Why your spreadsheets are lying to you about FBA inventory
Most people trying to track their FBA stock end up maintaining a Google Sheet that looks like it was built by someone who's never actually done a reorder. The numbers drift. The receipts get lost. Three months into selling, you realize you ordered 400 units based on sales data from six weeks ago, and now you've got 1,200 units sitting in a warehouse that won't sell through in time for the Q4 push. I've been doing this since 2016. The early days involved nothing more than a well-structured Excel file and a lot of guesswork. Then came tools that promised to solve everything. Some worked. Most didn't. The ones that worked did so by being boring and narrowly focused rather than feature-rich.
What an Amazon Fba Tracker Actually Does
At its core, a proper FBA tracker just needs to answer three questions: how many units am I moving per week, how many are currently in transit or at the fulfillment center, and when will I run out if I stop ordering right now. Everything else is decoration. The real value comes from connecting those three data points into a reorder alert that doesn't trigger at 47 units remaining when you need 300. That mismatch is what kills margins. You sit on dead stock for two months waiting for a sale that never comes because the algorithm decided your demand spike was seasonal when it was actually a one-time promotion. Here's what I actually use day to day. It's not a single tool. It's a small stack: Helium 10 for keyword and demand data, a custom tracking sheet that pulls from Amazon's inventory reports via API, and a simple reorder calculator I wrote in Python that handles lead time variability. The total time spent maintaining this setup is roughly 20 minutes per week once it's running. Before I had any of it, I was spending three to four hours every Friday just reconciling numbers that didn't add up.
Building something that actually works
Let me walk through the minimum viable setup. You'll need access to Amazon's Seller Central inventory reports. Go to Reports > Fulfillment by Amazon and download the Management Inventory Report and the Estimated FBA Service Fees Report. These are your ground truth. Every other dashboard pulls from these, which means if your dashboard disagrees with them, trust the raw reports. Here's a specific problem I ran into that took me three weeks to figure out. Amazon's Inventory Adjustment report occasionally merges adjustments from different warehouses into a single line item. If you're tracking per-FNSKU, this makes it look like you lost 40 units of one ASIN when actually 20 were moved between two regional fulfillment centers as part of their inventory optimization. My tracker was showing phantom losses that made me overorder by about 15 percent. The fix was to cross-reference the Movement Report and filter out any adjustment flagged as "inventory transfer" rather than "inventory loss or damage." Once I added that filter, my reorder quantities dropped by nearly a quarter on the affected SKUs. Now the actual tracking mechanism. Set up columns for: SKU, FNSKU, current Amazon-fulfilled inventory, units in transit, units at supplier, average weekly sales over the last 12 weeks (weighted more heavily toward the last four), and your reorder point. The reorder point formula that actually works is: average weekly sales multiplied by your total lead time in weeks plus a safety buffer. Lead time includes supplier manufacturing, shipping, customs, and Amazon receiving. That last one is the most variable. A standard container from China to a US FC can take anywhere from 35 days to 70 days depending on port congestion and Amazon's acceptance queue. Don't use 30 as your default.
Get the Full Details
For the weekly sales calculation, don't just average the last 12 weeks. Weight the most recent four weeks at 50 percent of the total, the middle four at 30 percent, and the oldest four at 20 percent. Demand trends matter more than historical averages, and this weighting captures that without overreacting to a single anomalous week. When you're evaluating tools versus building your own system, here's the thing nobody tells you. Most third-party trackers update their data somewhere between once and four times per day. Amazon's API allows more frequent updates, but the tools that charge you $80 to $200 a month don't use it. They batch requests to keep their own infrastructure costs down. Your reorder alerts will be stale by several hours. If you're running thin on stock and a competitor drops their price, you might not know until the next sync cycle. I built a lightweight script that pulls directly from Amazon's SP-API every six hours and flags anything within 14 days of stockout. It cost me about two weekends to set up and runs on a $5 monthly digital ocean droplet. The ongoing maintenance is minimal. I check it twice a week. The alerts are accurate within hours, not days.
Common mistakes that waste money
Tracking only what's currently in Amazon's warehouses and ignoring inbound inventory is the most expensive beginner mistake. You see 200 units on hand, order 300 more because your calculator says so, and then discover the 300 you already ordered three weeks ago is still stuck at the port. You now have 500 units coming in while your tracker told you to order 300 more. That's 800 units you didn't need. The second mistake is treating all SKUs the same. A product with a 4.6-star rating and 200 reviews behaves completely differently than a new listing with 3.8 stars and 12 reviews. Your tracker should factor in velocity, not just raw sales numbers. Slow-movers tie up capital. Fast-movers need aggressive reorder points. A single reorder algorithm for your entire catalog will under-order winners and over-order losers. Here's a counter-intuitive point about BSR tracking. Most people think a dropping BSR means higher sales and should trigger earlier reorders. That's backwards half the time. A BSR drop can mean Amazon is running a lightening deal on your product that you didn't set up. Those deals move inventory fast but end quickly. If you reorder based on that spike, you'll have excess stock two weeks later when the deal is over and velocity returns to normal. Cross-reference any sudden BSR movement with your promotions report before adjusting reorder quantities.
When tracking tools fail you
They fail when you're selling on multiple marketplaces. The US-only trackers don't account for your Canadian or UK inventory allocation. If you're using Pan-EU or Worldwide Fulfillment, your US tracker shows zero inventory while 600 units are sitting in Germany. You'll oversell in the US because the tool thinks you're out of stock. They also fail when Amazon changes their report format. This happens more often than Amazon advertises. I've lost two full tracking cycles because a column header shifted in the Inventory Adjustment report and my parser broke silently. Always validate that your data pipeline is actually pulling new numbers every day. A broken import that returns last month's data is worse than no import at all because it gives you false confidence. If you want something ready-made instead of building your own, InventoryLab and SellerBoard both handle FBA tracking well. SellerBoard is particularly strong on the demand forecasting side. It's not free, and it won't catch every edge case I described above, but it'll save you the development time. The trade-off is you're dependent on their update schedule and their pricing goes up as your SKU count grows.
The spreadsheet approach works if you're managing fewer than 20 SKUs and you're disciplined about downloading fresh reports every Monday and Friday. Beyond that, the manual process becomes unsustainable and the error rate climbs noticeably.
