Case study: how a Founder Pricing partner scaled their phone coverage
A car rental operation in Florida. Missed calls before and after OLISE. Lessons from the first 90 days.
OLISE Team
Account
This case describes a Founder Pricing partner from OLISE's early program. The aggregate figures are real for the measured period; specific names are kept private at the customer's request. Individual metrics will be published with explicit permission once 12 months of operation are complete.
Starting point
Independent car rental operation in Florida, mid-sized fleet. Coverage structure: two human agents during business hours, voicemail off-hours. Result: a meaningful share of off-hours calls went unanswered, and peak-hour calls were lost to a busy line.
The customer arrived with three symptoms:
- 1Measurable lost reservations on weekends.
- 2Recurring complaints in reviews about "they don't pick up the phone."
- 3Rising cost of trying to cover 24/7 with humans.
What changed with OLISE
After a three-week onboarding (fleet config, integration with their reservation PMS, tuning the assistant on real scripts), OLISE began answering 100% of calls. This is not marketing, it is physics: the AI does not get busy.
The assistant:
- Checks live availability against the reservation system.
- Captures customer details, dates, vehicle category, payment method.
- Creates the draft reservation in the PMS.
- Transfers to a human when complex (prior damage, coverage policy, VIP cases).
Observed metrics (90-day window)
The figures below correspond to the 90-day window after go-live and are specific to this partner. They are not representative of every business.
- Calls answered: from 71% pre-OLISE to 99.4% post-OLISE.
- Call→reservation conversion: improved by a double-digit percentage relative to the human baseline (exact figure data-placeholder pending audit close).
- Cost per answered call: material reduction relative to 24/7 human shifts. Final figure at 12-month mark.
- Post-call CSAT (SMS survey): held above 90% month over month.
What we learned
Onboarding is the product. 80% of value comes from how well you train the assistant on your real rules —cancellation policy, surcharges, active offers, internal vocabulary. We went from underestimating this to treating it as priority one.
Human transfer is a feature, not a failure. Designing well when and how to escalate is what separates a useful AI from one that frustrates customers.
Off-hours calls are where ROI shows. 2 a.m. on a Saturday is gold: high-intent customers, no human competition, ready to book. If you don't answer, your competitor does.
What we don't tell here
We do not publish absolute revenue or margin figures without explicit written permission from the customer. When this partner crosses 12 months and signs off, we will publish a full case study with closed numbers, video testimonial, and audit-grade data.
In the meantime: if you want to see the playbook we apply, write to us.
blog.related
blog.category.industry
AI phone operator vs chatbot: why it matters for luxe hospitality
Not all channels are equal. Voice has a different cognitive cost, a different service expectation, and a different tolerance for error.
blog.category.engineering
Why we picked Anthropic Claude for our AI phone operator
We compared Claude, GPT-4o, and Gemini on natural Spanish, latency, and real tool use. When voice is the channel, the gap shows.
blog.category.engineering
How we built prompt-injection defense in production
Three layers of defense, an output filter with regex, and why we never trust a single line of protection. Real case with code.