Every application arrives already reviewed
A household summary, an affordability check against the landlord’s threshold, and a document-by-document cross-check of what was stated against what the payslips and statements actually show — so a reviewer sees exactly what to chase, without opening a single PDF first.
- A plain-language summary of the household, ready on open
- Affordability worked out against the landlord’s own threshold
- Every payslip and statement read and cross-checked against the application
- A recommended next step — never a score, never a decision
The reading is the slow part
A stack of PDFs to open
Payslips, bank statements, references, IDs — for every occupant. Reading each one, then cross-referencing it against what the applicant typed, before you can say yes.
Figures that don’t quite match
The stated income is €4,950. The payslip says €4,870 net. Is that fine? Is the statement three months old? Easy to miss at speed.
Documents in another language
A Portuguese savings statement is still evidence — but only if someone can read it and reconcile the numbers.
Every application arrives already reviewed
Open a file and the AI review is waiting: who the household is, whether the rent is affordable, and how the stated income compares with what the documents actually show. A single recommended next step sits at the top — request documents, ask for a guarantor, ready to review.
- Plain-language household summary
- “Worth checking before deciding” notes
- A next step, not a verdict
Rent-to-income, worked out for you
Deterministic maths, shown the instant you open the file — before the AI pass even finishes. Rent against the whole household’s assessable income and the landlord’s threshold, with ±€100/€200 scenarios and a guarantor calculation when the ratio is over the line. Savings from the bank statements count as a positive factor.
- Assessable income = occupants in full + 50% of any guarantor
- ±€100 / ±€200 scenario rows and a max-affordable-rent figure
- Savings shown as “≈ N months’ rent” covered
Every document read, quoted and reconciled
The AI reads each payslip, bank statement and reference, quotes the figures printed on them, and flags where they disagree with the application — “stated €4,950 · payslips show €4,870 net/mo −1.6%”. A status dot and a one-line finding per file, grouped by occupant.
- Classified by content — catches a contract uploaded as a “payslip”
- Stale documents (over three months old) flagged automatically
- A discrepancy puts a badge on the Documents tab
A non-English document is detected and translated
The AI detects the language, translates the key content into a two- or three-sentence finding, and still reconciles the figures against the application. The language shows as a chip on the file.
- Detected automatically — no setting to toggle
- Figures reconciled, not just translated
- Shown inline on the document tile
Let the AI draft the reply
An approve / decline / request-documents / request-references / request-guarantor / follow-up intent writes a first draft that pulls the specific outstanding items straight from the review. The agent edits and sends — nothing goes out on its own, and a decline draft is barred from citing anything like a protected ground.
- The draft names the exact documents still needed
- Saved templates too, scoped to the company, client or property
- Agent reviews and sends every message
RTB and bankruptcy checks, with AI triage
Queue a check against the RTB tenancy-dispute register and the Irish bankruptcy (CSOL) records. Because those registers are searched by name, a hit is often a namesake — so the AI triages each match as likely-same-person, likely-different-person or unclear, with a reason, and staff record the outcome.
- RTB dispute register + CSOL bankruptcy records
- AI triage of name matches, with its reasoning shown
- Runs in the background — never blocks the application view
The AI advises. A person decides.
Screening touches GDPR Article 22 and the Equal Status Acts, so the review is built to stay firmly on the right side of both:
- Never scores, ranks, or recommends an approve/decline
- Barred from mentioning or reasoning about any protected ground — nationality, age, family status, HAP, disability, religion, civil status, race, Traveller-community membership — including inside an uploaded document
- Nationality, date of birth, children, HAP status and PPS number are never sent to the model
- No automation can change an application’s status from an AI value
- AI findings are purged on your company’s retention schedule
How a file moves through
- 1
The applicant builds their profile
One profile per household member in the portal, with documents uploaded and co-applicants or guarantors invited.
- 2
The review runs on submission
A single AI pass reads the household and its documents. The affordability maths is computed live and shows immediately.
- 3
You open a briefed file
Summary, affordability, per-document findings and a recommended next step — you confirm the figures against the documents yourself.
- 4
You decide, and reply
Approve, decline or request more — with a template or an AI-drafted email that already knows what’s outstanding.
Ready to make Real Enquiries your operating system?
Join the waiting list for pilot pricing, onboarding support and the launch-partner roadmap.