Choose staff-performed inspections when your team is already at the property, which is the normal shape of enterprise multifamily with on-site maintenance staff. Choose tenant-guided inspections when units are scattered across a metro and you are adding doors faster than you can add inspectors, which is the normal shape of single-family and small multifamily management. Add AI review to the tenant-guided model when the review itself has become the bottleneck: when someone on your team is opening every submission to judge damage versus wear, or when you want video evidence instead of photo checklists. Most growing portfolios end up mixing models, with tenant-guided capture on every door and staff visits reserved for exceptions.
This guide assumes you already know what the tenant-guided model is. If you do not, start with our hub explainer, What are tenant-guided inspections?, which covers the workflow, the term's history, and the category's mechanics. This page answers the next question: which model should your company actually run, and when does the answer change.
The decision in three questions
The whole choice compresses into three questions about your operation. Answer them in order; the first yes is usually your model.
1. Is your team already inside the building most days?
On-site maintenance staff, a leasing office in the lobby, units concentrated in a handful of properties. The marginal cost of an inspection is a flight of stairs.
2. Are your doors scattered, and is scheduling entry the real cost?
Single-family and small multifamily spread across a metro. Every inspection means notice, a drive, a calendar, and a tenant letting a stranger in.
3. Has reviewing submissions become its own job?
Someone opens every submission, judges damage versus wear, requests redos. That work grows one-for-one with doors, which is the exact scaling you adopted the model to escape.
The rest of this page is the evidence behind those three answers: who each model was built for, what each one costs as you scale, and where the legal posture differs.
The three models, profiled honestly
Staff-performed inspections
Best for on-site teamsA trained person from your team, or a hired inspector, walks the unit with a field inspection app. This is the incumbent model, and in its home segment it is not the losing side of anything. HappyCo (happy.co), the most established platform in enterprise multifamily, positions inspections as one part of a broader AI-led maintenance operation spanning maintenance ops, asset management, and asset evaluation, because in buildings with on-site staff, inspection is a maintenance workflow, not a scheduling problem. According to RentCheck's competitor comparison documentation (getrentcheck.com), HappyCo requires a minimum of 500 units with quote-based pricing, which tells you exactly the operator it serves.
Choose it when
- Maintenance or leasing staff are already on site daily
- The inspection needs trained judgment in person: habitability, code, insurance documentation
- You suspect lease violations a tenant will not photograph: unauthorized occupants, unreported pets, hoarding
- The tenant cannot or will not participate
Accept that
- Every inspection costs a person-visit: drive, keys, on-site time
- Occupied-unit entry carries notice requirements and tenant friction
- Coverage follows staffing: scattered doors get inspected less often than anyone intends
Tenant-guided, human review
Best for scattered doorsThe tenant captures through a guided app; a person on your team reviews every submission. This is the standard tenant-guided model, built for exactly the company the staff model underserves: the manager whose doors are spread across a metro and who is adding more. According to RentCheck's published pricing (getrentcheck.com), tenant-guided capture starts at $1 per unit per month, tenants are required to take photos inside the app so images arrive timestamped and unedited, and video capture is reserved for property manager inspections. The capture is cheap and defensible; the review is where your time now goes.
Choose it when
- Single-family or small multifamily, scattered across a metro
- You are adding doors and refuse to add inspectors one-for-one
- Routine documentation is the job: move-in, move-out, annual, mid-lease
- Deposit disputes keep coming down to missing move-in evidence
Accept that
- A person still opens every submission and judges damage versus wear
- Tenant capture is photos, not video, across the category
- Capture quality varies; redo requests are part of the workflow
- Specialist inspections still need a professional on site
Tenant-guided, AI review
Best when review is the bottleneckThe tenant still captures; software does the review, comparing the submission against that unit's own move-in baseline and reporting what changed. This is the emerging model, and the category's own tooling points at it: RentCheck's homepage (getrentcheck.com) now advertises AI damage detection whose stated job is to make sure critical issues are not missed and to prioritize which tenant inspections a manager reviews first. That is triage, a ranking of the queue a human still works through. Full AI review removes the queue: your team opens findings with evidence attached, not submissions. It also unlocks video as the tenant's capture format, because nothing has to be watched in real time by a person.
