How Much Should a Night Venue Spend on Advertising?

"What should I be spending on advertising?" gets answered in this industry with a percentage of revenue. Five percent, eight percent, whatever the person you asked happens to spend.

That answer is worthless, and not because the percentage is wrong. It is worthless because it is a single number applied to several channels that perform nothing alike. The venue spending eight percent is almost certainly overpaying one channel and underpaying another at the same time, and the aggregate figure hides both.

There is a real answer, and it is not a percentage. It is a ceiling per channel, in your currency, per month. This is how it is computed.

The number you are actually looking for

Not "what should I spend," but:

What is the most I can pay this specific channel next month and still make money on the guests it sends me?

Answer that per channel and the budget assembles itself. Channels below their ceiling get more. Channels above it get cut or renegotiated. There is no percentage-of-revenue rule anywhere in it, because there does not need to be.

The formula

Four quantities, then one line.

LTV       = average total revenue from a guest first seen via this channel,
            across their whole relationship with you
Margin    = the share of that revenue you keep after variable costs
CAC       = channel spend ÷ new guests attributed to that channel
Ceiling   = LTV × Margin × (1 − reinvestment reserve)

The ceiling is the most you can pay to acquire one guest through that channel. Multiply by the guests it can deliver in a month and you have that channel's monthly budget. If current CAC is below the ceiling, the channel is underfunded and you should buy more of it. If it is above, you are paying more to acquire a guest than that guest is worth to you.

Why LTV, and not first-visit spend

Almost everyone in nightlife evaluates advertising against what the guest spent the first night. That is the single most expensive mistake in the category, because it systematically underfunds the channels that bring people who come back.

A channel delivering guests at ¥8,000 first-visit spend looks worse than one delivering ¥12,000 — until you find the first cohort returns four times and the second never does. On lifetime value, the "worse" channel is worth three times more. Judging on first visit tells you to cut the good one.

The reinvestment reserve

The bracket at the end is the part people skip. If you spend the entire margin acquiring the guest, the acquisition is exactly break-even and the business has bought nothing. Hold back a share — a third is a common choice — so each acquired guest funds something beyond their own acquisition. The reserve is a decision, not a formula. Make it deliberately.

Why every venue overpays one channel

Because channel performance is judged on volume, and volume is the thing channels are best at manipulating.

A listing that sends a hundred people, ninety of whom order one drink and never return, will report a hundred. A referral source sending eight who become regulars reports eight. Both invoices are visible. Only one of the returns is. Without a guest record linking acquisition to lifetime revenue, the hundred wins every argument.

This is not a subtle effect. In most rooms it is the difference between the largest line of the marketing budget and the best one.

Attribution in a cash-heavy room

The obvious objection: most guests pay cash and sign up for nothing, so how is any of this attributed.

Honestly and partially. The practical anchors:

• The phone number taken at booking. Cheapest, highest coverage, and the single most valuable field in the business.
• The name on a reservation.
• The inbound call itself, if your phone system records where the number came from.
• A code, link or landing page per channel — the only fully clean signal, and worth insisting on with every paid source.

None is complete. Together they typically cover a meaningful share of guests, and — critically — they undercount every channel by roughly the same mechanism. That makes the ranking between channels trustworthy even where the absolute figures understate. Use it to decide where money moves, not as a P&L line.

Do not wait for perfect attribution to start. The ranking is available at fifty percent coverage and it is the ranking that changes the budget.

What the ceiling changes in practice

In a renewal conversation. A listing asks for a fifteen percent increase. Without the ceiling, you negotiate on feel. With it, you know whether you had room for thirty percent or should have cut them last quarter. The person on the other side of that call does not have your number.

In deciding what to test. A new channel is worth trying up to the ceiling of your current best channel — that is what a rational test budget is, rather than "let's put a bit into it and see."

In seasonality. The ceiling moves with LTV and margin, so quiet months genuinely support less spend, and the strong months support more. Most venues do this backwards, cutting ads when it is slow, which is often when the ceiling is highest because competition for attention is lowest.

In who you argue with. A promoter, an agency and a directory are all the same object in this math. One formula ranks them against each other.

Run it on your own numbers now

We put the calculation on our site, open, with nothing to sign up for and nothing transmitted anywhere:

→ Ad budget calculator

Put in what you spent per channel and what those guests actually spent. It returns the ceiling per channel as an amount. Use it and close the tab; nothing is stored.

You will need two things you may not have to hand: revenue attributed per channel, and some sense of repeat behaviour by channel. If you cannot produce those, that gap is the actual problem, and it is what the next section is about.

The system that produces this figure continuously

Running the calculation once, by hand, from a spreadsheet is worth doing this afternoon. Having it every month, from your own bookings, with nobody assembling anything, is a different thing — and it is what we built.

tasteck is a booking and analytics system for night venues. Its central feature is exactly this: guest records that persist across visits, first-touch channel attribution on those records, revenue by channel through to lifetime value, and the ceiling as an output — an amount, per channel, per month.

Why we are the ones who built it

Every other product in this category is a software company's model of the business. tasteck was built by operators who ran venues in this industry for sixteen years and grew from ¥200 million to ¥1.2 billion a year — six-fold, driven by attacking the problem with systems rather than by selling harder.

The arithmetic in this guide is not a content-marketing exercise. It is the arithmetic that produced that growth, run per channel, every month, cutting the channels that did not pay for themselves. The product is that process, made available to other rooms.

Ask it from ChatGPT

tasteck connects to ChatGPT over MCP. You ask your numbers as a question — "what's my ceiling on each channel this month," "which sources are producing repeat guests" — and the answer comes back in the chat. No dashboard to open, no report to find.

Measured against every vendor listed on Japan's principal nightlife-industry directory, this is the first implementation of it in the category. The same interface is callable from anywhere, so it is not tied to one assistant.

And the rest of the operation

Bookings, dispatch and driver status, staff scheduling and shifts, settlement and payouts, a reservation screen wired to inbound calls, and analytics across accounting, staff performance and guest segmentation. Multi-language is built in — the operating surface itself, with your language set put in place during onboarding.

From $34 a month for up to two venues. Thirty days free on every plan, cancel any time. → Pricing

Where to start this week

1. Capture a phone number at booking, every time. Nothing above works without a guest identifier, and this is the cheapest one in existence.
2. Give each paid channel its own link, code, or landing page. The only clean signal, and every channel worth paying will agree to it.
3. Compute LTV for one channel. Just one. The technique generalizes; the first is the hard one.
4. Run the ceiling on that channel and compare it to what you are paying. Expect a surprise in one direction or the other.
5. Take that number into your next renewal call. This is the step that pays for the other four.

---

On benchmarks. This guide quotes no industry averages for LTV, CAC, or margin. We do not have a dataset broad enough to stand behind, and these vary so widely by format, city and price point that publishing a range would be inventing precision. The point of the method is that it works on your figures rather than someone else's.