Electric Bike for Delivery Riders

Every e-bike sold falls into Class 1, 2 or 3, and for delivery work the class matters for one practical reason: the throttle. Class 1 is pedal-assist only. Class 2 adds a throttle, in the US capped at 20 mph same as Class 1; Class 3 raises the pedal-assist ceiling to 28 mph but usually drops throttle-only riding. On a route with fifteen stops in an afternoon, pulling away from a dead stop on throttle alone, no clipping back into the pedals for the length of a red light, is worth more than a higher top speed you'll rarely hold in city traffic. That's the case for an electric bike with throttle over a Class 1 or Class 3 bike for most delivery work.

Those figures are the US picture. Assist limits and throttle rules differ by country, and often by city once you're doing this for pay rather than for fun, so check the actual rule for wherever you're riding before you buy, not after. PeopleForBikes keeps the state-by-state US definitions current; our own breakdown of the class system covers the regional differences in more depth than a delivery-specific guide has room for.

In the catalogue here, the Rad Power RadRunner is the straightforward Class 2 answer: a 750 W hub motor, throttle, a 320 lb payload rating built around exactly this kind of use. If the route runs longer and faster, mostly road rather than bike lane, the RadRunner Max steps up to Class 3 and a 420 lb payload rating, trading throttle-only convenience for a higher assist ceiling on the open stretches. Whether that lands you on class 2 or class 3 for delivery comes down to the route, not the badge on the frame. (Both go by 'moped-style' in the trade. No, you don't need a plate. Yes, someone will still ask.)

Range is the number every delivery rider actually cares about, and the honest one is 45 to 110 km on a charge, mixed terrain, real rider weight, not the lab figure printed on the box. Where a bike lands in that range depends on battery size, motor, and how much of the shift is spent on throttle instead of pedalling, which is the trade-off nobody puts on the spec sheet: throttle-only riding empties a pack faster than pedal-assist, every time.

Charging a typical 500 Wh pack costs roughly $0.12 to $0.20 in electricity, and a genuine pack, tested to the standard UL Solutions uses to check for overcharging and short circuits, is good for 800 to 1,000 full charge cycles before it drops to 80% of new capacity, three to five years at normal use. A delivery shift isn't normal use, it's close to a full cycle most days you work, so the right ebike battery for delivery riders is oversized on purpose, not by accident.

If one pack genuinely won't cover the route, the fix is a second battery, not a bigger bike. The Super73 M1D is dual-battery capable out of the box, roughly doubling its real range for exactly this reason, and our full look at when a second pack is worth it runs the maths in more detail than fits here.

Nobody has ever lost a shift because the bike topped out at 20 mph instead of 28. Riders lose shifts because a rack rated for a laptop bag folded under three bags of groceries, or a frame flexed under a load it was never built to carry. Payload capacity, not top speed, is the number that actually matters once the bike is loaded.

The RadRunner carries a 320 lb payload rating, rider included; step up to the RadRunner Max and that rises to 420 lb. Most bikes in the range weigh 22 to 30 kg unloaded, cargo-rated bikes 30 to 40 kg, worth knowing if you're carrying the thing up a flight of stairs to charge it overnight rather than parking it in a garage.

Brakes matter here too, more than on a leisure bike: fifteen stops an hour with a loaded rack rewards hydraulic discs, not the cheap mechanical calipers that fade halfway through a wet shift. It's not a glamorous spec. It's the one that stops you sliding into the car ahead of you the exact moment its driver decides indicating is optional.

I track the running cost of anything with wheels, which is a personality trait more than a hobby at this point. I ran the numbers on my own second car back in 2019, worked out it cost more a month than most of the bikes we sell cost once, sold the car, and never looked back. Apply the same maths to a delivery route and the case for a bike gets even more one-sided, because you're riding it for pay, not just the school run.

Riding 100 km costs roughly $0.10 in electricity against $8 to $15 for the same distance by car, once fuel's counted, and that's before parking, tickets, or the ten minutes you'll spend circling a block a bike would have skipped entirely. Run a full shift's worth of kilometres through that gap and the per-shift saving against a car is a large part of why e-bike delivery works as a business model, not just a lifestyle choice.

Here's the opinion I'll defend: the price you pay isn't the cost you carry. A cheap bike that saves four hundred dollars up front and then needs a new pack in eighteen months, on a route ridden five days a week, is the expensive option, not the bargain. Total cost of ownership is the number that actually decides whether a bike paid for itself, and it rarely shows up on the price tag.

A bike ridden five or six days a week for hours at a time finds every weakness in a battery and a motor faster than a weekend rider ever will. This is exactly the environment cut-price marketplace e-bikes fail in fastest: uncertified cells, a motor no shop stocks parts for, a brand that's three letters and a shrug the day something breaks. Losing a bike mid-shift doesn't just cost the repair, it costs the whole day's earnings.

Every bike here carries a 1-year warranty on frame, motor and battery as standard, the battery guaranteed to hold 60% of its rated capacity at the one-year mark, workshop-checked before it ships rather than drop-shipped from a warehouse that's never seen it. The CPSC's own micromobility safety guidance exists largely because of bikes exactly like the ones this section is warning you off. If it's the wrong bike for the route, the 14-day ride-it-and-see return window is there before you've committed a month of shifts to finding out the hard way.

None of that makes a bike indestructible. It makes it the kind of indestructible a Nokia 3310 was indestructible: not immune to problems, just built by someone who expected you'd use the thing every day and doesn't go quiet when you give us a shout about it.

Here's the bit most delivery-bike guides skip, because their business model is convincing you to buy something. A bike is the wrong tool for a route that's mostly rural highway, spread further apart than a real-world 45 to 110 km range covers without a charging stop. That's a scooter's job, or a car's, not a bike's, and no battery upgrade fixes a route that's simply too spread out.

Same goes if the loads are consistently bigger than a bike frame is rated to carry: bulk grocery runs, furniture, anything that needs a van's floor space rather than a rack. Buying a heavier bike to chase a payload number a van clears without trying is a way to spend real money solving the wrong problem.

And if you're doing this occasionally rather than as the main job, a few deliveries on a Saturday between a car you already own and insure, the bike doesn't pay for itself the way the running-cost maths above assumes. That maths works because the bike replaces near-daily car use. It doesn't work nearly as well replacing three trips a month.

Get the route right first. The class, the battery and the payload rating all fall out of that one decision, not the other way round.