Choose it when
- Review volume has outgrown the person doing it
- You want each unit judged against its own baseline, consistently, every time
- You want video evidence of whole rooms, not photos of what a template asked for
- Growth plans assume inspections stop scaling with headcount at all
Accept that
- The model is newer than the other two; vet vendors on evidence quality
- A baseline has to exist before comparison pays off, so move-in capture discipline matters
- Staff visits remain the answer for specialist and exception cases
The cost logic: what each model scales with
Price lists make the models look closer than they are. The real difference is what the cost scales with as doors grow.
| Model | Cost shape | Public anchor | At 100 doors, inspected twice a year |
|---|---|---|---|
| Staff-performed | Per inspection. Drive time plus on-site time plus scheduling overhead, or a vendor invoice per visit. | Third-party rental inspections are sold as a service starting at $300 per inspection, per Hemlane's published pricing (hemlane.com). In-house visits cost less in cash and more in staff hours.~$300 / inspection outsourced | 200 person-visits a year. Outsourced, that is tens of thousands of dollars; in-house, it is a meaningful fraction of a full-time role before drive time. |
| Tenant-guided, human review | Per door for capture, per submission for review. The software is cheap; the review hours are the real line item. | Category capture pricing starts at $1 to $1.25 per unit per month, per RentCheck's published pricing (getrentcheck.com).~$1 / unit / month capture | Roughly $1,200 to $1,500 a year in software, plus a person opening and judging 200 submissions, requesting redos, and filing reports. |
| Tenant-guided, AI review | Per door, full stop. Neither the capture nor the review consumes staff hours on routine inspections. | AI-review vendors, RapidEye included, generally price by conversation rather than rate card; the comparison to run is per-door cost against review hours recovered. | Software cost, plus staff time only on the findings that need action and the exceptions that need a visit. |
The trap in the middle row is that the $1 per unit is the visible number and the review hours are the invisible one. At 30 doors, review is a coffee break. At 300, it is a role. The standard tenant-guided model moves the work twice, off your staff for capture and back onto them for judgment, and the second move is the one that quietly re-couples headcount to door count. That re-coupling is the specific thing the third model exists to remove.
The legal posture: entry, notice, and evidence
Two legal considerations genuinely differ between the models: getting in, and proving condition afterward.
Getting in. A staff inspection of an occupied unit is a regulated entry. Under California Civil Code Section 1954, a landlord must give written notice stating the date, approximate time, and purpose of entry, twenty-four hours is presumed reasonable notice, and non-emergency entry must happen during normal business hours unless the tenant consents otherwise. Every state draws these lines differently, but the operational shape is the same everywhere: notice periods, business-hours windows, and a tenant who may not want a stranger in their home. A tenant-guided inspection has no entry to notice, because the occupant is the one holding the camera, on whatever schedule suits them. For routine mid-lease condition checks across a scattered portfolio, that single difference is most of the argument.
Proving condition. Deposit deductions live or die on move-in versus move-out evidence, and state documentation requirements increasingly assume itemized, dated condition records. Here the models converge more than vendors admit: a well-documented staff inspection and a guided, timestamped tenant capture both produce a defensible record. The category's anti-fraud mechanics carry the tenant side. According to RentCheck's competitor comparison documentation (getrentcheck.com), resident inspections require all photos to be captured directly within the app, with no uploads from the camera roll, so the timestamp and the camera are bound together. What differs is coverage: the staff model documents the units staff reach, and the tenant model documents every door, which matters when the dispute is about a unit nobody drove to. The move-out and deposit system you build on top matters more than which human held the phone.
Where staff visits stay mandatory. No capture model replaces a qualified professional for habitability complaints, code compliance, insurance claims documentation, or structural concerns. And a tenant will not photograph their own lease violation; if you suspect an unauthorized occupant or an unreported pet, that is a staff visit with proper notice, whatever your default model is.
