6 min read
This is an illustrative story. Rosa is a composite based on common situations for solo florists, not a real person, and the numbers are realistic examples rather than reported results.
It is the Monday before Thanksgiving last year, and Rosa is standing in her tiny flower studio with a phone in one hand and a bucket of eucalyptus in the other. Her phone has 31 unread messages. Seventeen are some version of “Can I still get a centerpiece for Thursday?” Four are from people who asked last week, got a quote, and never confirmed. Two are angry, because she ran out of the burgundy dahlias they assumed she had saved for them.
Rosa runs a one-person floral business. She does weddings in summer, weekly office arrangements, and a holiday rush that is the most profitable and most chaotic stretch of her year. That Thanksgiving she made good money, but she also wasted close to two days on messages, ordered flowers on guesses, and threw away about 15 percent of her stems.
This year, she tried something different. Here is what she changed, and how AI did most of the heavy lifting.
The real problem was not the flowers
When Rosa looked back at the mess, she realized her design work was never the bottleneck. The bottleneck was taking orders by chat. Every order was a mini negotiation: what colors, what size, pickup or delivery, what budget, and did they actually want it? She could not order wholesale flowers until she knew her numbers, and she did not know her numbers until the last minute.
Her goal for this season was simple: have every holiday order confirmed and paid by a set date, so she could buy flowers once, with confidence.
Step one: AI helped her build a holiday menu
Instead of custom quotes, Rosa decided to offer a small menu. She opened ChatGPT and described her business, her price range and her style. Then she asked it to propose a holiday collection with three sizes of centerpiece and two add-ons.
The first suggestions were generic, so she pushed back with specifics: her signature mix of seasonal greens, the fact that she can only do 40 deliveries in one day, and that her customers love candles in arrangements but she worries about fire safety. After a few rounds she had a clean menu:
- The Small Table, a low centerpiece for up to six guests.
- The Long Table, a runner-style arrangement for larger gatherings.
- The Statement, a tall entry or sideboard piece.
- Add-ons: battery candles and a matching small hostess bouquet.
AI also helped her write short, honest descriptions for each one, including what flowers might be substituted if supply changed. That single sentence about substitutions headed off most of last year’s complaints.
Step two: one form replaced the message chaos
Rosa built a simple order form using Google Forms (Google’s Forms help pages cover the basics). She asked Claude to suggest the questions in the right order, and it recommended keeping it to eight fields: name, phone, menu choice, color palette from three options, pickup or delivery, delivery address, date, and any notes.
The key decision was the deadline. The form clearly stated: holiday orders close the Saturday before Thanksgiving. After that date, she would sell whatever was left as walk-in bouquets.
When people messaged her on Instagram or text, she replied with one saved message and the form link. No more back and forth about colors in a DM thread.
Step three: a deposit made orders real
Last year’s no-shows hurt the most, because unclaimed centerpieces are pure waste. This year Rosa created a payment link for each menu item using Stripe Payment Links, which let her accept payment without building a full online shop. The form’s confirmation page sent people straight to the right link.
She asked AI to help write the confirmation email that followed payment. It covered pickup times, care tips, and a polite line about the substitution policy. She edited it so it sounded like her, then saved it as a template.
Step four: AI turned orders into a buying list
This is where the system paid off. On the Sunday after orders closed, Rosa exported her form responses to a spreadsheet and pasted them into ChatGPT along with her own recipe card for each menu item: how many stems of each flower and greenery go into each size.
She asked it to total the stems needed by color palette, add a 10 percent buffer, and group the list the way her wholesaler sells: by bunch. In about fifteen minutes she had a buying list she trusted. She double-checked a few totals by hand, because AI can make arithmetic slips with long lists, and found one small error in the greenery count. The rest was right.
She also asked it to sort orders into a delivery route by neighborhood, which cut her delivery day down by over an hour.
What changed this season
Here is how Rosa’s holiday week compared with the year before:
- Message handling: from close to two full days to about three hours across the whole season.
- No-shows: from a handful of unclaimed pieces to zero, because every order was paid upfront.
- Flower waste: from roughly 15 percent of stems to under 5 percent, because she bought from real orders instead of guesses.
- Stress: she slept the night before Thanksgiving, which she says was the real win.
What she would do differently
Rosa admits a few things did not go perfectly. She opened orders too late, only ten days before the deadline, and some regular customers missed out. Next year she will announce the menu three weeks early with a short email to past customers. Before she does that, she plans to tidy her contact list, using the steps in our guide to cleaning a messy customer list with AI.
She also wishes she had added an FAQ section to the form. Several people still messaged her to ask about vase returns and whether arrangements were pet-safe. A handful of written answers would have saved more time, much like the approach in our candle maker story.
Could this work for your business?
You do not need to be a florist. The same pattern works for anyone who sells a seasonal product made to order: bakers, wreath makers, gift-box sellers, caterers. The recipe is the same: a short menu instead of custom quotes, one form, a deposit, and a firm deadline, with AI helping you write, total and organize. Our home baker story shows the same idea applied to custom cakes.
If you sell something seasonal, what is the one part of your holiday rush you would most like to fix this year? Tell us in the comments.
