Skip to content

Cold-Start Fill Playbook

A picture of success (working backwards)

A new building, opened in a market where nobody knew the name, is at 90% occupancy within 90 days of its doors opening. Bands found us because they were already looking for a real place to practice and we were there to meet them, or because an ad put us in front of them. They came in to see the room, some to play a free session in one, and the community manager closed most of the tours into signed lockouts. Ninety percent is the number that lets the building refinance on schedule. From there the market starts to run itself: the building stays full, a waitlist forms, and the next building opens into demand that already exists. This playbook is about the first building, the one the flywheel cannot fill for you.

What this is

A market-entry playbook with one job: take the first building in a cold market to 90% occupancy in 90 days, the threshold the financing model needs. Everything after that (a full building, a waitlist, the next building) is the proven Portland model and is out of scope here.

It matters because that first 90 days is the only part of our growth that marketing has to manufacture. In Portland we never marketed. The business grew one building at a time off a paid waitlist: a building fills, the waitlist grows, and when it is deep enough the next building opens and fills almost instantly from demand that already exists. Nine buildings, no ads. But that flywheel needs a full building feeding a waitlist to turn, and a brand-new market has neither. The first building, in a place with no brand and no waitlist, is the gap. Salem is the first time we have had to fill one cold; St. Louis and Seattle are next.

The thesis

In a cold market you fill the first building by generating demand fast, getting bands in the door for a first visit, and letting the visit do the converting. Paid and high-intent search are the demand engine, because they work on a clock and a brand-new market has no scene reputation or waitlist to draw on. The first visit is a tour or a free session: most bands can tell from seeing the space and getting the vibe whether it fits, and the community manager closes about 80% of tours. Because the close rate is already that high, the game is getting more qualified bands in the door, not converting them once there. Community, scene reputation, and the waitlist are what the market runs on afterward, and the CM builds them from day one, but they compound too slowly to fill the first building in 90 days, so we do not ask them to.

Plainly: paid generates the demand, a tour or free session gets them in the door, and the visit closes them, currently about four in five. The constraint is visit volume, not conversion. Community is the engine for the market's second act, not its first.

The engine and the supporting cast

There are only a few scalable ways to acquire customers, and the discipline is to commit, not spread. For the 90-day cold-start fill, in order:

The demand engine: high-intent capture and targeted paid. Bands actively searching for a place to practice are the highest-intent demand there is, and search (paid and organic) converts them; this is where our own results are already strongest. Targeted paid reaches the rest of the findable niche before they go looking. This is the lever that moves the number on a clock, and it is where the budget goes first.

The first visit, where it converts: a tour or a free session. The visit is the hinge of the funnel, and the CM closes about 80% of tours. Two on-ramps get a band there. A tour is the lower-barrier one and the current driver: most bands can tell from seeing the space and feeling the vibe whether it fits, so they do not need to prove it by playing. A free session (our settled 50-free-credits offer: a real-room play, a lockout-sized room with a PA) is the higher-engagement alternative for bands who want to play before they commit, and it promotes the hourly product too. The two can co-exist, one may outperform the other, or one may replace the other. What gets bands in the door, and what they do once there, is a core experiment, not a settled answer.

The closer: the community manager. Every location has one (Matador in Salem). The CM runs the place, the tours, the sales, and the local relationships, and closes about 80% of tours. Demand generation is the engine's job; conversion is the CM's, and it is already strong. This split matters: our CMs have always been closers and community-keepers, because Portland's demand was baked. A cold market is the first time the demand has to come from somewhere else, which is why the whole game here is getting bands in the door.

Built in parallel, not the 90-day engine: community and referral. Scene reputation and word-of-mouth are how an established market sustains itself, and the CM builds them from day one. They compound late, so we do not time the fill on them. Referral specifically is an open question, not a dead end: the give-50/get-50 reward is new and was never actually marketed, so we have no read on it. It is worth a real test, not reliance.

The hourly product is a separate motion. The same free session seeds it, but hourly is more individual; track it separately and do not assume lockout lessons transfer either way.

North Star, inputs, and guardrails

The metric the playbook serves is lockout move-ins per month, on pace to 90% by day 90.

