---
title: "Vonovia — DAX 40 Agent Success Score | Hyperize"
description: "Vonovia: AI Visibility 42.8/100, primary gap Interception. Hyperize DAX 40 Agent Success Index, measured under a public Task Selection Doctrine."
canonical: https://www.hyperize.ai/en/dax40-index/brands/vonovia
lang: en
last-updated: 2026-07-02
---

# Vonovia.

[Home](/en/) / [DAX 40 index](/en/dax40-index) / Vonovia

Original research · Real Estate



Last measured · 02 Jul 2026 Wave · Q2-2026-W3-SLOW Tier · proprietary Confidence · C

Brand ![Vonovia](/logos/dax40/vonovia.svg)

Vonovia SE

Agent success

## On the unbranded question the portals capture the search.

Bottleneck Interception

Intercepted before the brand is the answer.

4.4 /10

Agent Success Score

AI Visibility 43 / 100

Found & recommended by AI agents

AI Usability 41 / 100

Can an agent use the surface to get the job done

Coverage · 2 of 6 lanes measured Commerce lane · Wave Q2-2026-W3-SLOW

Commerce 4.4

Talent usability measured

After-sales —

Procurement —

Investor —

Press —

This page measures two lanes. Commerce: can an agent find Vonovia, and once it arrives, transact. Talent: can a candidate's agent find an open engineering role on Vonovia'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 Vonovia, free for your domain. Five agent classes, one real task, your score in 48 hours.

The test

## Portals own the unbranded search. The landlord surface itself is agent-ready end to end.

We asked five kinds of AI agent to find a 3-room rental in Munich under €1,500 cold and start a landlord contact on vonovia.de. A plain HTTP fetch read the Kaltmiete (€1,891), Warmmiete (€2,150), and address from the SSR listing without JS execution. The coding agent extracted the same listing data. The standard browser reached the listing detail in two of three runs; the only drop was the homepage search-to-listing handoff. The autonomous agent reached the contact overlay with all six visible inquiry fields and the 'Absenden' submit button. Five out of five agent classes reached at least the listing; the inquiry overlay opened end to end without authentication.

One label, many breeds. From a plain reader to an autonomous operator, the kinds behind ChatGPT, Perplexity, and Claude Code:

Plain reader reads your raw page text, no browser Succeeded

Search assistant finds you through search Not yet run

Coding agent a script hitting your site Succeeded

Computer-use agent clicks and types like a person Partial

Autonomous operator runs the whole task unattended Succeeded

Scope. This is one listing in one city. Vonovia operates hundreds of thousands of units across Germany, each with its own listing and inquiry flow.

Commerce lane

## Found, and able to transact?

Two questions, measured separately. A brand can be recommended and still un-buyable, or perfectly buyable and never found.

AI Visibility

When someone asks an agent to find a property, does it route to Vonovia?

43 / 100

Vonovia comes up about four times in ten. The rest of the time, an agent recommends an alternative first.

Discoverability · 18-datapoint audit

AI Usability

Once an agent is on Vonovia's site, can it book a viewing?

41 / 100

Plain reader and Coding agent and Autonomous operator reach contact-ready; Computer-use agent get there only in some runs.

Browser-class agents reach contact-ready; browser-class agents reach it only partially · breakdown above

Evidence · 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.

Talent lane

## The talent lane.

A candidate's agent, sent to find an open software-engineering role at Vonovia and prepare the application. Same brand, second surface: the career portal.

AI Usability · Talent lane

### Readable posting, unresolved apply step.

Both non-browser agent classes read the posting and locate the application entry; how far the form itself carries without JavaScript was not part of this pass.

Furthest close-state reached: spec match · frozen task locked https://jo

58 AI Usability / 100 · derived, not hand-rated

Plain reader fetches the posting page raw, no browser pass GET on the locked posting URL returns HTTP 200, 80930 bytes, plain server-rendered HTML (no JS required). Full job title, employer description, "Aufgaben" (tasks) and "Dein Profil"/"Anforderungen" (requirements) sections

Coding agent parses the posting into structured fields pass Structured data present as schema.org microdata (itemscope/itemprop), not a JSON-LD <script> block: itemtype="http://schema.org/JobPosting", jobLocation > PostalAddress (addressLocality=Bochum, postalCode=44803, addressC

Computer-use agent clicks through the career portal like a person untested

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 attempt or complete an actual application submission. No form was filled in, no account was created or logged into, and no data was submitted to Vonovia at any point. The apply dialog's login requirement and required artifacts (CV, cover letter, form length) remain unverified because they only render after the client-side JS dialog is triggered, which is out of scope for the HTTP/coding measurement class.

