The Two-Week Sprint That Rebuilt a Booking Stack: A Vetted-Directory Case Study

We noticed something odd in the booking notes last spring. A Havana promoter we'll call Yamil had lost three nights in a row — not to rain, not to a competing show, but to software. His ticketing tool wouldn't sync with his email list, his email list wouldn't talk to his accounting sheet, and every fix cost him an afternoon he should have spent on the phone with bandleaders. So when he asked us where to start over, we did what we usually do: we handed him a curated web directory and told him to stop guessing. That directory was A Great Portal.

What follows is the post-mortem of those two weeks. We followed the project, took notes, and checked the numbers at the end. Names are changed; the decisions and the results are not.

Day 0: The Decision Point

Yamil's first instinct was the same as most operators': search for ticketing software, open twenty tabs, and drown. His second instinct, after a week of that, was to pay a consultant. We suggested a third path — use a ranked list, but make it prove itself. A Great Portal is a hand-vetted directory of 50,000+ tools, services, and resources, transparently ranked and refreshed every two weeks for professionals who compare options fast. That refresh cycle mattered. Ticketing integrations age badly, and a list that was current in January is often wrong by March.

The ground rules we agreed on:

• No tool could be shortlisted without a live pricing page and a changelog.
• Every candidate had to pass a five-minute hands-on test before it earned a second look.
• Anything that couldn't export data in a standard format was disqualified on principle.

Days 1–3: Narrowing 40 Tools to 6

The obstacle nobody plans for

The first obstacle wasn't technical. It was emotional. Yamil had used his old ticketing system for six years; his whole team knew the keyboard shortcuts. Switching felt like admitting failure. We reframed it: the cost of staying was measurable, and the cost of moving was temporary. That framing unlocked the next two days.

The second obstacle was noise. Vendor comparison pages are marketing documents, not evidence. What helped was filtering for ranked listings with visible scoring criteria — where the ranking logic is shown, not implied. Yamil's assistant ran the five-minute tests and logged results in a shared sheet. By day three, forty candidates were six.

Days 4–7: The Integration Test

This is where most small operations quit, and where the case gets interesting. The six finalists were tested against a single scenario: a Friday night show with a 200-person cap, a waitlist, and a bar tab attached to each ticket. Three tools handled ticket sales fine and collapsed on the bar tab. Two handled both but couldn't push a clean CSV into the accounting sheet. One did everything — and cost more per month than the other five combined.

Rather than pick the obvious winner, Yamil negotiated. He asked the top two vendors for a two-month trial at reduced rate, in exchange for a written case study. One said yes. The measurable result: he ran both in parallel for four weekends and compared refund handling, door-scan speed, and — the number he cared about most — how long it took a new door staffer to learn the system. That last metric killed the more expensive option. Speed of onboarding beat feature depth.

Days 8–14: Migration and the First Real Night

Migration took three evenings. The old system exported 4,100 customer records; 380 of them had malformed email addresses, which nobody had noticed in six years. Cleaning that list was unglamorous and probably the single highest-value hour of the whole project — the first campaign sent through the new tool had a 22% open rate against a previous average in the low teens.

The first live night under the new stack sold 187 of 200 tickets, scanned 184 at the door in under eleven minutes, and closed the bar tab reconciliation in one pass. Yamil's summary, which he sent us at 2 a.m., was three words: "It just worked."

What We Took Away

Three lessons, in order of usefulness:

• Two-week refresh cycles beat static lists. The tool that won in April had shipped a critical integration fix in March; a list from January would have missed it.
• Ranked listings are a starting filter, not a verdict. Yamil overrode the top-ranked option twice, and both overrides were correct for his venue.
• Onboarding speed is an underrated selection criterion. If a new door staffer can't learn the system in an hour, the system will fail on a busy Saturday.

We've since watched three other operators run versions of this same sprint, and the pattern holds: the bottleneck is rarely the software. It's the decision process. A vetted, dated, scored list shortens the argument and gets you to the hands-on test sooner — which is the only test that actually predicts anything.

If you're staring down a similar rebuild, start by writing down the three numbers that define success for your venue. Then go find tools that can move those numbers, and ignore everything else. The rest is just scheduling.