I'm going to be honest with you. We've looked at the books of over 200 Amazon DSPs at this point, and the pattern is almost always the same. The owner knows their routes are making money — they can feel it. Revenue is coming in. Drivers are delivering. But when we actually sit down and trace where every dollar went that week? There's a gap. Sometimes a small one. Sometimes $8K or $12K a month.
The problem isn't that DSP owners are bad at business. It's that nobody set up their books to give them weekly answers. Their CPA sends a quarterly P&L, maybe 45 days after the quarter ended. By the time they see that a route was bleeding or rescue pay crept up, they've been losing money for five months straight.
I wrote this because we get asked the same question constantly: "How do I actually know my profit every week?" So here's the real answer. No theory, no fluff — just the process we run on Monday mornings for our clients.
Who this is really for
If you're running 15 to 50+ routes, using QuickBooks (or willing to switch), and you're tired of guessing whether last week was profitable — keep reading. If you're a brand new DSP with 5 routes, honestly you can probably track this in a spreadsheet for now.
Why Waiting for a Quarterly P&L is Killing Your Margins
Amazon settles weekly. Your drivers get paid biweekly. Fuel hits your card every single day. Van leases post monthly. But your CPA? Forty-five days after the quarter closes. Maybe sixty.
That's not a financial system. I've started calling it "the hope method" because you're basically hoping nothing went wrong for 90 days.
Here's what actually goes wrong during those 90 days. I'm pulling these from real DSP files we've onboarded — not hypotheticals:
- A route in north Dallas started needing rescues three times a week instead of once. The dispatcher knew but didn't flag it to the owner. By the time it showed up in the books, it had cost $7,200 over three months. Weekly tracking would've caught it at $600.
- A driver in Orlando was clocked into two overlapping shifts — his regular route and a "rescue" that was really just him sitting in a parking lot waiting. Payroll system didn't flag it because it has no concept of route assignments. Cost the owner about $1,400 a month for four months.
- Amazon adjusted two settlement line items — a fuel surcharge credit and a CX deduction — and the bookkeeper just booked the net deposit as revenue. The owner paid taxes on $3,800 he never actually received. We caught it eight months later during cleanup.
- A fuel card was being used for personal fill-ups. $180 a month. It showed up as "fuel expense" and nobody questioned it because fuel is supposed to go up during peak. Except peak ended in January and the spend never dropped.
Add that up across a typical DSP we onboard: $4,000 to $12,000 a month in leaks that a weekly look at the books would've caught in the first seven days.
The 5 Steps We Run Every Monday Morning
This is our actual process. We do it for every DSP client we have, and honestly, you can do it yourself if you're willing to put in about two hours a week. It's not complicated. It's just disciplined.
Pull the Amazon Settlement Report (Wednesday)
Amazon publishes your weekly settlement in the Logistics portal every Wednesday. Download the CSV. This is your revenue truth — not the bank deposit, not what you think you earned. The settlement. Look at the line items: base route compensation, fuel surcharges, performance bonuses, any CX deductions, rescue credits. If there's a manual adjustment line you don't recognize, dig into it before you move on. We've seen Amazon quietly adjust things here that cost owners thousands when nobody catches them.
Pull Payroll for the Same Week
Grab the payroll register from Gusto, ADP, Paychex, QB Payroll — whatever you run. The key here is matching it to the exact same 7-day window as the settlement. This is where a lot of people trip up because payroll runs biweekly or semi-monthly, and the periods don't align cleanly with Amazon's settlement week. You have to prorate. It's annoying. But if you don't, your cost-per-route numbers are garbage.
Reconcile Settlement vs. QuickBooks
This is the step most bookkeepers skip entirely, and it's the most important one. You need to match every single line item on the Amazon settlement to a revenue entry in QBO. Not just the deposit amount — the individual components. Because that deposit includes fuel surcharges that should be offsetting your fuel costs, not inflating your revenue. It includes deductions that need to hit expense accounts. We find discrepancies on literally every new DSP we onboard. Every one. Usually it's a few hundred bucks a week, but it compounds.
Map Every Cost to Its Route
This step is what turns boring bookkeeping into something actually useful. You take your dispatch manifest — the route assignment sheet — and use it to map each driver's wages to the route they actually ran. Fuel goes to the van, and each van runs a specific route. Rescue pay gets tagged to the route that was rescued, not the driver who did the rescuing. That distinction matters a lot because it tells you which routes are creating the problem. In QuickBooks, use Classes for this. One class per route. It takes some setup the first time, but after that it's just tagging.
Run a P&L by Class
In QBO, go to Reports → Profit & Loss by Class. Set the date range to the settlement week. What you'll see now is something most DSP owners have never seen: the actual profit or loss for every single route, broken out. Compare it to last week. Did Route 14 drop from 11% margin to 4%? Why? Did Driver Martinez cost $200 more than usual? Was it overtime or rescue? This report, run every Monday morning, is your profit scoreboard. It's the whole ballgame.
