Ireckonu, a provider of hospitality technology, has warned that rapid adoption of artificial intelligence across hotels will deliver limited benefits unless organisations first address fragmented guest data and legacy system design.
AI’s promise undermined by legacy architectures
Speaking on behalf of the company, chief executive Jan Jaap van Roon argued the underlying problem is not a shortage of AI tools but the quality and structure of the data those tools consume. He said most hotels hold guest information across multiple platforms, and many property management systems (PMS) were built to manage inventory — rooms and reservations — rather than build a holistic digital profile of a guest.
“For decades, the PMS has been the operational backbone of hotels, connecting an ever-growing ecosystem of technology platforms. But most PMS architectures were never designed to understand guests. They were built to manage inventory and room nights”
That fragmentation, the company warns, can produce duplicate entries, inconsistent records and a failure to recognise loyal customers. In practical terms, a frequent visitor may be treated as a first-time guest, undermining personalisation and customer experience, while generic communications may be sent to those who expect tailored service.
Where AI projects can go wrong
Hanna Schiller, Ireckonu’s vice-president of sales, said the sector is accelerating investment in AI — from virtual assistants to automation and predictive personalisation — but remains “not data-ready”. The core contention is that applying machine learning and automation to disconnected or poor-quality datasets risks amplifying errors and producing outcomes that damage brand reputation rather than improve service.
- Examples of AI deployment cited include virtual assistants, service automation and predictive personalisation.
- Primary data challenge: guest records spread across multiple systems not designed to be joined into a single identity.
- Potential negative outcomes: duplicate profiles, inconsistent records, failure to recognise loyalty, diluted personalisation.
The argument reframes a familiar technology debate: tools are only as effective as the data that feeds them. For hotels, that means the pathway to valuable AI is less about procuring more models and more about rebuilding or reconfiguring data flows to produce a trusted guest identity.
Practical implications for operators and suppliers
The warning has immediate practical implications. Hotels that rush to deploy AI features without resolving underlying data quality and integration will risk customer dissatisfaction and possible reputational harm. Conversely, those that invest first in consolidating and governing guest data may extract stronger, safer returns from later AI investments.
| Data issue | Likely consequence for AI |
|---|---|
| Fragmented guest records | Duplicate profiles; loss of loyalty recognition |
| PMS designed for inventory | Insufficient guest context for personalisation |
| Disconnected systems | Inconsistent communications; brand risk |
Technology vendors and hoteliers have debated integration and data governance for years; Ireckonu’s intervention stresses that these are not optional preconditions for AI but essential infrastructure. The company frames the solution as establishing a single, reliable digital profile for each guest so that AI-driven features can operate on authoritative, consolidated information.
That approach is consistent with practices in other consumer-facing sectors where identity and data quality underpin personalisation and regulatory compliance. For the hospitality industry, however, the legacy of multiple, specialised systems — each optimised for reservations, billing or guest services — presents a particular challenge.
As hotels continue to trial chatbots, personalised offers and automated service workflows, the industry will need to decide whether to prioritise rapid feature rollouts or to slow deployment until core data problems are resolved. According to the source material, Ireckonu believes the winner in the AI arms race will be those that secure trusted data, not those who merely accumulate more technology.
