Percentages slide off the brain, so let me make it concrete. Say you run 40 stops a day, five days a week, and 8 percent fail on the first attempt. That is a bit over three failed drops a day, sixteen a week, and somewhere near 800 a year. At a conservative $25 all-in per failure, you are watching roughly $20,000 a year evaporate into redeliveries, wasted fuel, and customers who never come back. And that is a modest failure rate on a small operation. Now flip it. Cut that failure rate from 8 percent to 4 and you have handed yourself back around $10,000 a year, most of it without driving a single extra mile. That is the prize sitting inside this one number, and it is why it deserves more attention than the fuel gauge.

Think of the day in three acts. Before the wheels turn, while they are turning, and after they stop. Nail all three and you have a last-mile route planning routine that holds up whether you are running two vans or ten.

1. The night before, or first thing: gather and import

Everything downstream depends on getting the day’s stops in cleanly. Pull every order into one list: the deliveries, the pickups, the same-day jobs that came in overnight. The fastest way to do this is to import the whole lot from a spreadsheet or straight from your order system, rather than typing addresses one at a time while the clock ticks.

This is also the moment to catch bad data. A wrong postcode or a missing flat number is a failed delivery in waiting, and it is a lot cheaper to fix at a desk than to discover on a doorstep. Clean the list now, and the rest of the day gets easier.

2. Build and optimize the routes

Now turn that list into routes. This is where a small team either saves an hour or loses one. Group stops by area so each driver gets a tight patch, sequence them for the shortest sensible run around traffic and time windows, and split the work so nobody finishes at lunchtime while someone else is out until eight.

Doing this by hand works for a dozen well-clustered stops. Past thirty it falls apart, because the number of possible orderings explodes and no human is finding the best one on a spreadsheet. Optimized routing typically saves 20 to 40 percent of drive time against booking order, which for a small team is the difference between three vans and needing a fourth. If you want the mechanics of this step, our guide on how to plan a courier route with multiple stops breaks it down.

3. Load the vans in the right order

Small teams skip this step and pay for it all day. If parcels go into the van in a random pile, the driver is digging at every stop, adding a minute here and there that stacks up to an hour by evening. A vehicle loading plan tells each driver where every parcel sits and has them load in reverse delivery order, so the first drop is right by the door and comes out first. It is unglamorous and it saves more time than almost anything else on this list.

4. On the road: status, notes, and navigation

Once the vans roll, the office needs to know what is happening without ringing every driver. Each stop should reach the driver’s phone through a driver mobile app with navigation, the parcel details, and the access notes (gate codes, buzzer numbers, “leave with reception”). As they work, drivers mark each stop done or failed with a tap, so you can see the round progressing in real time rather than guessing.

Keep the nav familiar, too. A good workflow does not force drivers into a strange navigation app. It optimizes the order, then hands each leg off to Google Maps or Waze, which the driver already knows cold.

5. Keep the customer in the loop

This is the step that separates a small team that feels professional from one that feels chaotic, and it costs you nothing once it is set up. Automatic ETAs and arrival notifications tell the customer when to expect the driver, which is the single most effective way to make sure someone is actually home. Fewer failed drops, fewer “where is my parcel” calls landing on your one office phone. Meanwhile real-time tracking means when a client rings to ask where their delivery is, you can answer in five seconds without interrupting the driver.

6. Handle the curveballs

No shift goes to plan. A customer reschedules. A pickup runs late. A same-day job lands at 2pm and needs slotting in. The whole point of a good workflow is that these are shrugs, not crises. A route that re-optimizes on the fly lets you drop the new stop into the nearest gap and re-sequence the rest, instead of rebuilding the day from scratch at the kerb. If a delivery fails, the driver logs why, and it flows back to the office to sort a redelivery cleanly.

7. Proof of delivery at every drop

Every completed stop should end with proof of delivery: a photo, a signature, a GPS-stamped record. For a small team chasing bigger clients, this is not optional. It ends “I never got it” disputes in seconds, and it is exactly the kind of professionalism that wins you the next contract. Capturing it should be one tap for the driver, not a form.

8. End of day: reconcile and review

When the vans are back, close the loop. Check that every stop is accounted for, sort any failed deliveries for tomorrow, and glance at the numbers. Route analytics show you which routes ran long, where drivers doubled back, and how your stops-per-hour is trending. Five minutes here is how next week’s routes get tighter. Small teams that skip the review keep repeating the same detour for months without ever noticing.

