5 min read
It is 9pm on a Tuesday and Maya is still at her kitchen table, seventeen browser tabs deep, trying to figure out whether a client’s honeymoon should route through Kyoto before or after the coast. She loves the actual work, matching people to places they will remember forever. She does not love that every custom itinerary eats an entire evening in research she cannot bill for. This is the version of Maya we are going to follow. She is an illustrative composite, built from the way many one-person travel advisors actually work, so the numbers are realistic rather than a single real person’s private books.
The problem hiding inside work she enjoyed
Maya runs a solo travel-advisory business. No assistant, no junior planner, just her and a laptop. Her bottleneck was never the booking or the client relationships. It was the front half of every trip: hours spent cross-checking neighborhoods, seasons, visa quirks, restaurant openings, and the dozen small facts that separate a good itinerary from an embarrassing one. A single complex trip could swallow six to eight hours of unpaid research. That ceiling meant she could only carry so many clients at once, and the overflow evenings were quietly burning her out.
What she tried first, and why it stalled
Her first instinct was the obvious one: work faster and later. That just meant more 10pm evenings. She considered hiring a part-time researcher, but the math never worked for a solo margin, and training someone on her taste would have eaten the very time she was trying to save. She had also heard the warnings about AI inventing facts, which made her nervous to trust it with something as detail-sensitive as international travel. That caution turned out to be the right instinct, and the key to using it well.
The setup that finally moved the needle
The unlock was not one magic app. It was using two kinds of AI for two different jobs, and keeping herself firmly in the loop.
For open-ended, factual digging, she leaned on an answer engine that cites its sources, the same approach we cover in our guide to using Perplexity as a one-person research department. Being able to click straight through to the source let her verify every claim instead of trusting a black box, which solved her accuracy fear. Tools like Perplexity are built around exactly that “show your work” habit.
For drafting and shaping, she used a general assistant such as ChatGPT to turn her verified notes into a clean first-draft itinerary in her own voice. She built one reusable prompt describing her tone, her typical client, and her non-negotiables, then fed it fresh trip details each time. The AI never talked to a client and never made a booking. It compressed the research and drafting, and she did the judgment.
What changed, in numbers she can feel
Within about two months, Maya’s research-and-draft time on a complex itinerary dropped from six or more hours to under two. She did not use the saved time to simply work less, though her evenings did come back. She used part of it to take on more clients, lifting her active load by roughly a third without hiring anyone, the same pattern we saw in a solo consultant’s playbook. More capacity, same forty-hour intent, no payroll.
A day in her new workflow
The change is easier to picture as a before and after. In the old version, a new trip request meant blocking a full evening: opening endless tabs, copying notes into a document, and slowly assembling an itinerary while second-guessing half the details. In the new version, she starts by handing the factual questions to her source-citing research tool, one focused question at a time, and pastes the answers she has verified into a running notes file. That first pass now takes forty minutes instead of three hours. Then she opens her drafting assistant, feeds it her saved voice-and-style prompt plus the verified notes, and asks for a structured first draft. She spends the remaining time doing the part only she can do: adding the personal touches, the “trust me on this restaurant” notes, the pacing that fits a particular couple. The AI handles the scaffolding. She handles the soul of the trip. Crucially, she never lets it contact a client or confirm a booking, so a stray AI error can never reach a customer unchecked.
Three lessons any solo owner can lift from Maya
Her craft is travel, but the pattern travels well.
- Split research from drafting. Use a source-citing tool for the “is this true” work and a general assistant for the “make this sound like me” work. One tool doing both jobs poorly is why people give up on AI.
- Verification is the job, not an afterthought. Maya’s edge was that she checked every fact against a cited source. AI made her faster, but her name still stood behind the itinerary. Keep the human on the parts that carry your reputation.
- Reinvest the time on purpose. Saved hours vanish if you do not decide where they go. She deliberately split hers between more clients and fewer late nights. Automating the back office, as a solo real estate agent did with her admin, frees capacity only if you claim it deliberately.
Where to start if you are the Maya in your field
Pick the one task that regularly steals an evening and does not require your signature skill. For Maya it was research. For you it might be first drafts, scheduling, or summarizing calls. Point a citing answer engine at the factual part, keep yourself on the judgment, and measure the hours before and after. If pairing AI with a smarter calendar is your bottleneck, our roundup of AI scheduling assistants is a fast next step. The goal is not to run your business on autopilot. It is to stop paying with your evenings for work a machine can rough out in minutes. Which task would you hand over first?