The 6 Numbers That Actually Matter
Once you're doing this weekly, don't drown in data. Focus on six KPIs. These are the ones we've found actually predict whether a DSP is making money or slowly dying. We surface all six in every scoreboard we send.
| KPI | What It Really Tells You | When to Worry |
|---|---|---|
| Revenue per Route | How much Amazon is actually paying you per route after adjustments | When it drops week-over-week without an obvious seasonal reason |
| Cost per Package | Your all-in cost divided by packages delivered — the purest efficiency number | When it creeps above $1.80 on standard routes |
| Rescue Pay Ratio | Rescue pay as a % of total driver wages | Anything above 8%. We've seen DSPs at 15%+ and they had no idea |
| Payroll as % of Revenue | Total driver payroll divided by total settlement revenue | Above 55% is tight. Above 60% means something structural is wrong |
| Fuel Cost per Van | Weekly fuel spend per vehicle | Sudden spikes, or one van consistently 30%+ higher than fleet average |
| Net Margin per Route | The actual profit for each route after all costs | Any route below 5% for two consecutive weeks needs attention |
Four Mistakes We See on Almost Every DSP File We Inherit
Booking the bank deposit as revenue
This one makes me genuinely frustrated because it's so common and so easy to fix. When Amazon's weekly deposit hits your Chase or Wells Fargo account, QuickBooks auto-imports it as income. But that deposit is not your revenue. It's the net result of base compensation, adjustments, credits, and deductions all mashed together. If you're booking the gross deposit as revenue, every single ratio you calculate from there is wrong. Your CPA's tax return is probably wrong too. You might be paying taxes on money Amazon deducted from you.
The fix: turn off auto-categorization for Amazon deposits. Reconcile from the settlement, not the bank feed.
Lumping rescue pay into "Driver Wages"
If your chart of accounts has one big line called "Driver Wages" and rescue pay lives there, you can't see it trending up. And it always trends up — slowly, week by week, until it's eating 10% of your payroll and nobody notices because the total wages number still looks "about right."
The fix: create a sub-account in QBO called "Rescue Pay – Driver Wages." Book every rescue shift there. We had a Dallas DSP owner who discovered $3,100 in monthly rescue pay he didn't know about just by splitting this account. He restructured two routes the following week.
Using a bookkeeper who's never seen a settlement report
I don't want to trash generalist bookkeepers — they're fine for most businesses. But a DSP is not most businesses. When your bookkeeper sees a $1,200 Amazon adjustment and books it as "Miscellaneous Income," that's a real problem. That adjustment might be a fuel surcharge credit that should be netting against your fuel expense. Or it might be a CX penalty that should be hitting a deductions account. The categorization matters because it changes your KPIs, your tax liability, and your understanding of what's actually happening in the business.
Honestly, if your bookkeeper has never opened an Amazon Logistics settlement CSV, you're paying them to categorize transactions without understanding what they mean. At 8–12% margins, you don't have room for that.
Tracking monthly instead of weekly
Monthly is better than quarterly, I'll give you that. But Amazon settles weekly. Your biggest leaks — rescue pay, unapproved overtime, fuel anomalies — compound weekly. A $600/week rescue pay problem costs $2,400 by the time you see it on a monthly report. With weekly tracking, you catch it at $600 and fix the root cause the next Monday. That's $1,800 saved. Multiply that by 12 months and you get why we're so intense about the weekly cadence.
Can You Do This Yourself?
Honestly? Yes. Everything I described above, you can do with QuickBooks Online, a spreadsheet, and about two hours on Sunday night or Monday morning. A lot of our most successful clients started exactly this way before hiring us.
Where it gets hard is consistency. Week one, you're motivated. Week three, peak is ramping and you're short two drivers and the dispatch board is on fire. Week six, you haven't reconciled in a month and now you've got three weeks of unmatched settlements. By week ten it's a mess and you're back to praying.
That's the real reason people hire us. Not because the process is complicated — it's because doing it every single week without exception, across 20 or 30 or 50 routes, while also running the actual business, is a lot. We've built the muscle for it across 200+ DSP files. We know where to look, what to flag, and how to build the Classes and accounts so the reports come out clean.
| Doing It Yourself | DSP Scoreboard | |
|---|---|---|
| Time per week | 2–3 hours | Zero — we handle it |
| QuickBooks setup | You build the chart of accounts and classes | We migrate and build a DSP-specific COA on day one |
| Settlement reconciliation | You match CSVs manually | We reconcile every line item, flag anomalies |
| Route-level P&L | You set up QBO classes | Pre-built, in your inbox every Monday at 7 AM |
| Cost | Your time + $30/mo QBO | From $299/month (beta pricing, 50% off for life) |
Either way, the point is to stop flying blind. Whether you DIY it or work with us, weekly profit tracking is probably the single highest-ROI thing you can do for your DSP right now. Higher ROI than buying another van. Higher ROI than hiring another dispatcher. Because you can't optimize what you can't see.
Start This Week — Here's Your Checklist
If you want to do this yourself starting right now:
- Open your QuickBooks Online file. If you don't have one, sign up — Simple Start at $30/month is enough for most DSPs.
- Go to Chart of Accounts and create a sub-account under wages called "Rescue Pay – Driver Wages." This alone will start giving you visibility.
- Download this week's settlement from the Amazon Logistics portal. Don't look at the bank deposit — look at the settlement CSV.
- Match every line on the settlement to a QBO entry. Anything that doesn't match gets investigated.
- Set up one Class per route in QBO. Start tagging expenses. It takes a couple weeks to get the habit, but once it's running, the P&L by Class report is transformative.
- Run that P&L by Class on Monday morning. That's your first scoreboard.
If you get to step 3 and think "I don't have time for this on top of everything else" — that's a perfectly valid realization. It's exactly why we exist. We'll do steps 1 through 6 every single week, and you'll get the finished scoreboard in your inbox Monday at 7 AM. No exceptions. No "we'll get to it next week."
Want us to find your leaks for free?
We'll look at your last 30 days of operations — settlements, payroll, fuel — and show you the exact dollar amounts leaking out. No cost, no strings, no 45-minute sales pitch. Just the numbers. If they're bad enough to fix, we'll talk about working together. If they're fine, you'll know. Claim your free audit here.