This question usually arrives in a specific form: we already paid for an ERP with a planning module, so why are we being asked to buy something else? It’s a fair challenge and deserves a straight answer rather than a sales one.
What ERP is Genuinely Built For
ERP systems are transaction and record systems. They exist to maintain a single accurate version of what happened — orders, receipts, issues, invoices, ledger entries — with the controls and audit trail that finance and compliance require. At this they are excellent, and nothing else in your stack replaces them.
Their planning modules are typically built on the same foundations: deterministic rules applied to master data. Fixed lead times, fixed safety stock, unconstrained MRP that assumes infinite capacity. This isn’t a defect. It’s a design appropriate to a system whose primary job is accuracy of record, not optimisation under uncertainty.
Where The Gap Opens
The gap shows up wherever a decision requires probability rather than a rule. Standard MRP treats a fourteen-day lead time as fact when it’s actually a distribution ranging from eleven to twenty-six days. It plans against a forecast without any view of that forecast’s reliability. It generates requirements without regard to whether the plant can physically produce them.
You can see the symptoms in most ERP-only operations: planners exporting to spreadsheets to do the analysis the system can’t, buyers overriding suggested order quantities from experience, schedules produced by the system and then rewritten on the floor. That workaround layer is the specialised capability, currently being provided by people at considerable cost.
How to Tell Whether You Need More
Some manufacturers genuinely don’t. If your product range is narrow, demand is stable, lead times are short and reliable, and you’re not capacity constrained, rule-based planning works fine and adding a layer is expense without return.
The case strengthens with complexity: many SKUs, volatile or seasonal demand, long or unreliable supplier lead times, multiple sites, a genuine capacity constraint, or significant inventory relative to revenue. If several of those apply and your planners are running the business out of Excel, the gap is already costing you — it’s just showing up as headcount and inventory rather than as a software line.
They Should Be Layered, Not Replaced
The right architecture keeps ERP as the transactional system of record and adds planning intelligence above it. The planning layer reads live transaction data, does the probabilistic and optimisation work, and writes decisions back as orders and schedules for ERP to execute and record.
This matters practically because it avoids the worst outcome: two systems each holding a partial version of the truth. Master data should have one owner, and it should be the ERP.
The Question to Ask a Vendor
Ask exactly how the integration works, in both directions, and what happens to data ownership. Vendors who answer clearly and specifically have done this before. Vendors who describe it as seamless without detail have not.
ticktick.ai runs as a planning layer above existing ERP systems, reading live transactions and writing back orders and schedules rather than duplicating the record.
