STR operations

Ask Claude, connected to Hospitable, where your vacation rental guests have come from since each listing started, and clean up the messy location data first

Andrew A., a host operating vacation rentals in Tbilisi, Georgia, posted in the Hospitable Community's MCP Builders forum a test of Claude Opus 4.7 against ChatGPT, both connected to Hospitable's MCP server, on one question: where have guests come from since each listing began, and how has guest origin changed over about five years, since regional shifts matter for a Tbilisi host. The prompt warns the model that guest location data from Hospitable and Airbnb is not normalized and asks the model to look at the data and think before charting. Andrew says Claude returned a chart, a breakdown and insights on the first run, flagged gaps in the dataset, and improved the dataset coverage on a second run, but used up his Claude usage allowance for the next three hours; ChatGPT needed step-by-step prompting and, on GPT-5.5, first worked with only about one-third of the data until the model paged through the full reservation set. A second host, Tom Beerley, pasted the same prompt into Claude for an Ocean City beach rental and learned that 117 stays with a known origin came mostly from Pennsylvania and Maryland, with the drive market of Pennsylvania, New Jersey, Maryland, Delaware and Virginia making up about 76% of each month's known stays.

At a glance

Tools used
ClaudeHospitableMCP
Category
STR operations
Audience
STR host
Difficulty
No-code setup
Source
Claude blows ChatGPT out of the water with the Hospitable MCP Server
Andrew A. (Hospitable Community) · Forum post · Apr 29, 2026
First step
Connect Claude to your Hospitable account through Hospitable's MCP server, then paste the guest-origin prompt below into a new chat with nothing else, and expect the first run to use much of your usage allowance.

Compliance note: Use guest-origin numbers for marketing and pricing decisions only, never to accept or reject a guest, because screening guests by national origin breaks Airbnb's nondiscrimination policy and fair housing law; keep guest names out of any chart you share.

Suggested prompt

Written by this directory, inspired by the source. Replace the [bracketed] parts.

Can you visualize over time since the very beginning of each listing (grouped is fine) where our guests have come from? I want to see a changing or constant 'demographic'. Watch out though, the location is not normalized that comes back from Hospitable / Airbnb. So look at the data then think it through first!
Then: Tell me how many stays had no usable guest location, and list the three origin regions whose share changed most from the first year to the last year.

Share on X Read the source

← All use cases