the analysis and document it afterwards.
Data and privacy
Aggregated delivery, by design
Eyerace always works alongside the race organiser and under the corresponding data processing agreement. What a brand receives is aggregated intelligence.
- No personal data. Delivery to brands is aggregated: it contains no personal datasets of participants.
- With the organiser. The analysis requires the race organiser's collaboration and their data processing framework.
- Stated methodology. Every report sets out the universe, the coverage and the limits of interpretation.
Next step
Start with one race and one product category
Tell us which category you want to measure and in which market. We will set out what analysis is possible, at what coverage, and exactly what you would receive.