Role Metric
North Star Lockout move-ins / month, on pace to 90% by day 90
Input First visits booked (tours and free sessions)
Input Demand-to-visit rate (leads and claims that become a booked visit)
Input Visit-to-lockout close rate (tours ~80% today)
Guardrail Junk-lead rate
Guardrail CAC vs ceiling (generous; LTV is high, ~$16K in Portland, set per market)
Guardrail Member durability / early churn
Candidate to validate Referral / word-of-mouth share (efficacy untested)

The number

"Fill it" becomes a target only as arithmetic. Work backwards: studios to reach 90% by day 90 → lockout move-ins per month → first visits per month (at the ~80% tour close) → demand per month (leads and free-session claims that become visits). Current pace versus required pace is the gap, and the gap is the brief. Each market runs its own back-calc in its instance, from analytics.occupancy_monthly and its own measured rates, not in this document.

The experiment loop

We move the inputs through a weekly cycle: analyze, generate ideas, prioritize by ICE (impact, confidence, ease), run one or two, review against the North Star.

  • Fixed envelope. The per-building monthly budget (set per market) is reallocated, not grown. Test by splitting or shifting spend.
  • One or two levers at a time. Low volume cannot read many changes at once. ICE-rank and run the top one or two.
  • Decision-rules, not p-values. At a handful of tours a month, significance is unreachable, so pre-commit to a minimum count plus a directional threshold, and hunt step-changes. With LTV this high, cost is rarely the kill criterion; tour and move-in volume is.
  • Explore, then exploit. Early, run for volume to learn what converts, with downstream cohort measurement as the guardrail. Tighten toward quality as the converting segments appear. Volume-first is a learning phase, not a setting.

The first thing to test is the on-ramp itself: tour versus free session versus both, what gets bands in the door and what converts. The 80% tour close says conversion is not the constraint, so most experiments should target visit volume at the top of the funnel.

Log every run: hypothesis, lever, the input it targets, ICE score, the metric and threshold, result, decision.

Measurement substrate

Nothing above is repeatable without it, and the list is finite.

  • Qualified-lead is a learning lens, not a gate. Bias to volume, screen only obvious junk, and learn quality downstream from what converts. Do not throttle the top of the funnel.
  • Attribution made real. Leads counted by source and campaign; tours made attributable; conversions linked back to the ad or free session that started them.
  • The scorecard. One running view per building: spend → first visits (tours and free-session claims) → lockout move-ins, with the tour close rate tracked and band claimers kept distinct from drop-in hourly. It is the budget justification (spend continues while it is producing) and the readout every experiment reads from.

FAQ

Internal

  • What is the win? 90% occupancy by day 90, within the budget envelope. The waitlist and the flywheel are the next goal, not this one.
  • How do we know it is working? The scorecard. Results gate the spend.
  • Who does what? Paid and capture generate demand; the free session lowers the barrier; the CM converts and builds local community for the phase after.
  • What if the first building misses 90 in 90? The readout tells us which input is the constraint, and that lesson is what the next market inherits.

External (the band's view)

  • What do I get? A real, private room that is yours, on your schedule. Come see it, or play a free session in a room just like it.
  • Why not the garage or where we practice now? Named plainly, without running down the alternative.

Traps

  • Expecting the Portland flywheel to fill the first building. It cannot; there is no full building or waitlist yet. That gap is the whole reason this playbook exists.
  • Treating community or referral as the 90-day engine. They compound late, and referral is untested. They are the second act, built in parallel, not the fill.
  • Asking the CM to be the demand generator. The CM closes and builds community; the engine generates the demand.
  • Optimizing conversion when it is not the constraint. Tours already close around 80%; the leverage is visit volume, not the close rate.
  • Chasing cheap CPL. We are not CAC-constrained; cheap junk volume corrupts the signal and the scorecard.
  • Treating skepticism as a universal. It may appear in small, guarded scenes and may not; it is not a Portland trait and not a foundation. Confirm per market before designing around it.
  • Scaling volume without downstream measurement, which poisons the signal.

Portability

What makes this a playbook and not a one-off is the split between what is constant and what is set per market.

Constant: the objective (first building to 90% in 90 days), the thesis, the engine and supporting cast, the North Star with its inputs and guardrails, the loop, the scorecard, and the back-calc method.

Per-market parameters (in each market's instance, not here): target date, budget, market depth and tier, brand-cold baseline, available channels and local partners, and whether scene skepticism is real there. Instances live in cold-start-instances/, filled from the template.

The exit. A market leaves this playbook when its first building is full and a waitlist has started. From there it runs on the proven Portland model, and marketing throttles down.