Where are my ecommerce orders being lost?

Why is "orders being lost" too vague a starting question?

Because it can mean five genuinely different things depending on where you look in the pipeline: leads that never became orders, orders that never got confirmed, confirmed orders that never shipped, shipped orders that never delivered, or delivered orders where cash was somehow never actually collected despite the parcel changing hands. Each of these has a different cause and a different fix entirely, with almost no overlap between them. Asking "where are we losing orders" without first splitting the funnel into these separate stages usually leads to fixing something that was not actually the biggest leak in the first place, simply because the biggest leak was never isolated from the smaller ones sitting alongside it.

What is the fastest way to find the answer this week?

Pull the last two to four weeks of orders and tag each one with the last stage it reached before stopping: lead only, placed but not confirmed, confirmed but not shipped, shipped but not delivered, or delivered and paid in full with nothing left outstanding. Count how many orders sit at each stage right now, today, without any further movement. The stage with the largest count of orders stuck in it, relative to how many entered that stage in the first place, is your biggest leak by a meaningful margin worth acting on first. This does not require new software of any kind to accomplish, a spreadsheet with a single status column per order is genuinely enough to answer the question with real confidence.

What does each stopping point usually indicate?

A store with a large "placed, never confirmed" bucket has a completely different fix, the confirmation process itself and how it is run, than one with a large "shipped, never delivered" bucket, which points instead at address collection and carrier performance on the ground. Guessing between these two without the actual count in front of you is how effort ends up spent repeatedly on the wrong fix month after month with little to show for it.

What if orders are stuck between confirmed and shipped?

This particular gap is worth calling out specifically because it is often invisible until someone actually measures it directly rather than assuming the pipeline is flowing smoothly: orders sitting confirmed but not yet turned into a carrier parcel for days at a time without anyone noticing. This usually means a manual step, someone has to log into the carrier's system and create each parcel by hand one at a time, is not keeping pace with confirmation volume as it grows over the weeks. It shows up as good confirmation numbers paired with a mysteriously lower delivery rate than the confirmation number alone would predict, because the real problem sitting underneath is a shipping backlog, not the confirmation or delivery process itself failing in any way.

How do you turn this into an ongoing habit rather than a one-time check?

Recount the stage breakdown weekly, using the exact same five stopping points every single time without redefining categories midway through, so the numbers stay genuinely comparable week over week as time passes. A single snapshot tells you where the leak sat last week, which is useful information but fairly limited on its own. A recurring weekly count tells you whether the fix you applied actually moved orders out of that stuck stage into the next one, which is the only reliable way to know a change genuinely worked rather than simply assuming it did based on a general feeling that things seem better lately.