Mixing models, and the signals it is time to switch
The clean version of this decision is a portfolio that runs one model. The real version is a default model plus exceptions. The most durable split we see: tenant-guided capture as the default on every door for move-in, move-out, and mid-lease documentation, with staff visits triggered by flagged findings, non-responsive tenants, specialist needs, and any building that has on-site staff anyway. The models are answers to different unit shapes, and most growing portfolios contain more than one shape.
You notice
The move
Mid-lease inspections keep slipping because scheduling entry is nobody's favorite job, and some doors have not been seen in over a year.
Staff-performed to tenant-guided
You are quoting growth plans in inspector hires: every 150 new doors implies another person driving a route.
Staff-performed to tenant-guided
Submissions sit unreviewed for days after tenants complete them, and approvals happen in batches when someone finds time.
Human review to AI review
Two reviewers would call the same scuff two different things; damage-versus-wear calls vary by who opened the submission.
Human review to AI review
Deposit disputes turn on what a photo template never asked for, and you wish you had the whole room on record.
Photos to video, which requires AI review
You acquired a concentrated building with on-site maintenance staff.
Run staff-performed there; keep tenant-guided on scattered doors
The last signal in that table deserves its own sentence, because it is the one the category talks about least. Video is the better condition record: it captures the room, not the checklist's guesses about the room. The reason tenant capture is photo-only across the standard tenant-guided category is not that video is worse evidence; it is that a human reviewer cannot afford to watch it. Once the review is software, that constraint disappears, which is why video walkthroughs and AI review arrive as a package rather than as separate upgrades. This is the model RapidEye runs for long-term rentals: the tenant records a video walkthrough on their phone, RapidEye reviews it against the unit's own move-in baseline, and your team gets a report of what changed. Details on the long-term rentals page.
Quick FAQ
Should tenants do their own rental inspections?
For routine move-in, move-out, and mid-lease condition documentation in scattered single-family and small multifamily portfolios, yes. A guided app produces timestamped, in-app-captured evidence without a site visit, and your team reviews remotely. Staff still belong on site for specialist inspections, suspected lease violations, and buildings where a team is already present every day.
When are staff-performed inspections the right choice?
When your team is already at the property, which is the normal shape of enterprise multifamily with on-site maintenance staff, and when the inspection needs trained judgment in person: habitability and code issues, insurance documentation, suspected unauthorized occupants or pets, or a tenant who cannot or will not participate. Field inspector platforms exist because that job is real.
What does each inspection model cost?
They scale differently. A staff or vendor visit is a per-inspection cost: third-party rental inspections are sold starting around $300 each, and in-house visits cost drive time plus on-site hours. Tenant-guided capture is priced per door, starting around $1 per unit per month in published category pricing, but a person still spends review time on every submission. AI review keeps the per-door price shape and removes the per-submission review time.
Can you mix inspection models across one portfolio?
Most growing portfolios should. The durable split is tenant-guided capture as the default on every door, with staff visits reserved for exceptions: flagged findings, non-responsive tenants, specialist inspections, and buildings with on-site teams. The models are answers to different unit shapes, not competing religions.
What changes when AI reviews tenant-guided inspections?
The review stops scaling with door count. In the standard model a person opens every submission and judges damage versus wear, so review headcount grows with doors. With AI review, software compares the tenant's capture against the unit's own move-in baseline and reports what changed. It also makes video practical as the tenant's capture format, because nobody has to watch it in real time.
Sources
Sources are named at the publisher level with their root domain, rather than linked or titled; every figure is verifiable at the named source.
- Published pricing, competitor comparison documentation, and homepage, RentCheck, 2026getrentcheck.com
- Multifamily maintenance platform homepage, HappyCo, 2026happy.co
- California Civil Code Section 1954, California Legislative Informationleginfo.legislature.ca.gov
- Published pricing for third-party inspection services, Hemlane, 2026hemlane.com