Using Market Data and AI Rate Suggestions to Price Self-Storage

I was skeptical of dynamic pricing for a long time.

My reference point was airfare, where the price jumps $80 because you looked at it twice. That's what "algorithmic pricing" meant to me: software built to find the moment a customer is desperate. I didn't want to run a storage facility that did that to a woman moving her mother into assisted living.

The sales pitches didn't help. Every revenue-management deck I saw promised a revenue lift and got vague when asked how. When you pushed, the method was usually to advertise a low street rate, get the tenant's things into the unit, and then raise the rent step by step. Tenants hate surprise increases more than almost anything else in this business. The alternative to paying is renting a truck and spending a Saturday moving boxes they haven't opened since 2019. Pricing that depends on that hassle is closer to a toll booth than a strategy.

That was my view five years ago. Part of it was wrong.

Stale rates are the real problem

Ask an operator how they set rates and you'll usually hear something like this. They drove past the competitor, read the banner, and picked a number ending in 9. They might revisit it in a year. If a tenant complains, they take five dollars off.

The number is usually close on the day it's set. The problem is that it goes stale without anyone noticing. The REIT facility two miles away drops its 10x10 web rate on a Tuesday and adds a first-month promo. You find out in March, when move-ins dry up and you can't tell why. Your rate was reasonable when you set it, but you learned about the competitor's change months late.

That changed how I thought about it. Using data to squeeze existing tenants is still a bad idea, and I'd still tell you not to do it. Using data so you aren't the last operator in your market to see a rate change is ordinary good management.

What the data looks like

The market dataset behind StorageFlow tracks about 34,700 facilities. That's 7.4 million live price points and 5.2 million active promotions, refreshed daily and deduplicated across sources. Deduplication matters because the same facility often lists different rates on different channels, and you want one accurate rate instead of four conflicting ones.

How often the data updates matters more than how much of it there is. A snapshot from last quarter only tells you about last quarter. When a competitor starts a price war, you want to see it the same day. At that point, you can still decide whether to match it, undercut it, or wait it out.

Companies like StorTrack sell comparable market data for $300 to $1,000+ per month, for each market. An operator in three markets pays that three times. We include it in StorageFlow at no extra charge, including on the free tier, because we think every operator should be able to see their market. The base paid plan is $39 a month, and that's all I'll say about pricing here.

Most rate mistakes come from learning about a competitor's change weeks or months late.

What I still don't trust it to do

I'm still skeptical about some of it, so here are the limits.

AI suggests a rate, and you decide whether to use it. The product is built to work that way. The model doesn't know your gate motor is failing. It doesn't know half your climate units rent through word of mouth from the church down the road. A rate suggestion with the comps attached is information for you to weigh.

Airline-style pricing is a choice. Good data doesn't force you to reprice hourly or hit existing tenants with surprise increases. Operators who do that would do it without the data. Data only makes your existing pricing approach better informed, including a conservative and fair one.

It won't fix a bad facility. If your doors are rusted and your reviews say "couldn't reach anyone for a week," a well-tuned street rate won't help much. Fix the callback problem first. We have software for that too, but it's a topic for another post.

Where to start

Don't start by repricing everything. First, benchmark your current rates against live comps and find the unit type that's furthest from market in either direction. Often the problem unit turns out to be overpriced, like a 5x10 priced $20 above anything nearby. It sits vacant and costs more than a rate increase would bring in. Change that one rate and watch occupancy for a month. After that, move to the next one.

The software reads 7.4 million prices every day so you don't have to drive past the competitor's banner. The decision stays with you.

If you want to see your own market this way, it's in the dashboard from day one, free for a single facility.