adidas.
Akamai 403 at homepage entry across every measured agent class.
Found & recommended by AI agents
Can an agent use the surface to get the job done
This page measures two lanes. Commerce: can an agent find adidas, and once it arrives, transact. Talent: can a candidate's agent find an open engineering role on adidas's own career surface and prepare the application. After-sales, procurement, investor and press lanes run on different surfaces and are not yet measured.
Your brand isn't measured yet
What does an AI agent do with your website?
The same measurement as adidas, free for your domain. Five agent classes, one real task, your score in 48 hours.
Documented hard-block. Sub-pilot finding, measured below the G6 threshold.
We asked five kinds of AI agent to add an Ultraboost 5 (men's, size 43) to a cart on adidas.de. Every single one hit the Akamai 403 page at homepage entry: 'Leider können wir im Moment keinen Zugriff auf unsere Seite geben.' Reference URL: error 0.cad5ce17.1775309423.3d04a9b. The block was stable across all three standard browser runs; the autonomous agent observed the same blocked product page on the start URL. No product detail, cart, login prompt, or checkout form became reachable. The Five-Agents-Five-Answers passing count is 0/5; per G6 this is a sub-pilot finding.
One label, many breeds. From a plain reader to an autonomous operator, the kinds behind ChatGPT, Perplexity, and Claude Code:
Blocked somewhere on the path: Plain reader, Coding agent, Computer-use agent, Autonomous operator. A customer whose assistant runs on one of those breeds never finishes the task.
Found, and able to transact?
Two questions, measured separately. A brand can be recommended and still un-buyable, or perfectly buyable and never found.
When someone asks an agent to buy a product, does it route to adidas?
adidas comes up about a third of the time. The rest of the time, an agent recommends an alternative first.
Discoverability · 18-datapoint auditOnce an agent is on adidas's site, can it check out?
No agent class reaches a working cart.
No agent class reaches a working cart, final payment is a separate step · breakdown aboveEvidence · 70 / 100 A measure of how provable and consistent the result is, grounded in cross-method ground-truth agreement (the methods that ran returned the same price), not a separate measured run. A confidence layer on the two scores above, not a third sales axis.
The talent lane.
A candidate's agent, sent to find an open software-engineering role at adidas and prepare the application. Same brand, second surface: the career portal.
Readable posting, unresolved apply step.
Both non-browser agent classes read the full posting server-side as plain HTML prose (no JSON-LD, no JS rendering needed for the JD itself); the linked application entry point cannot be resolved by a raw HTTP request and requires browser-side JavaScript execution to determine whether it leads to a login wall or a direct application form.
Furthest close-state reached: spec match · frozen task locked 2026-07-02
The score above is provisional where a class shows untested: untested classes are excluded from the calculation, so the pending check can only confirm or raise the value, never silently lower it.
AI Visibility · in measurement — the audit run for this lane has not landed yet. No Talent-lane composite is shown until both axes are measured.
What this does not yet cover
This measurement did not submit an application, upload any file, create an account, or log in. Only the GET-reachable posting-detail page and a raw GET on the linked apply-entry URL were inspected; the apply URL's 301 redirect to the domain root was observed at the HTTP layer only — no browser/JS execution was used to follow the client-side route further. Whether login is required, what artefacts (CV/cover letter) are needed, and the real form length remain unconfirmed and are deferred to the Phase C browser pass.
Fairness note
This measurement used a live adidas engineering posting (Connectivity Platform Engineer, Req 541628, Herzogenaurach), frozen on 2 July 2026 with an archived copy, so a later closing of the role does not change the result. No application was submitted and no candidate data was used; if the application flow requires a login, that is recorded as a fact, not counted against adidas. The job ad itself is fully machine-readable as plain HTML on adidas's own job server; the one open question is the application entry point, which could not be resolved without in-browser JavaScript and will be verified in a follow-up browser check that can only confirm or improve the current score.
What this means for adidas.
Adidas is not failing on product or surface design here; it is unreachable to every agent class we measured. The Akamai-shaped 403 at homepage entry is the deal-breaker. This is a documented hard-block, the sub-pilot finding most operators would prefer to know early. Until that entry-gate distinguishes humans from agents in a configurable way, no product, cart, or checkout measurement is possible on this surface.
Agent-class entry policy, not surface polish. A WAF rule that admits identified, declared agents to the same surface humans use, an agent-readable catalog and add-to-cart contract, and citeable evidence that the surface accepts agent traffic so an answer engine learns it can recommend adidas at all. None of that is possible while the homepage returns 403 to every agent class.
The proof is not a different shoe. It is the same Ultraboost 5 add-to-cart task running clean from at least three of five agent classes on the next wave, re-measured.
Commission an audit.
Where the BrandScore opens the question, an audit closes it. An interpretive engagement on your full surface, scored under the same methodology.
Get the auditFound with us.
Strategic partnership for brands building agent success as a long-term capability, not a one-off engagement.
ApplyAudit an adjacent property.
The BrandScore covers the primary domain. Get the same methodology applied to an adjacent property: a country site, a sub-brand, a category beyond the DAX-40 slate.
Get a SnapshotThis is adidas. What about your brand?
You just saw how an AI agent treats a DAX 40 brand.
