National self-storage move-in rates landed at $96.44 in Q4 2025, down 10.7% year-over-year, according to Yardi Matrix data. That figure is near pricing levels last seen in 2016 and 2017. National occupancy sits at 77.0% at stabilized facilities, flat from a year ago. New supply equal to 2.5% of existing inventory is still under construction. The street rate market is not recovering in 2026.
But a segment of the industry is holding NOI steady anyway, not by filling units cheaper, but by pricing the tenants already inside more precisely than the market's conditions would otherwise allow. That is the real story of AI revenue management in 2026: not what it does to street rates, but what it does to the existing tenant ledger.
What Happens When the AI Turns to Existing Tenants?
Most conversations about AI pricing in self-storage focus on street rates: the advertised price that shows up on a Google search or a comparison site. That rate has been falling for 18 months. AI can optimize it, but it cannot create demand that isn't there.
The more consequential use of AI pricing right now is in-place rent management: systematic, data-driven rate increases on tenants already in the building. This is not a new concept. What AI does is make it precise enough to execute at scale without triggering the move-outs that erode the gains.
Cubix Asset Management published execution data from 2024 that shows what precision looks like at scale. The company executed 14,700 rate increases across its portfolio in a single year and recorded only a 1.7% move-out rate attributable to those increases. The dollar result was $160,000 in added profit. That is not a modeled estimate. It is a figure pulled from actual tenant behavior after actual rate increases were applied across a 50-plus property portfolio totaling more than 3.3 million square feet.
Our 2024-2025 results show that smart automation and data-driven pricing can drive NOI gains even when the broader market is flat.
- Sean Venezia, Partner, Cubix Asset Management
The low move-out rate is the proof of concept. In-place rent increases have historically been constrained by operator fear of vacancy: raise rates too aggressively and the tenant leaves, the unit sits empty, and the incremental revenue disappears. AI pricing does not eliminate that trade-off. It narrows it by segmenting tenants on dimensions that predict price sensitivity, and then executing only the increases calibrated to what each segment will absorb.
How AI Segments the Existing Tenant Base
Platforms like Veritec and Prorize analyze existing tenants across multiple variables simultaneously: tenure, payment history, unit type, unit size, competing options in the local market, seasonal timing, and occupancy in that specific unit category. The output is not a facility-wide rate increase. It is a recommendation set that distinguishes between the three-year tenant on autopay in a 95%-occupied unit type and the six-month tenant in a climate-controlled unit where the market has three cheaper alternatives nearby.
Veritec's Multiple Signal Modeling (MSM) system processes this kind of multi-variable analysis across more than 8,000 stores and 70-plus operators. The company reports 9 to 14 percent annual revenue increases for clients running its system, with the high end driven by its Value Pricing module, which adds a 10%-plus revenue layer through targeted in-place pricing recommendations.
The traditional approach to in-place increases was simpler: raise all tenants a fixed percentage annually, absorb the move-outs, accept that some increases were too aggressive and others too conservative. AI replaces the blanket approach with individualized recommendations. The result is more total revenue with fewer move-outs, because the increases that get executed are calibrated to what the specific tenant in the specific unit will actually accept.
Inside Self-Storage's reporting on AI pricing methods confirms the underlying framework: segmentation by tenure, payment consistency, and unit-type occupancy allows operators to push rate on low-risk tenants while holding rate on tenants where the move-out probability is high enough to make the increase a net negative.
What 10 Federal's Numbers Actually Show
10 Federal Storage's Q1 2025 results are the most complete public data set for what AI-driven operations deliver end-to-end. The company's 10FSSAC3 portfolio, covering 15 facilities and 713,458 square feet, posted a 9.0% year-over-year increase in same-store revenue and a 45.1% surge in net operating income, alongside a 12.8% reduction in operating expenses. Its 10FSSAC4 portfolio added 11.9% same-store revenue growth and a 42.9% NOI increase.
Those numbers do not come from a rising market. They come from a company that has automated significant portions of its revenue management function, including a dedicated Chief AI Officer role covering voice agents, pricing algorithms, and predictive maintenance, along with an AI chatbot handling 80% of customer FAQs while cutting call-center staff by 25%.
