When a Tenant-Screening Report Becomes a Housing Decision
This series examines RentGrow, Inc.’s tenant-screening reporting business and the relationship among the company, rental applicants, landlords, property managers, and the other reporting agencies that supply data. The central question is simple but consequential: when a screening report is wrong, where did the error enter, and who must make it right? The answer requires more than asking whether a record exists. It requires looking at how records are matched, grouped, described, challenged, updated, and communicated to the person making a housing decision.
A complaint filed by the United States in July 2026 alleges that RentGrow’s ScreeningWorks Pro reports combined information such as credit data, criminal records, eviction records, rental history, and historical addresses. The reports were sold to rental property owners and management companies to help determine whether an applicant met, conditionally met, or did not meet rental criteria. The complaint is an allegation, not a final judgment. But it gives a concrete case through which to understand a broader point: automation does not make the decisions inside a reporting pipeline disappear.
The report is part of a housing decision
A tenant-screening report can look like a document about an applicant, but its practical role is relational. RentGrow assembles and evaluates information; a landlord or property manager uses the resulting report as a factor in deciding whether to rent; and the applicant must live with the consequences of the report, often without seeing every input that shaped it. Upstream consumer reporting agencies, public-record vendors, and data sources sit behind the report. A change at any one point can alter what the landlord sees.
That is why the phrase “background check” can be misleading. It suggests a search that simply retrieves facts from a government database. The complaint describes something more involved: information is gathered from multiple sources, combined, and used to produce an eligibility assessment. The CFPB’s advisory opinion on background screening makes the same basic distinction at the industry level. Screening companies use different identifiers, databases, matching criteria, and updating procedures. The report is therefore a constructed account of a person’s history, not a transparent window onto reality.
The construction does not mean that every report is arbitrary. It means that accuracy has several dimensions. A report can contain a real court filing but describe it without its disposition. It can find a record that belongs to someone else. It can count one proceeding more than once. It can retain information that is no longer legally available to the user of the underlying government record. Or it can tell the applicant that an update was sent to the landlord when the landlord receives only a message that the screening result did not change.
The basic relationship looks like this:
The diagram is not a claim that every RentGrow report passed through every failure point. It is a map of where meaning can change in the kind of reporting relationship described by the complaint and the CFPB. The applicant’s information may be accurate at the point of collection and still become misleading after aggregation or presentation. A correction may occur inside the reporting system and still fail to reach the decision-maker. A landlord may receive a screening result without knowing which part of the report produced it.
Accuracy is more than finding a matching entry
The most intuitive test for an automated report is often: did the system find a record? That test is too weak for housing screening. The more useful questions are: Did the record belong to this applicant? Does it represent one event or several? What happened after the filing? Is the information still reportable? Can the applicant identify its sources? If the record was corrected, did the correction reach the people who relied on it?
The complaint alleges that RentGrow reported duplicate criminal and eviction information in ways that made applicants appear to have more records or a longer eviction history than they actually had. The CFPB’s guidance explains why that matters. Different stages of one court proceeding—such as a filing and a judgment—may appear in different data sources. If they are displayed as separate proceedings, the report can imply that several events occurred. The problem is not merely that the page is untidy. The count and arrangement of entries can change the apparent severity of a person’s history.
The complaint gives the mechanism a human scale. In one disputed report, three displayed eviction events allegedly represented two proceedings because two events belonged to the same case. In another, six listed records allegedly reduced to two once duplicates were recognized. A separate example described fifteen criminal charges across seven records, with several entries duplicating one another; the complaint says the applicant was initially found not to meet the landlord’s criteria and later accepted after the records were removed. These examples remain allegations, not a representative sample of RentGrow’s reports. Their value is more precise: they show how a system can change the apparent number of events without inventing a new court record. The landlord’s decision may respond to the displayed count, even though the underlying history has not changed.
