ReviewDesk guide · By the Ingenious AI team · Published 14 November 2026 · Reviewed 14 November 2026
What must landlords and screening services disclose when screening is automated?
Generate a free adverse-outcome letter — no sign-up Build a free pre-use notice See pricing
The short answer. The applicant must know a screening tool will assess them, what it weighs — payment history, eviction records, income against rent — and what happens if the answer is no: specific reasons, a correction route for the wrong-file errors that plague housing data, and a human who can look again. Whether you screen in-house or buy it as a service, the decision about the applicant is yours, and so is the disclosure. ReviewDesk runs it as a housing workflow with the notice, letter and inbox connected.
What is automated housing-screening disclosure?
Automated housing screening is the use of tools to score or recommend on tenancy applications — a consequential decision in the Colorado frame and a significant one in California's. Disclosure duties attach to the process: pre-use notice of the tool and its factors, adverse-outcome explanation with principal reasons, correction of incorrect records, and human review. The duties sit alongside — not instead of — existing housing law.
Why does housing screening generate so many wrong decisions?
Because its data sources are the messiest in consumer decisioning. Eviction records attach to names, not people, and namesakes collide constantly. Court outcomes are recorded without their resolutions — a filing that was dismissed reads like a filing that was granted. Rental payment history is fragmented across bureaus and landlords. Income verification penalises the self-employed, the tipped and the newly arrived, whose money is real but irregularly documented. Any honest screening process starts from the premise that a meaningful share of its adverse data is wrong, and builds its correction machinery accordingly.
That premise explains the regime's shape: reasons specific enough to identify the offending record, correction routes that accept documentary proof, and humans who can distinguish a filing from a judgment. A landlord whose process cannot make those distinctions will make housing decisions on noise — and the applicants harmed will be disproportionately those least equipped to fight a form letter.
Who owes the disclosure — landlord or screening service?
The applicant experiences one decision, usually wearing the landlord's name. Screening services supply the tool and the file; landlords and agents make or adopt the decision. The safe operating model puts the landlord's name on the notice and the letter, with the service's role described honestly inside them, and the review route answered by whoever can actually change the outcome. Contracts should allocate the mechanics — whose workflow, whose inbox, how the service's factor data reaches the letter — but allocation cannot outsource the applicant's experience of being told.
Portfolio landlords running the same service across properties should standardise on one workflow per jurisdiction, with property-level branding where needed. The multi-business pattern is covered in the agency guide; the vendor questions that keep a screening service honest are in the vendor guide. What the applicant must never encounter is the shell game — landlord blaming screener, screener blaming data, nobody owning the answer.
What belongs in the housing pre-use notice?
The tool's involvement and its factor groups in plain housing language: rental payment history, court records, income relative to rent, tenancy history where used. The recommendation's role — does the service decide, recommend, or inform a person who decides? The data sources by category, so a wrong record can be traced. And the aftermath: reasons if declined, correction with proof, review by a person. Publish it in the application flow itself; rental applications are often completed on phones, in hallways, under time pressure, and a notice that requires a desktop treasure hunt has not been given.
ReviewDesk's free workflow covers one housing operation end-to-end, and the notice builder drafts the page from your description of the screening you actually run. Describe the screening you would defend in front of the applicant — if a factor embarrasses you in the drafting, the factor is the problem, as the scope discussion in the scope-test guide keeps insisting.
How specific must decline reasons be in housing?
Specific enough to identify the record. 'Eviction history' is a category; 'an eviction filing from March 2023 in the county record we received, shown without an outcome' is a reason an applicant can work with — they can produce the dismissal order and be re-assessed the same week. Income reasons should carry the arithmetic: verified income, the rent, the multiple the criteria apply. Criminal-record factors, where used, need the same discipline — the record, its date, its disposition as held — because disposition is where housing data most often lies by omission.
The letter generator enforces this shape by asking for the reasons per applicant from the screening record. Landlords adopting the workflow after years of form declines usually find the first week uncomfortable — writing real reasons takes minutes per case — and the second week revelatory, because half the disputes they used to dread simply resolve when applicants can see and fix the actual record.
How should correction and review run for applicants?
