How it started
We run Double Rainbow Suites: a 2BR pool house in Fort Lauderdale and two efficiency suites in Wilton Manors, which can also be booked together as a 2BR. Our Airbnb year-to-date export looked bad, down about 16% on revenue and 22% on nights. I asked Claude to tell me how we did year over year, how we compare to similar properties, and what to expect next year.
What the research found
- Airbnb alone told the wrong story. Claude pulled every reservation for both years from Hospitable across all channels. Total revenue was actually flat, and nights were up 8%. Booking.com more than tripled, and Vrbo was new this year. Bookings had shifted channels; we hadn’t lost demand.
- The real problem was summer. January–April was up about 20%, but May–October was down about 16%. The house took the biggest hit, with June and September nights down 70–80%.
- Claude read our Airbnb Performance dashboard in the Claude desktop app’s built-in browser. We were priced 20–45% above similar listings and still beat them on occupancy from January to April. From August on, at 66–92% above similar listings, views dropped about 76% and occupancy fell to half of theirs.
- It wasn’t Airbnb’s fee change. Claude found the exact date we switched to host-only fees from the service fee on each reservation. Our 18.5% markup roughly offsets that fee for guests, and our best months came right after the switch.
- The cause was our dynamic pricing minimums. The custom seasonality was set to cut summer prices 20–39%, but the minimum prices blocked it. The house’s minimum was above what guests had actually paid the summer before. The combined 2BR could never be cheaper than booking the two suites separately, so it got zero bookings.
- Settings were inconsistent between near-identical units: different seasonality, demand sensitivity, minimum-stay rules and Booking.com instant-book settings.
What we changed
- Lowered the minimum prices on the house and the combined 2BR, and moved all units to Hospitable’s recommended seasonality and demand sensitivity.
- Set dynamic minimum stays to 3–4 nights far out, dropping to 1 night close in (the “Decrease sooner” setting), so gap nights fill.
- Created reusable event rule-sets (+10% to +40%, some with a 2-night minimum) and applied them to local events Hospitable’s pricing wasn’t picking up, mostly Pride, White Party and Pig Week. Every event date also got a calendar note with the event name.
- Because a rule-set’s minimum stay is fixed, scheduled tasks swap each 2-night-minimum rule-set for a price-only version 9 days before the event. The price increase stays, and the minimum stay drops to 1 night.
- A monthly scheduled task reads our website’s events page, adds notes and rule-sets for new or changed events, and sends me a summary.
- Scheduled 30- and 60-day reviews to check the results and recommend the next adjustments.
What made it work
The Hospitable MCP gave Claude every reservation, payout breakdown, fee, calendar day, pricing setting and minimum-stay setting for every channel, all in one place. The built-in browser covered what the MCP doesn’t, like Airbnb’s comparison with similar listings and Hospitable’s dynamic pricing and rule-set screens. A data review that started as one Airbnb export turned into a full diagnosis and working fixes in one afternoon.
Wish list for the Hospitable MCP: changing rule-sets and dynamic pricing settings, plus a history of markup and pricing changes. Most of the clicking Claude did in the browser would then be a single tool call.
Want a shorter version, or one with your revenue numbers in?
when speaking about dynamic pricing, list it as Hospitables Dynamic pricing so everyone including the hospitable team can learn from it
Here’s the revised post. Every pricing mention now names Hospitable’s Dynamic Pricing.
How we used Claude + the Hospitable MCP to diagnose a summer slump and rebuild our pricing
How it started
We run Double Rainbow Suites: a 2BR pool house in Fort Lauderdale and two efficiency suites in Wilton Manors, which can also be booked together as a 2BR. Our Airbnb year-to-date export looked bad, down about 16% on revenue and 22% on nights. I asked Claude to tell me how we did year over year, how we compare to similar properties, and what to expect next year.
What the research found
- Airbnb alone told the wrong story. Claude pulled every reservation for both years from Hospitable across all channels. Total revenue was actually flat, and nights were up 8%. Booking.com more than tripled, and Vrbo was new this year. Bookings had shifted channels; we hadn’t lost demand.
- The real problem was summer. January–April was up about 20%, but May–October was down about 16%. The house took the biggest hit, with June and September nights down 70–80%.
- Claude read our Airbnb Performance dashboard in the Claude desktop app’s built-in browser. We were priced 20–45% above similar listings and still beat them on occupancy from January to April. From August on, at 66–92% above similar listings, views dropped about 76% and occupancy fell to half of theirs.
- It wasn’t Airbnb’s fee change. Claude found the exact date we switched to host-only fees from the service fee on each reservation. Our 18.5% markup roughly offsets that fee for guests, and our best months came right after the switch.
- The cause was our minimum prices in Hospitable’s Dynamic Pricing. Our custom seasonality in Hospitable’s Dynamic Pricing was set to cut summer prices 20–39%, but the minimum prices blocked it. The house’s minimum was above what guests had actually paid the summer before. The combined 2BR could never be cheaper than booking the two suites separately, so it got zero bookings.
- Hospitable’s Dynamic Pricing settings were inconsistent between near-identical units: different seasonality, demand sensitivity and minimum-stay rules, plus different Booking.com instant-book settings.
What we changed
- Lowered the minimum prices in Hospitable’s Dynamic Pricing on the house and the combined 2BR, and moved all units to Hospitable’s Dynamic Pricing recommended seasonality and demand sensitivity.
- Set the dynamic minimum stays in Hospitable’s Dynamic Pricing to 3–4 nights far out, dropping to 1 night close in (the “Decrease sooner” setting), so gap nights fill.
- Created reusable event rule-sets (+10% to +40%, some with a 2-night minimum) on top of Hospitable’s Dynamic Pricing. We applied them to local events it wasn’t picking up on its own, mostly Pride, White Party and Pig Week. Every event date also got a calendar note with the event name.
- Because a rule-set’s minimum stay is fixed, scheduled tasks swap each 2-night-minimum rule-set for a price-only version 9 days before the event. The price increase stays, and Hospitable’s Dynamic Pricing takes the minimum stay down to 1 night.
- A monthly scheduled task reads our website’s events page, adds notes and rule-sets for new or changed events, and sends me a summary.
- Scheduled 30- and 60-day reviews to check the results and recommend the next adjustments.
What made it work
The Hospitable MCP gave Claude every reservation, payout breakdown, fee, calendar day, pricing setting and minimum-stay setting for every channel, all in one place. The built-in browser covered what the MCP doesn’t, like Airbnb’s comparison with similar listings and the Hospitable’s Dynamic Pricing and rule-set screens. A data review that started as one Airbnb export turned into a full diagnosis and working fixes in one afternoon.
Lessons for Hospitable’s Dynamic Pricing users
- Check your minimum prices every season. If they’re set too high, they quietly cancel the seasonal and last-minute discounts.
- Hospitable’s Dynamic Pricing picked up some events (the Boat Show, July 4) but not local LGBTQ events like Pride or White Party. Event rule-sets filled that gap.
Wish list for the Hospitable team:
- Let the MCP change rule-sets and Hospitable’s Dynamic Pricing settings.
- Keep a history of markup and Hospitable’s Dynamic Pricing changes.
- Make rule-set minimum stays expire automatically a set number of days before the dates.
Most of the clicking Claude did in the browser would then be a single tool call.
