Running a taxi business without a dashboard is like driving without a speedometer. You know you're moving, but you don't know if you're accelerating or about to stall. Total rides and total revenue tell you the size of your business. They don't tell you if it's healthy.
Here are the 10 numbers that do — and what each one is warning you about when it moves in the wrong direction.
1. Completed Rides This Week (vs. Last Week)
What it tells you: Whether your business is growing, flat, or shrinking.
Why week-over-week matters: A single week's number is meaningless. The trend is everything. If rides are up 5% week-over-week consistently, you're growing. If they're flat or declining for 3+ weeks, something is wrong — fewer riders, fewer drivers, or both.
Action trigger: 3 consecutive weeks of decline → investigate. Is driver availability down? Are riders churning? Is a competitor running promotions?
2. Rides per Driver per Day
What it tells you: How productively your fleet is being used.
Healthy range: 8–15 rides per driver per day (varies by market and shift length).
Below 6: Either you have too many drivers for your rider volume, or your dispatch system isn't distributing rides efficiently. Drivers are sitting idle, getting frustrated, and thinking about leaving.
Above 18: Drivers are overloaded. Wait times are probably fine (drivers are always busy), but driver fatigue and burnout increase. Cancellation rates will creep up as tired drivers decline rides late in their shifts.
This is the number that connects rider growth to driver retention. If this number is healthy, both sides of your marketplace are balanced.
3. First-Attempt Acceptance Rate
What it tells you: How often the first driver offered a ride actually accepts it.
Healthy range: 80–90%.
Below 70%: Drivers are rejecting rides. The reasons: too far from the pickup, fare too low for the distance, or the ride is to an area drivers don't want to go. A low acceptance rate means every ride request bounces between 2–3 drivers before someone takes it — which adds 2–5 minutes to the rider's wait time.
Fix: Check which rides are being rejected. If it's long-distance pickups, your dispatch radius is too wide. If it's low-fare rides, your base fare may be too low. If it's rides to specific zones, those zones need better driver incentives.
4. Average Wait Time (Rider Perspective)
What it tells you: How long riders wait between requesting a ride and being picked up.
Healthy range: 3–7 minutes.
Above 10 minutes: Riders start cancelling. Above 12 minutes, they open a competitor's app. Wait time is the most visible metric to riders, and it directly correlates with repeat bookings. Riders will tolerate a slightly higher fare far more easily than a 15-minute wait.
Watch for peak vs. off-peak gaps. If your average wait time is 5 minutes overall but 14 minutes during Friday evening rush, you have a supply problem at specific times — not a general one. This is where surge pricing and scheduled driver incentives help.
5. Cancellation Rate (Rider + Driver Combined)
What it tells you: What percentage of confirmed bookings never become completed rides.
Healthy range: 5–10% combined.
Above 15%: You're losing rides you already won. Every cancelled ride wastes driver time (they drove toward the pickup for nothing) and frustrates the rider (they have to rebook). High cancellation rates are a symptom of: wait times too long (rider gives up), fare too low (driver regrets accepting), or driver too far away (distance recalculation).
Separate rider cancellations from driver cancellations. If riders are cancelling, the problem is wait time or pricing. If drivers are cancelling, the problem is dispatch quality or fare attractiveness.
6. Revenue per Ride
What it tells you: Your average fare, including base fare, distance charges, time charges, and surge.
Why it matters: If your total revenue is flat but rides are increasing, your revenue per ride is declining — which means your fares are effectively getting cheaper. This happens when rides get shorter (riders taking more short trips), surge pricing is underused, or you've lowered fares to compete on price.
Track this alongside rides per driver. A driver doing 12 rides/day at $6/ride earns $72. The same driver doing 15 rides/day at $4.50/ride earns $67.50 while working harder. Revenue per ride tells you whether more rides actually means more earnings for your drivers.
7. Monthly Active Riders
What it tells you: How many unique riders completed at least one ride in the last 30 days.
