Back to Blog Listing

The 3PL-to-ERP gap and what it costs eCommerce brands per quarter

The 3PL-to-ERP gap and what it costs eCommerce brands per quarter
Nicole Ogonowska Jul 11, 2026 5 min read

Written by: Nicole Ogonowska, IT Growth Manager, Digital Colliers

Ask your 3PL how many units of SKU-1043 are on the shelf. Ask your WMS. Ask your ERP. You'll get three answers. Sometimes the gap is a rounding error. Often it's material enough that finance closes the quarter with a stock figure the ops team doesn't recognise, and ops ships against a picture finance doesn't trust.

That gap has a price. It shows up as slower cash conversion, cautious reorder decisions, and a category manager who won't run a promo because they can't confirm coverage. In a market where UK eCommerce grew only around 3% in 2024, being cautious about the wrong SKU is expensive.

The three inventory truths

Most mid-market brands are running at least three systems that each believe they hold the source of truth for stock.

  • The 3PL portal shows what the warehouse physically has, minus whatever the pickers haven't yet scanned out.
  • The WMS (sometimes the same system, sometimes a layer above) tracks bin-level state and in-flight movements.
  • The ERP or finance system holds the number the CFO closes the books against, usually fed by nightly or weekly batch.

Each system is right about something and wrong about something else. The 3PL knows what's on the shelf but not what's been invoiced. The ERP knows what's been invoiced but is running on yesterday's snapshot. The WMS knows what's in motion but not always what's been financially recognised.

Why they disagree

The drift comes from timing, taxonomy and returns. Timing is the obvious one. Batch feeds mean the ERP sees a version of reality that's 12 to 36 hours stale. If you ship a lot on Mondays, your Tuesday finance report is fiction.

Taxonomy is subtler. The 3PL uses its own SKU codes for cartons and inners. The ERP uses your master SKU list. Somewhere there's a mapping table, and if a new bundle SKU was added last quarter and nobody updated the map, that unit is invisible to one system.

Returns are where it gets ugly. Online return rates run around 19 to 20% of gross sales, and UK apparel sits at 25 to 40% depending on category. A returned unit lives in limbo for days. Physically at the 3PL. Not yet graded. Not yet restocked. Not yet credited in the ERP. Multiply that across thousands of units and you have a permanent shadow of inventory nobody wants to commit to.

What the reconciliation actually looks like

In most teams I've seen, the reconciliation is a person. Someone in finance or ops pulls a CSV from the 3PL on a Friday, pulls another from the ERP, and spends half a day in a spreadsheet chasing variances. If the variance is under some tolerance, they sign off. If it's over, they escalate, and the escalation dies in a Slack thread because nobody can prove which system is right.

That process was tolerable when margins were healthier. It's less tolerable now. Customer acquisition cost across DTC brands is up roughly 40% since 2023, and around 30% of SKUs at a typical multi-channel brand already lose money per order after returns and ad spend. If you're paying more to acquire a customer and shipping a unit you didn't know you had, or worse, telling them you have stock you don't, the economics compound the wrong way.

The cash conversion cost

The gap shows up in three places on the P&L.

  1. Working capital sits in reorders you placed against pessimistic stock figures.
  2. Revenue leaks through oversells and cancellations when the ERP thought you had 40 units and the shelf had 12.
  3. Promo revenue that never happened, because the merch team wouldn't commit without confirmed coverage.

Each one is small per event. Over a quarter, across a few hundred SKUs, brands routinely find seven-figure swings hiding in the reconciliation gap.

How to shrink it

The operators pulling ahead in 2026 aren't buying a new ERP. They're doing three cheaper things.

First, they move the 3PL-to-ERP feed from nightly batch to something closer to event-driven. Every ship confirm, every receipt, every return grade fires an event. Reconciliation stops being a Friday task and becomes a continuous drift signal.

Second, they own the SKU map. The mapping between 3PL codes, WMS bins, ERP master, and marketplace listings lives in one place, versioned, with a person accountable for it.

Third, they treat returns as first-class inventory state. In-transit, received, graded, restocked, written off. Each state is visible to finance and ops at the same time.

None of this needs a rip and replace. It needs a small integration layer, clear ownership, and the discipline to stop trusting the Friday spreadsheet. Brands that don't get there in 2026 will keep closing quarters against numbers they can't defend, while the ones who do will be redeploying cash the rest of the market has locked in reconciliation limbo.

Related Posts