HOW THE PAY WORKS
Three numbers make a delivery fee
Bolt's own published structure for a courier's earnings has exactly three parts:
- a pickup fee
- the distance from the restaurant to the customer
- a drop-off fee
Nothing else is in the base.
On top of that: bonuses, a surge multiplier shown live on the map, occasional compensation, and customer tips where the market supports them. Off the top: deductions and VAT if registered.
There is no time in the formula. Waiting at a counter for an order that isn't ready earns nothing — not less, nothing. The only way to be paid for time lost is a discretionary "compensation," and it has to be asked for by name.
Distance is paid restaurant-to-customer only. The ride to the pickup, and the ride back to wherever the next order comes from, are unpaid by design. Over the first week, the app recorded roughly 70% of the distance the odometer actually measured — that gap is the business model, not a bug.
The weekly fee band
A flat fee is taken off the week's total earnings once, based on which band the week's gross fell into. It only goes up as the week's total rises — never down.
The trap: crossing a band edge costs the full new fee, not the difference. One week this rider sat 18.85 RON below an edge — crossing it by even 1 RON would have made the next ~11 RON of driving earn nothing. Stopping just under an edge, or riding well past it, are the only two good options. Stopping just inside the gap is the one genuinely bad outcome.
DAY BY DAY
Four weeks, day by day
Delivery fees by day, 07 September to 04 October. Label = deliveries / RON. A rented moped throughout — one bike for 07–09.09, another from 10.09. The app’s vehicle setting read “Bicycle” on the first days only.
| Date | N | Fees | Tips | Gross | Km | Acc. |
|---|
- Date — the calendar day
- N — deliveries completed that day
- Fees — Bolt's own delivery fees earned, before tips (RON)
- Tips — app tips and cash tips combined (RON)
- Gross — fees + tips added together, before any costs come out (RON)
- Km — distance Bolt counted for that day’s orders. Bolt counts only the order legs; the distance actually ridden runs about 1.4–1.9× higher
- Acc. — acceptance rate: the share of offered orders accepted. Colored by rate: 95%+, 85–94%, under 85%
Tips split two ways that matter: in-app tips are inside the weekly fee band and lose 10% commission; cash tips never touch the app, never touch the band, and keep every leu. Across the four weeks cash tips came to 607.41 RON against 232.74 RON in the app — more in hand than on screen, and worth tracking separately, because they behave completely differently.
RON PER HOUR
The number that actually matters
Fees per delivery are easy to publish and easy to misread — the real question is what an hour of being online is worth, waiting included. The online clock comes from captured app blocks where they exist; otherwise it is the first-to-last span of the day minus the offline time reported.
Over all 28 days: 26.05 RON/h — 5'418.73 RON take-home over 208 hours online.
Same rider, against Romania's own statutory minimum wage (HG 146/2026: 4'325 RON gross / 2'699 RON net per month, 168-hour month). On the same basis a 26-day month comes to about 3'321.46 RON net, after a month of rent and fuel.
Neither number tells the whole story. None of this sits inside either figure — the courier carries all of that risk personally, on both sides of the comparison:
- no paid holiday
- no sick pay
- no employer pension contribution
- no accident cover
Pay per kilometre, week by week
Delivery fees divided by the kilometres Bolt counted fell every week of the four:
- Week 1, 09.07–09.13: 4.23 RON per Bolt-km
- Week 2, 09.14–09.20: 3.90 RON per Bolt-km
- Week 3, 09.21–09.27: 3.72 RON per Bolt-km
- Week 4, 09.28–10.04: 3.62 RON per Bolt-km
On 04.10 a stacked second order — a second pickup added to a delivery already under way — paid about 6.50 RON. These are measurements from one account. The cause sits in Bolt’s pricing and is not confirmed by Bolt.
WHAT ACTUALLY ARRIVED
Every number above is money earned, not money received
This is the part most courier write-ups leave out, and it is the part that decides whether the work was worth doing. Everything on this page up to here is what the app recorded and what the fleet owed. It is not what reached a bank account.
Twenty-eight days were worked out of twenty-eight, 343 deliveries, every shift completed and not one day off. For the three weeks settled so far, the fleet's own figures and the rider's day-by-day record agree to within one bani.
The costs ran on time; the income did not. Rent was paid every week in advance, in cash, on time. Fuel was paid at the pump. The thermal bag and powerbank were paid on day one. None of those suppliers waited to see whether the fleet paid.
- the payment schedule that was missed was never in the signed contract — it existed only in messages
- the contract itself says payment is monthly, which is the gap the fleet can argue from
- three different people promised payment on the same day, in writing; nothing arrived that day
- the platform declines to help: fleet-managed payouts are not its concern
- until the money came, the rider carried the gap alone and kept riding to avoid losing the account
So read the earnings figures on this page for what they are — an accurate measure of what the work produces per hour, per kilometre and per delivery. When that money arrives is a separate question, and it depends entirely on which fleet stands between the platform and the rider. Ask about payment terms before the first shift, get them in the contract rather than in a chat, and treat a fleet that will not put its schedule in writing as the risk that it is.