Guest CRM for Rooms Where Everyone Pays Cash

A coffee chain knows how often you come, what you order, and when you stopped. A night venue doing ten times the revenue per head usually knows nothing at all — not who was in the room last Saturday, not whether they had been before, not whether they came back.

This is not a technology gap. It is a habit gap. The technology to hold a guest record has existed for thirty years; what is missing in most rooms is the thirty seconds at booking where someone writes down who this is.

The cost of that gap is specific and large: without a guest record, you cannot tell which of the things you pay for actually brings people who come back. Every other consequence in this guide is downstream of that one.

What a guest record actually needs

Far less than people assume. Five fields carry almost all the value.

Field	Why it earns its place
Phone number	the identifier. Everything joins on this
First seen date	the clock for every retention number
First-touch source	how they found you — the field that decides the budget
Visit dates	the entire repeat picture is derived from this list
Spend per visit	turns visits into value

Names are nice. Emails are nice. The five above are the ones that pay. A room with only these, kept consistently, is ahead of nearly every competitor with a fancier system used inconsistently.

Why the phone number specifically

It is the only identifier a guest in this industry will reliably give you, it is the one they use to book, it survives them changing card, and it is how you will contact them. Email addresses in nightlife are given falsely at a high rate; phone numbers usually are not, because the guest wants the confirmation.

Capture it at booking, before the visit, when the guest has a reason to give it. Asking at the door or on the way out has a far lower success rate.

The four numbers a guest record makes possible

Nothing here is exotic. All four are impossible without the record and trivial with it.

Returning share

Returning share = guests tonight who have been before ÷ guests tonight

What it decides: how much of next month's budget goes to acquisition versus to reasons for coming back. A room running at a very low returning share is renting its traffic and will keep paying that rent forever.

Repeat interval

Repeat interval = median days between a returning guest's visits

What it decides: when to contact someone, and what "lapsed" means in your room. This second part matters more than people expect. If your guests naturally return every four months and your system calls anyone past ninety days inactive, you will spend the year chasing people who were never gone — and miss the ones who genuinely drifted.

Set the active window from your own repeat interval. It is a business decision, not a technical constant.

Channel lifetime value

Channel LTV = total revenue from guests first seen via a channel,
              across their whole relationship

What it decides: the entire marketing budget. This is the number that ends arguments about which listing to renew, and it does not exist without first-touch source on the guest record.

The reason it changes decisions so sharply is that channels differ far more in retention than in volume. A source delivering fewer, stickier guests routinely beats a high-volume one by a multiple — and looks worse on every report that stops at the first visit.

Cohort decay

Of guests first seen in month M, what share visited again in months M+1, M+2, M+3…

What it decides: whether anything you changed worked. Comparing this month to last month is noise. Comparing the March cohort's second-month return rate to the June cohort's is signal, because the two groups are matched on where they were in their own relationship with you.

Getting the data in a cash room

The practical answer is to capture at the moment the guest wants something from you.

At booking. By phone, by message, on a form — this is the highest-yield moment in the entire funnel, because the guest needs you to have their number. A booking flow that does not persist that number into a record is throwing away the only free data in the business.

On the inbound call. If the phone system passes the caller's number to whoever answers, the record can start before anyone speaks. This removes the transcription step entirely, and transcription is where capture rates go to die.

Through a per-channel link. Each paid source gets its own link or code. It is the only attribution signal that is unambiguous, and any channel worth paying will agree to it.

Not at the door. Door capture rates are poor and the queue punishes you for trying. Use the door for the count, not the record.

The consent and retention part, briefly

You are holding personal data on guests in a discreet industry, which means: collect what you use and nothing else, say what it is for, keep it for a defined period, and be able to delete on request. The five-field record above is small on purpose — a lean record is both more useful and less exposed than a large one. If your jurisdiction has specific rules for this sector, they govern; this paragraph is a design principle, not legal advice.

What changes once the record exists

Marketing stops being an argument. Channels are ranked by the lifetime value of the guests they delivered. The ranking settles it.

Contact gets timing. A guest at twice their normal interval is a message worth sending. One at half of it is not. The same list, sent by interval instead of by calendar, performs differently enough to be worth the effort of building it.

Staff performance gets a real denominator. A host who books returning guests is building something that belongs to the room. Without the record, they look identical to one running on cold volume.

You can tell what a guest is worth. Which is the precondition for knowing what you can pay to get another one — the subject of the ceiling calculation below.

The system built around this record

Everything above is achievable with discipline and a spreadsheet. It stops being achievable at around the point where you have several sources, several thousand guests, and staff who need to see the record while a guest is on the phone.

tasteck is a booking and analytics system for night venues, and the guest record is what it is built around — not a module bolted on.

What that means concretely: guest records that persist across visits, first-touch source attribution captured at booking, revenue by source through to lifetime value, cohort and repeat-interval analysis, and a reservation screen wired to inbound calls so the record starts from the phone call without anyone retyping a number.

The output nobody else produces

The lifetime value figures feed one thing that no other product in this category outputs:

The maximum you can spend on each channel next month, in your currency.

Not a performance chart — an amount, per source, that you take into the renewal negotiation. That is the whole reason the guest record is worth keeping, and it is the reason the product was built.

Built by people who ran the rooms

tasteck was built by operators who ran venues in this industry for sixteen years and grew the business from ¥200 million to ¥1.2 billion a year, six-fold, by attacking it with systems rather than by selling harder. Everything else in this category is a software company's model of the business. The reason the guest record sits at the centre of this one is that it was the thing that produced that growth.

Ask it from ChatGPT

tasteck connects to ChatGPT over MCP: ask your guest data as a question and the answer comes back in the chat — returning share this month, which sources produce repeat guests, who has gone past their usual interval.

Measured against every vendor listed on Japan's principal nightlife-industry directory, this is the first implementation of it in the category, and the same interface is callable from anywhere rather than being tied to one assistant.

Multi-language is built in, the operating surface rather than a translated brochure, with your language set put in place during onboarding. Alongside it: bookings, dispatch and driver status, staff scheduling, settlement and payouts, and analytics across accounting and staff performance.

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

Start here, this week

1. Capture a phone number at every booking. One field. Nothing else in this guide is possible without it, and no software is required to begin.
2. Add a first-touch source field with a short fixed list of options. Free text will defeat you within a month.
3. Give every paid channel its own link or code.
4. After sixty days, compute returning share and repeat interval. These two numbers will tell you more about the business than the revenue line did.
5. Then compute channel LTV and run it through the ceiling.

The ceiling calculation is open on our site with nothing to sign up for: Ad budget calculator. Nothing is transmitted anywhere — use it and close the tab.

---

On benchmarks. No target figures for returning share, repeat interval, or channel LTV appear in this guide. We do not have a dataset broad enough to publish them, and they vary enormously by format and price point. Your own first sixty days of records are a better benchmark than anyone else's average.