The NOI figure is the one worth isolating. A 45% NOI surge in a flat-to-declining market means the gains are not coming from tailwinds. They are coming from the spread between revenue optimization and cost reduction, both executed through the same technology infrastructure.
The Rate Gap Between REITs and Everyone Else
January 2026 data from Yardi Matrix shows that publicly traded REITs advertised rents 7.5% lower than non-REIT operators nationally. That inversion reflects a deliberate strategy: large operators are accepting rate compression to protect occupancy during the supply overhang, betting that revenue per available foot on a full building beats revenue per available foot on a partially occupied one at higher advertised rates.
Independent operators cannot execute that strategy. They don't have the capital reserves or the portfolio diversification to absorb rate compression across dozens of markets simultaneously. What they can do is use AI pricing to hold rate on the tenant base they already have, reduce acquisition cost by improving conversion at current price points, and minimize the move-outs that would force them to replace tenants at today's lower street rates.
The gap between REIT and non-REIT advertised rates is partly a strategic choice and partly a technology advantage. REITs have run AI pricing infrastructure for years and have the data to model occupancy elasticity across thousands of units at a time. Independents who deploy comparable tools close the analytical gap, even when the balance sheet gap remains.
What the Move-In Rate Tells You About the Real Risk
The 10.7% drop in move-in rates year-over-year is a problem, but it is a bounded one. The operators who built their revenue models around continuously rising street rates are the ones exposed. The operators using AI to layer in-place revenue increases on top of a stable existing tenant base have partially decoupled their NOI from the street rate market.
That decoupling is not complete. High move-outs erode the existing tenant base and force operators back into the street rate market to replace departing tenants at rates that are 10% lower than they were a year ago. AI-driven in-place pricing only works if the increase recommendations are calibrated well enough to preserve occupancy while lifting revenue.
The Cubix result, 14,700 increases at 1.7% attributed move-out, suggests the calibration can get precise enough to matter even in a compressed market. For operators who have not yet built that precision into their workflow, the compounding cost of running static in-place increases while the street rate environment stays soft is growing every quarter.
The Numbers Worth Tracking
- National move-in rates: $96.44 in Q4 2025, down 10.7% year-over-year, near 2016-2017 pricing levels
- National stabilized occupancy: 77.0%, flat year-over-year heading into Q2 2026
- Cubix: 14,700 in-place rate increases in 2024, 1.7% attributed move-out rate, $160,000 in added profit
- 10 Federal's 10FSSAC3 portfolio: 9.0% same-store revenue growth, 45.1% NOI surge, 12.8% expense reduction in Q1 2025
- Veritec: runs pricing across 8,000-plus stores and 70-plus operators; reports 9-14% annual revenue increases
- January 2026: REITs advertised rents 7.5% lower than non-REITs nationally
- New supply equal to 2.5% of existing inventory remains under construction, sustaining street rate pressure
Precision Is the Hedge Against a Soft Market
Street rates will stay under pressure as long as new supply keeps delivering into a 77% occupancy environment. Operators waiting for the market to recover before executing revenue management are giving up ground every quarter. The operators using AI to work the existing tenant base at the precision Cubix and Veritec have demonstrated are generating revenue that has nothing to do with whether move-in rates go up or down next quarter.
The AI pricing story in 2026 is not about what the tools can do. That's been established for years. It's about whether the operator has deployed them deeply enough to capture the gains available inside the fence, not just at the door.
Sources
- Using AI to Set Your Self-Storage Rental Rates: Benefits, Challenges and Integration Guidelines, Inside Self-Storage
- Cubix Announces Unified AI-Powered Platform to Drive Next Phase of Performance in Self-Storage, PR Newswire
- Cubix Creates Unified AI Platform Integrating Self-Storage Technology Vendors, Inside Self-Storage
- 10 Federal Delivers Q1 2025 Results, Secures $100M Credit Facility, Modern Storage Media
- Veritec Revenue Management System, Veritec Solutions
- Self-Storage National Report, March 2026, Multi-Housing News
- Self-Storage Market Outlook, Yardi Matrix