The possible consequences are also specific. The complaint alleges that inaccurate reporting can prolong a housing search, increase application fees, lead to higher rent or security deposits, or result in denial of housing. The CFPB describes similar consequences more broadly, including restricted housing choices and the time and expense of correcting information. These are not proof that every applicant in the RentGrow case suffered each harm. They explain why the accuracy question cannot be separated from the decision that follows the report.
This is also why source visibility matters. The complaint alleges that RentGrow used LexisNexis Accurint address-history and name information to help locate public records, but did not identify that source in certain file disclosures. Additional identifiers may help a screening company find a record that an application alone would not reveal. They can also create an unseen dependency: an applicant may be disputing a record without knowing which address, name, or other detail helped produce the match. A system that can find more information needs a meaningful way for the person affected to understand and challenge the basis of what it found.
Source disclosure has a practical purpose beyond giving a report a longer list of vendors. The complaint describes the FCRA file-disclosure requirement as a way for a consumer to identify an inaccuracy at its source and reduce the chance that the same error will be supplied again. If a key identifier or data provider is omitted, the report may be internally traceable to the company that assembled it while remaining difficult for the applicant to correct. That is the difference between an explanation available somewhere in a system and an explanation available to the person whose housing application depends on it.
The legal question is a chain of responsibilities
The FCRA treats the role of a consumer reporting agency as more than a technical service. The complaint alleges that RentGrow regularly assembled and evaluated information for consumer reports and acted as a reseller by combining information from other reporting agencies. On that account, the relevant duties include reasonable procedures for maximum possible accuracy, disclosure of information and its sources when a consumer requests a file, and specific procedures for handling disputes.
Those duties do not turn the complaint’s allegations into established facts. They do show why “the data came from a vendor” is not a complete answer. A reporting company may depend on upstream information while still having responsibilities for how it assembles, displays, discloses, and routes that information. The CFPB’s advisory opinion likewise describes background-screening accuracy as a continuing procedural obligation. It highlights preventing duplicate entries, including available dispositions, and excluding records that have been expunged, sealed, or otherwise legally restricted from public access when the user could not obtain them directly from the government recordkeeper.
This chain of duties follows the information rather than stopping at the first database. An upstream agency may furnish a record; RentGrow may merge it with other information and turn it into a screening report; a property manager may use the report as one factor in an eligibility decision. Each role is different, but the applicant encounters the combined result. A dispute must therefore do more than ask whether a vendor supplied a line item. It must ask whether RentGrow’s own assembly or presentation introduced the problem, whether the source agency needs to investigate it, and whether the eventual correction will be visible to the person making the decision. Those are accountability questions even when no one person wrote the underlying record.
The practical issue, then, is not whether automation is allowed to assist with tenant screening. The issue is whether the people and institutions operating the process can account for the transformations between source and decision. A model, matching rule, vendor feed, layout choice, dispute label, or notification template can all affect the result. Calling the process automated does not identify which choice caused an error, and it does not provide the applicant with a remedy.
What the documents establish—and what they do not
The RentGrow complaint supplies a detailed set of allegations: duplicate records, incomplete source disclosures, disputed items treated as invalid, and statements about landlord notification that allegedly did not match what property managers received. It also describes the statutory theories and asks for an injunction and civil penalties. Those allegations are valuable evidence of the mechanisms regulators say they are challenging, but they are not a final adjudication.
The CFPB document serves a different function. It is an official advisory opinion interpreting existing FCRA requirements for background-screening reports. It helps explain why duplication, missing dispositions, legally restricted records, and obsolete information can be accuracy problems. It does not independently prove that RentGrow committed every act alleged in the complaint, nor does it establish how common such failures are across the industry.
That distinction is part of the subject, not a footnote to it. A defensible account of tenant-screening automation must preserve both the report’s influence and the evidence’s limits. The central judgment this series supports is therefore practical: before trusting a screening output that may affect housing access, ask where the information came from, how it was matched and grouped, what its current status is, how a dispute moves, and whether a correction reaches the decision-maker. The next question is how the report’s internal roles and legal classification make those checks possible—or make their absence consequential.