About RideCompare
Ride-hailing prices are opaque, variable, and almost impossible to compare fairly. RideCompare is a community effort to change that — one receipt at a time.
What we do
Collect
Riders submit fare details — price, distance, ETA, city — after real trips on Uber or Lyft.
Aggregate
We pool submissions by city and time window, compute rolling averages, and derive surge proxies from price-per-mile deviation.
Summarise
An AI generates a plain-English weekly digest per city so you don't have to read charts.
Methodology & caveats
All data is self-reported. We don't integrate with Uber or Lyft APIs. Users submit fares manually or via screenshot scan. Data quality depends on honest, accurate submissions.
We use safe proxy metrics. "Surge frequency" is estimated from price-per-mile deviation (>25% above city median), not a direct surge flag. "ETA accuracy" is self-reported deviation, not GPS-verified.
Small samples = wide error bars. Cities with fewer than 50 submissions are labelled Preview and should be treated as directional, not authoritative.
AI summaries may be imperfect. Weekly digests are generated by an LLM (GPT-4o) from aggregated stats. They are informational and may contain errors. We never claim causality — language like "may suggest" or "trend shows" is intentional.
Frequently asked questions
Is this affiliated with Uber or Lyft?
No. RideCompare is fully independent. We have no relationship with either company, and the data is entirely user-submitted.
How accurate is the data?
Accuracy improves with sample size. Cities marked 'Preview' have fewer than 50 submissions and should be treated as directional, not definitive. We show confidence caveats on all stats.
What does 'surge proxy' mean?
We don't receive surge flag data directly. Instead, we flag a ride as potentially surged when its price-per-mile is more than 25% above that city's rolling median — a common proxy used by researchers.
Why do my submitted numbers look slightly different?
Stats are shown as rolling 7-day averages. A single submission moves the needle slightly but won't dramatically change a city's average on its own.
Can I delete my submission?
Yes — email us at ridecompareapp@gmail.com with the city, date, and approximate fare and we'll remove it within 48 hours.
How is ETA accuracy calculated?
We compare the estimated pickup time shown at booking against the actual wait time the user reports. Positive values mean Uber/Lyft underestimated the wait.
Get in touch
Questions, data corrections, or press enquiries — ridecompareapp@gmail.com
Looking for collaborators
RideCompare is a one-person side project right now. If any of this resonates with you, I'd love to work together — part-time, async, no pressure.
Full-stack / data engineer
Help build real API ingestion, scale the backend, or improve the receipt parser. Python / TypeScript background ideal.
Data analyst
Design better aggregation models, improve surge detection, and find patterns in the submission data as it grows.
Growth / community
Help grow the rider community, run city-specific campaigns, or build partnerships with transit and commuter groups.
Interested? Reach out at ridecompareapp@gmail.com with a quick note about yourself.
Help improve the data
Every submission makes city averages more accurate. It takes 30 seconds — or just scan your receipt screenshot.
Submit a Ride