Fairness note

Vonovia hosts its job postings directly on its own careers site, so this measurement required no detours via subsidiaries or job boards, and machine-readable agents could retrieve the full tested posting and its structured data on the first request. Software roles are a small minority of Vonovia's hiring, which is dominated by trades, facility and administrative positions; the IT role was chosen solely to keep the test identical across all 40 companies. Browser-based agents were not tested here, so the finding that automated requests cannot reach the application form applies only to the agent classes actually tested, and no application was ever submitted.

What's next

## What this means for Vonovia.

Diagnosis

On product discovery, Vonovia is not failing on the surface; the listing reads clean and the inquiry overlay opens for every measured agent class. It loses at the entry point: the unbranded Munich-rental question routes to ImmoScout24 and Immowelt before it routes to Vonovia. Found via the portals, completed by every agent that arrives.

What changes the outcome

Discovery packaging, not surface repair. Answer Pages for the rental questions a tenant asks before naming a portal, Concept Pages that own the landlord-vs-portal vocabulary, Reference Pages that concentrate Vonovia's per-city listing authority, so the unbranded Munich question names Vonovia before ImmoScout24 frames the shortlist.

What proof looks like

The proof isn't more listings. It's agents surfacing Vonovia on broad rental prompts, not only when the landlord is already named, re-measured each wave.

Audit · €1,900

### 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 audit](mailto:hello@hyperize.ai?subject=Audit%20%C2%B7%20Vonovia)

Founding · €4,500

### Found with us.

Strategic partnership for brands building agent success as a long-term capability, not a one-off engagement.

[Apply](mailto:hello@hyperize.ai?subject=Founding%20Program%20%C2%B7%20Vonovia)

Snapshot

### Audit 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 Snapshot](mailto:hello@hyperize.ai?subject=Snapshot%20%C2%B7%20adjacent%20property%20for%20Vonovia)

This is Vonovia. 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.

✓

**Measured, not guessed.** Real agents against your real site, no questionnaire.

✓

**Publicly comparable.** Your score in the same grid as the measured DAX 40 brands.

✓

**Next wave closing.** Start now to be in the next index round.

Channel position

## Listing portals capture 83% of property demand before Vonovia.

Across the 18 AI Visibility responses, 83% routed the Munich-rental question through a displacing portal (Immowelt in 14, ImmoScout24 in 11, wg-gesucht in 11, Kleinanzeigen in 9); vonovia.de was named directly in 9 of 18 (50%). The portals dominate the unbranded informational probe; on the branded comparative and transactional probes vonovia.de appears alongside them. Vonovia and Scout24 share the identical informational probe ('Wie finde ich eine 3-Zimmer-Mietwohnung in München bis 1.500 Euro kalt?'), so the contrast measures whether agents surface the landlord or the portal on a single query: the portal mentions outnumber landlord mentions by a wide margin.

50% direct

83% via intermediary

Intermediaries Immowelt · ImmoScout24 · wg-gesucht · Kleinanzeigen

Frozen task slate

## Hyperize-selected tasks.

One task from the public sector grid. Task list is frozen before each wave runs.

Vonovia — 3-Zimmer-Mietwohnung München (Kontakt)

Close state

contact-ready

Bottleneck

HTTP reads Kaltmiete + Warmmiete + address from SSR; coding extracts the same; standard browser 2/3 (drop is homepage search-to-listing handoff); autonomous agent opens the landlord-owned inquiry overlay end to end. All five agent classes reach at least the listing. The dominant gap is discovery: the unbranded query routes to ImmoScout24 and Immowelt.

