How do I do replenishment planning in weclapp with AI, following my own purchasing rules?
In many retail businesses, reordering follows one of three paths: Either reorder points are set once in weclapp and hardly touched again. Or stock lists are regularly moved to Excel and calculated by hand. Or a separate forecasting tool is attached to the ERP, with its own data space, its own article mapping and its own subscription.
All three paths share the same blind spot: the information a good ordering decision needs is already in the ERP. It just isn't brought together in one place and isn't prepared in the language of your own purchasing rules. That is exactly what wals.pro AI 4 weclapp takes care of: an agent that does replenishment planning according to your rules.
What an ordering decision really needs
“How much of this article should I reorder?” is not a single number. The honest answer depends on several things at once:
- How much is freely available in the warehouse, after reservations?
- What is already ordered from the supplier and on its way?
- How much is already committed by open sales orders?
- How fast is the article selling right now, and how was it last year?
- What are the reorder point, target stock, minimum order quantity and pack size?
- Which supplier, which purchase price, which replenishment lead time?
weclapp knows all of this. Until now you had to laboriously click it together for each article or train a second system for it. Your assistant now gets it at a glance: a decision-ready replenishment file per article with stock, incoming supply, demand, supplier and sales velocity. For large assortments, it looks at the articles section by section.
The rules are yours
Classic replenishment tools mainly sell a statistics engine: ABC classes, seasonality, safety stock based on variance. That is valuable, but it isn't worth a separate system that has to maintain your master data a second time.
An agent can make this trade-off itself if it gets clean data and your rules. With us, your rules are written in plain text, as a work instruction (SOP) in your tenant, not as a configuration toolkit. A made-up example:
“A articles: high service level, short coverage. C articles: only order when below the reorder point. Garden seasonal articles with a surcharge during the garden season. Supplier X is blocked. Always a full pack size, never below the supplier's minimum quantity.”
The agent reads these rules, lays them over the data and proposes concrete quantities. As a neutral starting point, it also calculates the native weclapp formula: trigger on the reorder point, replenish up to the target stock, minimum order quantity, full pack size. It refines this baseline proposal according to your rules. You see every line before anything happens. The assistant does not change the reorder point or target stock on your articles.
From proposal to purchase order draft
A reorder proposal is only half the job if it turns back into manual work. That's why the process continues: from the proposal lines you have confirmed, a purchase order draft is created in weclapp, bundled per supplier, as soon as you approve the creation. The assistant takes price, currency, minimum quantity and pack size from the supply source, and quantities are rounded up cleanly.
What is created is a draft, not a confirmed purchase. Nothing goes to the supplier without you. Confirming the purchase order is a separate, deliberate step with its own approval. How approvals work is explained in the help article Approvals.
Result: one system less
The real gain isn't a prettier forecast. It's that the whole chain of stock, demand, proposal and purchase order stays in the system that already holds the truth about your goods. No second article mapping that drifts apart with every new variant. No export-import ping-pong. No additional subscription for a tool that sees only a slice of your data. This is not an attack on specialized planning tools, but a question of where the task is best placed: where the data already lives.
What it is not
It is not automatic purchasing and not a replacement for your purchasing decision: the proposal and the creation each need your yes, and the assistant works according to the rules you give it. Replenishment planning is included in the Business plan; until December 31, 2026, it is available in every workspace.
It is for online retailers who use or are introducing weclapp as their ERP, have real lead times and pack sizes, and need not another planning system but a reordering process that knows their own rules. You can find what else wals.pro AI 4 weclapp can do on the product page and in the overview of the weclapp MCP server. Connect wals.pro AI to your weclapp at ai.wals.pro or book an appointment at wals.pro/termin, and we'll look together at how this works with your real article data.