Generative AI Use Cases In Real Estate Marketing: Adoption and Constraints in Germany
Victor Kolominsky-Rabas, Leopold Müller, Felicia Perpina, Niklas Kühl
cs.AI · cs.HC
Quick verdict
Eleven-interview study maps current vs potential GenAI use cases in German real estate marketing; credible qualitative work, limited by AI-user-only sample.
Summary
This qualitative study examines how German real estate agents (REAs) actually use generative AI in marketing practice. The authors conducted eleven semi-structured interviews (June-August 2025, averaging 29.6 minutes) with agents recruited via convenience and snowball sampling, all of whom were active AI users, and analyzed transcripts with Gioia-style inductive coding in two cycles separating current from potential use cases. The result is a grounded map spanning six activity clusters: marketing communication dominates (47.7% of participant-level first-order references), followed by general communication (16.5%), property analysis (14.7%), acquisition preparation (9.2%), general knowledge support (7.3%), and sales (4.6%). Writing exposé texts is the only widely established use case; image editing, social media posts, and floor-plan visualization remain emergent and unevenly adopted. Reported interaction is predominantly human-in-the-loop - GenAI drafts and retrieves while agents verify and decide - and constraints are attributed less to model capability than to listing/document integration, data availability, and GDPR/EU AI Act compliance concerns, notably in client screening. Main limitations: a small, AI-user-only convenience sample precludes claims about non-adoption; self-reported use is not verified against artifacts; no inter-coder reliability is reported; and the percentages are descriptive only.
Review complete
Assessment at a glance
4
risk flags identified
● high 0
● medium 2
● low 2
5 strengths · 5 weaknesses · 5 author questions
AI text signal = Binoculars zero-shot detector score over 20 sampled passages (Fast-DetectGPT co-reported) — absence does not prove human authorship, and a flag does not imply misconduct.
Levels reflect verification complexity — not a quality score. Evolve never rates your paper.
Key claims under review
- GenAI is already used across the German real estate agent workflow, with marketing communication the most prominent cluster (47.7% of participant-level first-order references).
- Writing exposé texts is the only widely established GenAI use case; all other applications are emergent and unevenly adopted.
- Reported interaction is predominantly human-in-the-loop: GenAI drafts, structures, and retrieves while agents curate, verify, and remain accountable.
- Adoption constraints stem more from listing/document integration, data availability, and GDPR/EU AI Act compliance concerns than from model capability.
Strengths
Clear GenAI boundary definition
Section 2.2 explicitly separates GenAI from adjacent analytics/automation/rule-based software before coding, which prevents category inflation - a common flaw in use-case mapping studies.
Transparent, two-cycle Gioia coding with explicit non-saturation claim
Follows Gioia et al. [16] with two coding cycles (current vs. potential), lists all first-order concepts with agent IDs (Figures 1-3), and explicitly states that full theoretical saturation is NOT claimed (Section 3) - commendable honesty.
Honest, well-scoped limitations
Genuine cross-country extension rather than repetition
Rich participant context
Weaknesses
Small, AI-user-only convenience sample constrains external validity
Figure 5 percentages invite over-interpretation
No inter-coder reliability for the Gioia coding
Self-report without artifact triangulation
Compliance claims rest on lay perceptions
Full Review Report
$1.99 · one-timeUnlock the complete preliminary analysis Evolve produced for this paper.
Report preview
Scope & Claims
The paper answers two questions: which GenAI use cases German real estate agents (REAs) already apply (RQ1) and which potential use cases GenAI offers for real estate marketing (RQ2). It positions itself as the first empirical, task-level account of GenAI in German REM, complementing the U.S. sentiment study of Kriegbaum et al. [22]. Three contributions are claimed: (1) a grounded map of current and potential use cases, (2) evidence that interaction is predominantly human-in-the-loop, (3) identification of adoption constraints around listings/documents, data availability, and compliance. The single-country scope is justified institutionally (land register excerpts, development plans, notarial closings, GDPR/EU AI Act). Claims are calibrated modestly throughout: the authors speak of a *grounded map* and *tentative* implications and explicitly decline theoretical saturation (§3).
The remaining sections cover Methodology Assessment, Statistical Evidence, Results & Reproducibility, Limitations & Risks, and Verdict & Recommendations.
- All 5 strengths & 5 weaknesses, analyzed in detail
- 4 risk flags with severity, section & reviewer questions
- Methods & statistical evidence assessment
- 5 questions a peer reviewer would ask
- Full six-dimension evaluation report & recommendations
One-time purchase · this paper only
One-time purchase, this paper only. AI-generated preliminary analysis — not peer review.
Frequently asked questions
How many real estate agents were interviewed for this study?
Eleven German real estate agents, recruited via convenience and snowball sampling between June and August 2025, with a mean interview length of 29.6 minutes. All were active AI users by design, and the sample averaged 11.2 years of industry experience with 6-50 closed transactions annually.
What is the most common generative AI use case among German real estate agents?
Writing exposé (property listing) texts. The paper reports that all other identified use cases - image editing, social media posts, floor-plan visualization, acquisition aggregation, screening - remain emergent and unevenly adopted across agents.
Can this study tell me how many agents in general use generative AI?
No. The authors state the study is descriptive and exploratory; they did not examine non-users, so it cannot explain resistance or adoption barriers. Percentages in Figure 5 are explicitly descriptive, not representative adoption rates.
Is the paper's data or code publicly available?
No data availability statement, interview guide, codebook, or transcripts are reported as shared. The taxonomy is described in enough detail (first-order concepts with agent IDs) to support qualitative replication with a new sample, but the raw material is not reproducible from code.
About this review
This is an AI-generated preliminary review by Evolve.Pub, produced from the openly available text of the paper. It is not peer review and does not represent the authors' views. The canonical record of the paper is at the publisher link above.
- Figures 1-5 and Table 1 are flattened into running text by PDF extraction; figure topology was reconstructed from the surrounding prose, which is fully present.
- No truncation detected: the provided text ends with the complete reference list ([1]-[37]) and the version line, so no end-of-paper content is missing.
Go deeper on this paper
Wrote a paper? Stress-test it before reviewers do.
Upload your manuscript for the same structured scrutiny — reasoning, statistics, evidence, figures, and reproducibility.