TL;DR
Multi-channel sellers in the Philippines run inventory across Lazada, Shopee, TikTok Shop, their own Shopify or web store, and one or two physical stores. Most of them keep five separate inventory states and reconcile them twice a day. The first oversell incident usually triggers the OMS conversation. Here is how we run the consolidation without breaking peak-season operations.
The shape of the problem
A typical mid-sized e-commerce operation we work with has roughly these inventory homes:
- Lazada seller center inventory
- Shopee seller center inventory
- TikTok Shop seller inventory
- Shopify store inventory
- Warehouse management spreadsheet (the actual physical count)
Five places. Four of them update on different schedules. The warehouse spreadsheet is the physical truth, but it lags everything by hours.
When a customer buys the last unit of a SKU on Shopee, three things should happen: the item ships, the warehouse count decrements, and the other four channels reflect the new zero. In practice, what happens is the item ships, the warehouse count gets manually updated at end-of-day, and one or two other channels still show stock available for ten to thirty minutes after the actual physical zero. Someone buys the now-unavailable unit. The team has to cancel and apologize. Repeat this pattern five to twenty times a week during peak season.
Why the obvious fix is risky
The obvious fix is to install an inventory sync platform that connects all five sources. There are vendors who do this. We have seen them work and we have seen them fail. The failure mode is always the same: the migration weekend.
Migration to a centralized inventory truth means freezing every channel for several hours, reconciling counts, then turning sync back on. If your peak season is November to January, you cannot afford that downtime. If you do it during off-peak, the team is not stressed enough to surface the edge cases. Either way, you find out about the edge cases on a Saturday night two weeks after go-live.
The pattern we ship instead
We do not flip everything to centralized at once. We sequence in three phases.
Phase one: pick a primary source of truth. Usually the warehouse management system, even if it is currently a spreadsheet. The OMS we deploy reads physical inventory from there and treats it as canonical. Phase one ships in two to three weeks. At the end of phase one, the OMS knows the truth but the channels do not yet defer to it.
Phase two: read-and-warn for one week. We let the OMS observe channel inventory levels and warn when they drift from the truth, but the OMS does not push corrections yet. The team sees a Slack alert when Lazada and the warehouse disagree by more than two units. They confirm or override. This is the most important week of the engagement. The team learns to trust the OMS by watching it be right repeatedly. The OMS learns the team’s edge cases.
Phase three: outbound sync. Once trust is established, the OMS starts pushing corrections to channels. Lazada, Shopee, TikTok Shop, and Shopify each get a webhook contract. We turn on one channel at a time, a week apart. If something goes wrong on Lazada, Shopee is still on the old workflow and the operation does not stop.
By the end of phase three, all four channels and the warehouse are sync’d off one source. Total elapsed time: ten to twelve weeks. No frozen weekend.
The numbers we typically see
After full rollout:
- Oversell incidents per week: from twelve to sixteen down to under one
- Inventory reconciliation labor: from four hours per day across two staff to under thirty minutes per day for one
- Order-to-ship cycle: down by 20 to 40 percent because warehouse pickers stop chasing the canonical SKU list across multiple windows
- Returns rate: down by a couple of percentage points because fewer wrong items get shipped (the SKU master is now actually canonical)
The biggest qualitative win is the team’s confidence. The customer service team stops apologizing for cancellations they did not cause. The warehouse stops getting blamed for inventory that was wrong on the channel, not in the bin.
What an OMS is not
It is not a replacement for your warehouse management. The WMS still owns the physical layout, the picking strategy, the bin assignments. The OMS sits above WMS and handles the orders-and-inventory-across-channels layer.
It is also not a replacement for your accounting system. Orders flow from the OMS to accounting nightly, not the other way around.
And it is not a magic solution if your warehouse count is dirty. We have walked into engagements where the warehouse spreadsheet was off by 8 to 12 percent against a physical recount. No amount of OMS sophistication fixes that. We do a physical recount in week one, full stop. The OMS goes live with truth, not aspiration.
Should you build or buy?
For most operators we work with, the answer is hybrid: buy the marketplace integrations (these are commodity and well-understood), build the orchestration layer (this is where your operation is unique). Off-the-shelf OMS platforms force a one-size-fits-all warehouse model. If you have a Manila hub plus three regional drop points, plus a third-party fulfilment partner for cold-chain, off-the-shelf bends awkwardly around that.
We typically ship a custom orchestration layer with vetted off-the-shelf marketplace connectors underneath. You get integration speed and operational fit at the same time.
What to do before the first OMS conversation
A few cheap audits that will sharpen the conversation:
- Pick one SKU at random. Within thirty minutes, can you tell me the inventory count from each of your five sources? If not, that is your starting baseline.
- Pull last month’s cancellation log. How many were oversell-driven? That number is your floor for the OMS ROI.
- Time the warehouse team’s morning routine. How much of the first hour is reconciliation versus picking?
If the audits surface a real operational pain (not just an aspiration to “have a single dashboard”), an OMS engagement will pay for itself within three to four quarters. If they don’t, you do not need an OMS. You need cleaner inventory hygiene first.
When you are ready to talk through the sequence for your operation, book a consultation. We bring the playbook; the first conversation is on us.