On housing timescales, which are brutal: an applicant declined wrongly this week loses the home to another applicant next week, and a remedy next month is a remedy for a different life. Triage corrections daily, verify documents fast — court dispositions and payment proofs are usually single-lookup checks — and where the home itself is gone, keep the corrected approval portable across your own portfolio and say so in the outcome letter. A person who loses a specific home to your data error but gains a clean, evidenced approval for the next one has been treated as well as the situation allows.
Review proper — where the data is right but the decision is challenged — needs a person with letting authority and the full tenancy picture, applying the standard in the human-review checklist. Overrides should be recorded with the same care as declines, because the file's integrity protects the next applicant as much as this one.
What records protect a landlord?
The sequence, on one timeline: notice version, application, screening factors as returned, the recommendation and the landlord's decision — recorded separately where they differ — the letter, corrections, reviews and outcomes, kept for three years per the records guide. The separation between the service's recommendation and the landlord's decision matters more than it looks: where a landlord overrules a decline and lets the home, or adopts a decline the service merely suggested, the file should show whose decision it was.
That record is also the answer to the portfolio question every housing provider eventually faces — are we declining on data we would not defend? Sample the letters quarterly against the screening returns. Where a factor generates repeated successful corrections, the factor's weight, source or verification step is the fix, and the fix belongs in the risk conversation, not just in individual case files.
Your five actions for automated housing-screening disclosure
- Put the landlord's name on the notice and the letter; describe the screening service's role honestly inside them.
- Name factor groups and data-source categories in the notice, published inside the application flow.
- Write decline reasons that identify the record — date, type, disposition as held — with income arithmetic shown.
- Triage corrections daily; where the specific home is lost to your error, make the corrected approval portable across your portfolio.
- Record the service's recommendation and your decision as separate events on one three-year timeline.
Worked example: the dismissed filing that read as an eviction
A fictional applicant — Priya Shah — applies for a two-bedroom home with a portfolio landlord in Colorado. Screening returns an eviction filing from 2023; the recommendation is decline. The generated letter identifies the record precisely: filing, county, date, 'shown without an outcome in the file we received'. Priya's correction arrives the next morning: the court's disposition — the case was dismissed when the landlord withdrew it, a fact the screening file never carried.
Verification takes one lookup. The re-assessment, on the corrected record, clears the remaining factors — income 3.2 times rent, four years of on-time payments — and the approval issues that afternoon, day two of the correction. The specific home is still available because the correction moved at housing speed. The screening service receives the case in the monthly data-quality pack, and the workflow gains a standing rule: filings without dispositions are flagged for human check before any decline. One applicant's paperwork improved the process for every applicant after her.
Frequently asked questions
Is tenant screening a consequential decision?
Yes — housing is named among consequential decisions in Colorado's framework and significant decisions in California's. Automated screening therefore carries the notice, explanation, correction and review duties those frameworks attach.
Who is responsible — the landlord or the screening company?
The applicant faces one decision, normally under the landlord's name. Allocate mechanics by contract — workflow, inbox, factor data for letters — but the notice and letter should carry the decision-maker's name and a review route answered by someone who can change the outcome.
What is the most common screening error?
Court records without dispositions — filings shown as if they were judgments — and namesake merges. Reasons that identify the exact record let applicants produce the disposition and get re-assessed quickly.
How fast should housing corrections move?
Daily triage, verification in days. Housing remedies decay fast because homes let to others; where the specific home is lost to your error, keep the corrected approval portable across your portfolio.
Is this guide legal advice?
No. ReviewDesk is templates and workflow tooling, not legal advice. Duties are summarised from the regimes in our research and sit alongside existing housing law; consult counsel for your situation.
Keep reading: How do notice and human-review duties apply to automated insurance underwriting and claims? · What notice and review duties arise when software helps make healthcare decisions about patients? · What must schools and universities disclose when software screens applicants or students? · Do you have to tell employees when monitoring or productivity software scores them? · How can a tenant applicant request human review of an automated screening decision? · All ReviewDesk guides · ReviewDesk home · Try the free desk · Free letter generator · Free notice builder · Pricing
ReviewDesk is built by Ingenious AI. Templates and workflow tools — not legal advice.