The same measurement runs free against your domain: five agent classes, one real buying task, your Agent Success Score in 48 hours, in the exact format of this page.
No intermediary stands between agents and adidas. The gap is being found, not the channel.
Across the 18 AI Visibility responses, no displacing intermediary captured the running-shoe question: brand-direct surfaces (Nike, ASICS, Brooks) get named as alternatives, but those are competitors, not intermediary capture. adidas.de appeared in 5 of 18 (28%). The story is brand absence on the unbranded category probe, not portal displacement — at the surface itself the Akamai 403 prevents any agent from completing the path even when the brand IS named.
Hyperize-selected tasks.
One task from the public sector grid. Task list is frozen before each wave runs.
adidas Ultraboost 5 (cart) — sub-pilot probe
- Close state
- a working cart
- Bottleneck
- Stable homepage-level Akamai 403 across all measured agent classes; failure is hard-block, not product unavailability. No product, cart, or checkout interaction reachable. Sub-pilot per G6 (0/5).
Fairness note
Wave 3 Slow Lane (a fleet wave Consumer Retail, sub-pilot). Single measured task (Ultraboost 5 add-to-cart, cart_ready close state). Hard Akamai 403 block at homepage entry across every agent class measured; 0/5 reach close state, sub-pilot per G6. AI Visibility 35.0 (18/18 valid datapoints across 3 providers); AI Usability is low (derived from the per-class access profile — every agent class is blocked at the entry gate, so the score reflects depth-of-block, not cart depth). Channel split grounded via channel-derive.py (0/18 displacing intermediary; brand competitors Nike/ASICS/Brooks excluded per Resolver as competitors not capture). The page carries the Pilot probe chip and noindex via confidence=D — measured, but below the G6 threshold for a normal scored brand.
How the score was produced.
Discoverability is audit-pipeline-derived. 3-provider sample (openai, perplexity, anthropic), 3 query variants per task, 2 runs per variant · 18 valid datapoints scored against a five-state handoff cascade. [S1]
AI Usability is derived from the access-profile above (usability-derivation/v1): how far the best agent reached (close state) modulated by how many agent classes succeeded. The per-class profile is the truth; the score is a reproducible summary of it, not a separate rating. Fleet phases (HTTP / Coding / Browser / ACT) produce the profile. [S2]
Agent Success Score = (AI Visibility × 0.20) + (AI Usability × 0.70) + (Evidence × 0.10)
On a 0–100 scale, displayed 0–10. AI Usability bundles the agent's reach + completion; AI Visibility is audit-derived discoverability. Weighting is public; the per-prompt derivation is not.
Measurement scope
Confidence D · one measured task on a 3-provider track (openai, perplexity, anthropic). Confidence promotes to B with a second task plus a fourth provider on the next wave.
Measurement timeline.
Each wave appends; nothing overwrites. Frozen Wave Rule.
- Entry · 01
23 May 2026
Wave · ProtocolWAVE-Q2-2026-W3-SLOW
ars-methodology/v1.1
First v3 measurement. Documented hard-block finding: stable Akamai 403 at homepage entry across every agent class (text, coding, all three standard browser runs, autonomous). No product, cart, or checkout interaction reachable. Five-Agents-Five-Answers passing 0/5, sub-pilot per G6. AI Visibility 35.0 (18/18 valid datapoints, 3 providers, DE). Shipped with confidence D (Pilot probe chip + noindex) — measured below the G6 threshold.
Evidence and provenance.
Public methodology references and internal evidence pointers behind every claim above.
- [S1]Accessed · 23 May 2026
AI Visibility audit · Adidas Wave Q2 2026
Internal · Hyperize evidence
- · AI Visibility score 35.0 (18/18 valid datapoints across 3 providers)
- [S2]Accessed · 23 May 2026
Hyperize fleet · a fleet wave Consumer Retail (access profile)
Internal · Hyperize evidence
- · documented Akamai 403 hard-block at homepage entry across all measured agent classes
- · Five-Agents-Five-Answers passing 0/5, sub_pilot true
- · reference error 0.cad5ce17.1775309423.3d04a9b
- Accessed · 23 May 2026
Public · hyperize.ai
- · fairness declaration
- · sub-pilot doctrine (G6)
Read the doctrine. Challenge the score. Extend the slate.
Task Selection.
The fairness doctrine behind the slate above. Five failure modes, six criteria, public before each wave.
Read ChallengeDisagree with this score.
Send evidence under public Fairness Review. Failed reviews are documented with the named failure mode.
Challenge ExtendSubmit your own task.
Open Surface Run · additive measurement. The Hyperize-selected slate stays frozen; your task gets the same methodology.
SubmitEditorial coverage
The DAX 40 Agent Success Index is a point-in-time snapshot of the agent-success of public digital touchpoints. Results are not statements about product quality, company performance, service quality, or the legal obligations of the brands named. Brand names and logos remain the property of their respective owners and are used solely for identification and reporting purposes in the context of editorial coverage (§ 23 MarkenG, Art. 5 GG).
Brands wishing to respond, engage, or correct a factual error may contact hello@hyperize.ai. Responses received are published in full alongside the findings. Full methodology and editorial-coverage notice: coverage statement.