Your base rate is only half the invoice. Experienced couriers protect their margin with surcharges for the stuff that genuinely costs extra, and leaving these off is one of the quietest ways to bleed money. The usual suspects:

Urgency. Same-day, rush, and sub-one-hour dispatch carry a premium, often 20 to 60 percent over standard. If someone needs it there in an hour, that convenience is worth paying for. After-hours, weekends, and holidays. Usually a flat after-hours fee, somewhere around $25 to $75, or a weekend surcharge of 10 to 25 percent. Waiting time. If the pickup is not ready and your driver is stood around, bill for it. The industry norm is roughly $1 to $2 a minute after a short grace period. Oversize or heavy items. A bigger vehicle or an overweight parcel costs more to carry, so a handling fee is fair, for example a per-pound charge over 50 pounds. And fuel. A fuel surcharge, commonly 5 to 15 percent, lets you ride out price swings without rewriting your whole rate card every month.

For context on where the totals land: same-day courier jobs frequently run anywhere from $25 to $300 or more once distance, vehicle, and surcharges stack up. Specialised work sits higher still. Standard medical routes often run $30 to $45 for a short same-day drop, while STAT medical deliveries can hit $90 to $160 or more, because the timing is not negotiable.

People new to the trade tend to fixate on the hourly rate and miss that the pay model shapes almost everything.

An employed, hourly job is the steady option. You know what lands in your account, holiday and sick cover usually come with it, and a slow week is the company’s problem, not yours. The catch is the ceiling: you cannot really out-earn the rate, no matter how sharp you get.

Self-employment flips that. Every efficiency you find, every extra drop you fit in, every dead mile you cut lands in your pocket instead of the company’s. The ceiling is high. But you are also buying your own fuel, insurance, and van, and there is no one to cover a quiet spell. This is the model where “earn more per shift” stops being a nice idea and becomes the difference between a good month and a scary one.

Gig and app-based work sits somewhere in between, and it is the most variable of the lot. Pay per drop can look decent in a busy hour and dreadful in a dead one, and you are largely at the mercy of the platform’s demand. Plenty of couriers stack a couple of platforms, or use gig work to fill gaps around contract runs.

There is no single “best.” But if your income is tied to output in any way, the rest of this guide is where your raise actually comes from.

If you want to cut failures, it helps to know the handful of ways a parcel actually dies on the round. In my experience talking to couriers, they are nearly always one of these five.

Nobody is home. The classic. A signature-required parcel and an empty house, and the whole trip is wasted. Wrong or vague address. The flat number is missing, the street has three similar names, the postcode is a digit off. Access problems. A gate code that lived in an email instead of on the stop, a buzzer that does not work, a business that shut at four. No safe place, so a leave-safe instruction turns into a “carded and returned.” And timing, where the driver arrives during the one two-hour window the customer asked them to avoid.

Notice how few of those are really about driving. They are about information and communication, which is exactly why routing software helps more than people expect. The right details on the stop, the right order of stops, and a text to the customer knock out most of these before they happen.

The trouble starts the moment the operation gets serious. A Google Maps courier route runs into a set of hard limits that a growing courier simply cannot design around. Let me go through them, because each one shows up as a real cost on a real day.

The ten-stop cap is the headline. Google Maps lets you add up to ten stops on a route, full stop. A proper courier round is several times that, so you end up chopping the day into chunks of ten and stitching them together by hand, which is its own little time sink every morning.

No route optimization is the quieter, bigger problem. Even inside those ten stops, Google Maps drives them in the order you typed them. It does not work out the best sequence. So you are still the one guessing the order, and every guess that is slightly off is a bit of backtracking and a bit of fuel you did not need to spend. People fixate on the stop cap and miss this one, and this one costs more.

No mid-route re-planning. Dispatch calls at 2pm with an urgent pickup, and Google Maps has no way to fold it into your existing round and re-sequence the rest. You rebuild the plan in your head at the kerb, which is exactly when you are least able to.

No proof of delivery. There is nowhere to snap a photo, grab a signature, or stamp a time. So when a customer insists the parcel never arrived, you have nothing but your word.

No live tracking or customer heads-up. You cannot see where your drivers are, and customers get no automatic “arriving soon” message, which means more “where is my parcel” calls landing on whoever answers the phone.

No multi-driver dispatch. Google Maps plans one route for one phone. It has no idea your team exists, so there is no way to balance the work across drivers or push routes out from an office.

And no manifest import. You type or paste stops one at a time. There is no clean way to drop in a spreadsheet or pull from your order system, which at volume is soul-destroying.

None of these are flaws, exactly. They are just jobs the app was never built to do. Google Maps is a navigation tool. Courier work needs a planning tool sitting behind it.