Why this is better than total downloads: Downloads mean nothing. Monthly active riders tell you how many people are actually using your platform. If you have 5,000 downloads and 400 monthly active riders, your activation rate is 8% — which means 92% of people who downloaded your app never came back (or tried once and didn't return).
Watch the trend: Growing monthly active riders means your retention is working. Flat or declining means you're acquiring new riders at the same rate you're losing existing ones — which is the most expensive way to stay the same size.
8. Driver Churn (Monthly)
What it tells you: What percentage of your active drivers stopped accepting rides this month.
Healthy range: Below 5% monthly.
Above 8%: You're losing drivers faster than you can replace them. At 10% monthly churn, you replace your entire driver fleet annually. This is expensive (recruitment costs), disruptive (service quality drops), and demoralising for the drivers who stay.
Pair this with rides per driver. If rides per driver is below 6 and driver churn is above 8%, the cause is clear: drivers aren't getting enough work. No retention programme fixes an empty shift.
9. Driver Online Hours vs. Revenue Hours
What it tells you: How much of a driver's online time is spent on revenue-generating rides versus waiting for the next trip.
Healthy range: 60–75% utilisation (revenue hours ÷ online hours).
Below 50%: Drivers spend more time waiting than driving. This is the leading cause of driver dissatisfaction. If you're at 40% utilisation, your drivers are earning for 3.5 hours out of every 8 they're online. That's below minimum wage in most markets — and they know it.
Above 85%: Sounds good, but it means drivers have no downtime between trips. Fatigue increases, cancellations rise, and service quality drops.
10. Rider Repeat Rate (30-Day)
What it tells you: What percentage of riders who completed a ride this month also rode last month.
Healthy range: 40–55%.
Below 30%: Your service isn't sticky. Riders try you once and don't come back. Before spending on acquisition, figure out why retention is low. Common causes: long wait times, poor driver quality, app crashes, or pricing that doesn't match the experience.
This is your most important growth metric. High repeat rates mean organic growth (repeats + referrals) exceeds churn. Low repeat rates mean you're on a treadmill — spending money to acquire riders who leave, requiring you to spend more to replace them.
The Monday Morning Dashboard
| # | Metric | Healthy | Warning | What It Diagnoses |
|---|---|---|---|---|
| 1 | Completed rides (WoW change) | +3–10% | Flat or declining 3+ weeks | Growth trajectory |
| 2 | Rides per driver per day | 8–15 | Below 6 or above 18 | Supply-demand balance |
| 3 | First-attempt acceptance rate | 80–90% | Below 70% | Dispatch quality |
| 4 | Average wait time | 3–7 min | Above 10 min | Rider experience |
| 5 | Cancellation rate | 5–10% | Above 15% | Booking friction |
| 6 | Revenue per ride | Market-dependent | Declining trend | Pricing health |
| 7 | Monthly active riders | Growing | Flat or declining | Retention effectiveness |
| 8 | Driver churn (monthly) | Below 5% | Above 8% | Driver satisfaction |
| 9 | Driver utilisation rate | 60–75% | Below 50% | Idle time / earnings quality |
| 10 | Rider repeat rate (30-day) | 40–55% | Below 30% | Service stickiness |
Print this table. Check these 10 numbers every Monday morning. If all 10 are in the healthy range, your business is running well. If any 3 drop into warning territory simultaneously, you have a systemic problem — not a one-off fluctuation.
The operators who succeed long-term aren't the ones with the flashiest apps or the biggest marketing budgets. They're the ones who know their numbers, spot problems early, and fix them before riders and drivers leave.
KPI benchmarks informed by Hyvikk, Zoom.taxi, and Onde fleet analytics research (2025–2026). Utilisation rate targets referenced from fleet management industry standards (Visual Planning, SimplyFleet). Wait time and cancellation rate correlations with rider retention from UXCam and ride-hailing platform analytics. Benchmarks are generalised — adjust targets to your market, fleet size, and operational maturity.
