6 min read
This is an illustrative story. Marisol is a composite based on common situations solo event planners describe, not a single real person. The tools and steps are real and you can copy them.
It is 11:40 on a Thursday night, nine days before a 120 guest vineyard wedding, and Marisol is sitting in her car outside a rental warehouse with her laptop balanced on the steering wheel. She has 63 unread emails. The florist needs a final stem count. The caterer wants the dietary list, which lives in three different text threads. The couple’s mother has sent a voice message asking, for the fourth time, what time the shuttle leaves the hotel. And Marisol, who runs her event planning business entirely on her own, still has two other events on the calendar that month.
Six months later, her weeks look different. Not magically empty, but calm enough that she took on a fourth event in the same busy month and still went to bed before midnight. Here is how she got there, one fix at a time.
The real problem was not the workload
When Marisol finally sat down to look at where her hours went, the surprise was that the planning itself, the part she loved, took the least time. The drain was everything around it:
- Answering the same questions from different people in different channels.
- Rewriting the same timeline into versions for the couple, the vendors and the day of coordinator.
- Chasing vendors for confirmations, then chasing clients for decisions.
- Typing up notes after every call, usually late at night.
She is far from alone. In a YouGov survey of nearly 4,000 small business owners commissioned by Meta, 76% reported that AI had a positive effect on their productivity, yet 86% said small businesses need more guidance to get the full benefit. Marisol was squarely in that second group. She had ChatGPT on her phone. She just used it like a fancy thesaurus.
Fix one: one project per event, one source of truth
The first change was structural, and it took an afternoon. Marisol created a separate ChatGPT project for each event. A project is simply a folder inside ChatGPT that keeps its own files, instructions and conversations together, so the assistant only “sees” what belongs to that wedding.
Into each project she uploaded the contract, the vendor list with contacts, the venue rules and the current timeline. Then she wrote short project instructions:
You are the assistant for the Ortega and Kim wedding on November 7. Answer only from the files in this project. If something is not in the files, say so and suggest who to ask. Keep messages warm, short and clear.
Now when a question came in, whether “what time is the shuttle” or “is the venue OK with sparklers,” she pasted it into the right project and had an accurate draft in seconds. The key phrase was “answer only from the files.” It stopped the assistant from inventing a plausible but wrong shuttle time, which in event work is worse than no answer at all.
Fix two: write the timeline once, publish it three ways
Marisol’s master timeline was a detailed spreadsheet: every vendor arrival, every cue, every contingency. Nobody except her needed all of it.
So she asked the assistant inside each event project to produce three versions from the same master:
- For the couple: a one page, friendly schedule with only the moments they need to be present for.
- For each vendor: just their load in, setup window, service times and the contact person on site.
- For the day of coordinator: the full cue sheet with phone numbers and the rain plan.
When the timeline changed (and it always changes), she updated the master and regenerated the three versions in a few minutes instead of an hour. She still read every version before sending it. Twice in the first month she caught a wrong time the assistant had copied from an outdated row, which taught her to delete old timeline files from the project instead of piling new ones on top.
Fix three: call notes that write the follow up for her
Client calls were Marisol’s favorite part of the job and the biggest source of late night typing. She started recording calls (always asking permission first) with an AI note taker, then pasting the transcript into the event project with one request: “List every decision, every open question and every task, with who owns it. Then draft a recap email to the couple.”
The recap went out within an hour of hanging up, while the conversation was still fresh for everyone. Clients loved it, and arguments about “but I thought we said” mostly disappeared. If you want to set this up yourself, our guide on turning every client call into a follow up email and task list walks through it.
Fix four: friendly, consistent vendor chasing
Every Monday, Marisol asked each event project a single question: “Which vendor confirmations are still missing, based on the files?” The answer became her chase list. She kept a short set of reminder templates that the assistant personalized: first nudge, second nudge with a deadline, and a final “I need this today” message that stayed polite.
The same approach worked on her own invoices. Instead of avoiding awkward payment reminders, she had calm, specific drafts ready. Our guide to chasing unpaid invoices with AI without sounding desperate covers the wording she adapted.
What did not work
Not everything stuck, and the misses are useful:
- Letting AI pick vendors. Asking the assistant to “suggest the best florist nearby” produced generic lists and one business that had closed. Vendor relationships stayed her job.
- Auto sending anything. She experimented with sending drafts straight out and stopped after one message went to the wrong mother of the bride. Every message now gets a human read.
- Mixing events in one chat. Early on she used a single long conversation for everything, and details from one wedding leaked into another. Separate projects fixed it.
The results, six months in
Marisol did not track her time perfectly, but her rough log told a clear story. Admin that used to take about fifteen hours a week settled closer to seven. Client recaps went from “sometime this week” to “same day.” Most importantly, she took on one more event in her peak month without hiring help, and she raised her planning fee for new clients, because her service now felt more organized and more responsive than it had before. If you are thinking about pricing in the same way, read how to move from hourly to project pricing when AI makes you faster.
Small businesses everywhere are reporting similar wins, which is why owners have even taken their AI stories to Capitol Hill. But the lesson from Marisol’s story is not about a specific tool. It is about structure: one organized place per client, clear instructions, and a human who reads everything before it goes out.
Try her first step this weekend
Pick your busiest client or project. Create one dedicated project or folder in your AI assistant, upload the three documents you check most often, and add the instruction “answer only from these files.” Then use it for one week. What is the question you answer most often that this could take off your plate?