Fairness note

Wave 3 Slow Lane (a fleet wave Real Estate). Single measured task (3-Zimmer-Mietwohnung in München bis 1.500 Euro kalt, contact\_ready close state via landlord-owned inquiry overlay). AI Visibility 42.78 (18/18 valid datapoints across 3 providers, unbranded informational + branded comparative + branded transactional); AI Usability derived from a fleet wave per-class access profile (5/5 agent classes reach the listing; 2/3 standard browser runs reach listing detail; the autonomous agent opens the contact overlay end to end). Channel split grounded via channel-derive.py (intermediary in 15/18 AI Visibility responses, own-domain in 9/18). Confidence C, single task; Fairness Review pending the sector fairness grid.

Methodology

## 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. [\[S1\]](#sources)

AI Usability is derived from the access-profile above (usability-derivation/v1.2): 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. [\[S2\]](#sources)

Formula

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 C · 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.

History

## Measurement timeline.

Each wave appends; nothing overwrites. Frozen Wave Rule.

1.  Entry · 01
    
    23 May 2026
    
    Wave · Protocol
    
    WAVE-Q2-2026-W3-SLOW
    
    ars-methodology/v1.1
    
    First v3 measurement. AI Visibility 42.78 (18/18 valid datapoints, 3 providers, DE, unbranded informational + branded comparative + branded transactional). AI Usability derived from a fleet wave per-class access profile (5/5 agent classes reach listing; HTTP reads SSR listing; coding extracts; standard browser 2/3 reaches listing detail; autonomous agent opens landlord-owned inquiry overlay end to end). The unbranded informational query is intentionally identical to Scout24's, so the Third-Party Interception contrast measures landlord-vs-portal surfacing on a single probe (15/18 portal capture, 9/18 landlord direct). Confidence C, single task.
    

Sources

## Evidence and provenance.

Public methodology references and internal evidence pointers behind every claim above.

1.  \[S1\]
    
    AI Visibility audit · Vonovia Wave Q2 2026
    
    Accessed · 23 May 2026
    
    Internal · Hyperize evidence
    
    -   · AI Visibility score 42.78 (18/18 valid datapoints across 3 providers)
    -   · Third-Party Interception read (15/18 portal capture, 9/18 landlord direct mentions)
2.  \[S2\]
    
    Hyperize fleet · a fleet wave Real Estate (access profile)
    
    Accessed · 23 May 2026
    
    Internal · Hyperize evidence
    
    -   · how each kind of agent fared (HTTP reads SSR listing, coding extracts, browser 2/3, ACT opens inquiry overlay)
    -   · the close state reached (contact\_ready, landlord-owned inquiry overlay visible)
    -   · the tested product (3-room Munich rental, Kaltmiete €1,891, Warmmiete €2,150)
    -   · Five-Agents-Five-Answers passing 5/5, sub\_pilot false
3.  \[S3\] [Task Selection Doctrine](https://www.hyperize.ai/en/methodology/task-selection)
    
    Accessed · 23 May 2026
    
    Public · hyperize.ai
    
    -   · fairness declaration
    -   · Third-Party Interception framing

Related

## Read the doctrine. Challenge the score. Extend the slate.

[Methodology

### Task Selection.

The fairness doctrine behind the slate above. Five failure modes, six criteria, public before each wave.

Read](/en/methodology/task-selection) [Challenge

### Disagree with this score.

Send evidence under public Fairness Review. Failed reviews are documented with the named failure mode.

Challenge](mailto:hello@hyperize.ai?subject=DAX%2040%20Pilot%20%E2%80%94%20Vonovia%20Reference%20Page%20Challenge) [Extend

### Submit your own task.

Open Surface Run · additive measurement. The Hyperize-selected slate stays frozen; your task gets the same methodology.

Submit](mailto:hello@hyperize.ai?subject=Open%20Surface%20Run%20%E2%80%94%20Vonovia)

Last updated · 02 Jul 2026 Next review · 30 Sept 2026 Wave · Q2-2026-W3-SLOW Tier · proprietary Confidence · C Index score · 4.4/10 [Machine-readable record](https://www.hyperize.ai/en/dax40-index/brands/vonovia.json)

Universe: DAX 40 composition as of 2026-07-01, reviewed after each Deutsche Boerse index review.

Editorial 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](mailto:hello@hyperize.ai). Responses received are published in full alongside the findings. Full methodology and editorial-coverage notice: [coverage statement](/en/imprint#dax40-disclaimer).

---

Canonical: https://www.hyperize.ai/en/dax40-index/brands/vonovia
Machine-readable index: https://www.hyperize.ai/llms.txt