← case studies

How Axia Partners Cut Park Comps From 80 Hours to Same Day

Jake Jones had just finished an 80-hour comp by hand when he ran the agent as a check. It found every comp he had, plus about twice as many he hadn't, in a day.

80 hrs → same day

Cut from each park comp analysis

2–4 months

Implementation Time

Not disclosed

Project Cost
the challenge
Axia Partners underwrites RV park and outdoor hospitality acquisitions, and a full comparable analysis took 80 analyst hours per park, roughly three weeks of elapsed time. At the firm's acquisition pace that math stopped working, and Axia had planned to hire two analysts purely to keep comps ahead of the pipeline. Separately, a portfolio-wide comp refresh across every park took 21 days, so in practice it rarely got run.
what they built
SageCreek built Axia a comp analysis agent running on Claude Opus 4.6. Purpose-built browser automation collects booking and occupancy data for the target market, the agent runs the same analytical steps Axia's analysts ran by hand, and it writes up the study. It was built side by side with acquisitions lead Jake Jones so the output matches how the team underwrites, and it was checked against a completed 80-hour manual comp before going into use.
SageCreek built the agent alongside Jake Jones, Axia's acquisitions lead, rather than from a written spec. The starting point was the workflow the analysts already ran: which comparables get pulled, what booking and occupancy data matters, how the numbers get normalized, and what the finished study has to say for the deal team to act on it. Purpose-built browser automation gathers the data from the sources the analysts had been visiting by hand, and the agent, running on Claude Opus 4.6, executes the same analytical sequence and drafts the written study. The proof came from Jones himself. He had just finished an 80-hour comp analysis by hand and ran the agent against it as a check. It returned every comparable he had found plus about twice as many he had not, with more information on each. His reaction: "I don't want to be corny, but this is life changing." Mapping the workflow took about a week and the build took roughly two months. Jones now runs comps himself instead of queueing analyst time, the two planned analyst hires came off the plan, and the 21-day portfolio-wide refresh runs overnight for about $200 a month. Axia is extending the same agent into underwriting. Disclosure: Greg Butterfield is a partner at SageCreek and a general partner at Axia Partners.
best fit for
Best for real estate and outdoor hospitality investors running an active acquisition pipeline, where comparable analysis is the bottleneck and the underlying data sits in booking and occupancy platforms rather than a licensed comps database. Strongest where an experienced analyst can sit with the build, and where a completed manual comp exists to validate against.
Ai ROLE
The agent performs the comparable analysis itself. It collects booking and occupancy data for the target market, runs the analytical steps Axia's analysts previously ran by hand, and drafts the written comp study. Checked against a completed manual comp, it found every comparable the analyst had found and about twice as many he had not. The acquisitions lead now runs comps directly instead of queueing analyst time, and the same agent is being extended into underwriting.
impact

80 hrs → same day

of analyst work per park comp, now returned the same day

~2x the comps

found versus the same market worked by hand, with none missed

2 analyst hires

called off; the 21-day portfolio refresh now runs overnight
implementation complexity
Medium. A browser-automation data layer plus an agent that reproduces an existing analyst workflow, built collaboratively with the acquisitions lead and validated against a completed manual comp. No legacy system integration or data infrastructure program, and running cost is roughly $200 a month. Inferred editorially from the Solution Description and Approach, not stated by SageCreek.

Connor McLeod

Partner, SageCreek
SageCreek
Partner at SageCreek, a Utah firm building AI agents and decision tools for private equity firms and their portfolio companies.
Get an intro
industry
Real Estate
Hospitality & Travel
business organization
Executive & Strategy
Operations
AI TYpe
Data Synthesis & Reporting
Process Automation (RPA + AI)
Decision Support & Scoring
value type
Time Savings
Headcount Avoidance
frequently asked questions
How did a real estate PE firm cut RV park comp analysis from 80 hours to the same day?
Axia Partners worked with outside experts to build a comp analysis agent running on Claude Opus 4.6. Purpose-built browser automation collects booking and occupancy data for the target market, and the agent then runs the same analytical steps the firm's analysts ran by hand before writing up the study. A park comp that took 80 analyst hours now comes back the same day.
What AI models and tools were used to build the comp analysis agent?
The agent runs on Claude, and the results published here were produced on Opus 4.6. The reasoning layer is model-agnostic and gets moved to whichever model performs best at the time. The browser automation that feeds it is purpose-built for the client's data sources and does not swap out.
What results did Axia Partners achieve?
A park comp that took 80 analyst hours now returns the same day. Checked against a completed manual comp, the agent found every comparable the analyst had found by hand and about twice as many he had not. The 21-day portfolio-wide refresh now runs overnight for about $200 a month, two planned analyst hires were called off, and the agent is being extended into underwriting.
How long did the comp analysis agent take to build?
About one week to map the existing workflow and roughly two months to build, which puts it in the two to four month range. It is in production rather than a pilot.
Who is this AI comp analysis approach best for?
Real estate and outdoor hospitality investors with an active acquisition pipeline, where comparable analysis is the bottleneck and the underlying data sits in booking and occupancy platforms rather than a licensed comps database.

Have a similar challenge?

Tell us what you're working on and we'll match you with the right expert.
TELL US WHAT YOU'RE EXPLORING