Slurp + AI: what it actually looks like in 2026

Slurp is the cloud POS the smaller Malaysian F&B chains run — bubble tea shops, kopitiams, small restaurant groups, the kind of business that picked Slurp because it was the cloud POS that actually worked at 3 outlets, or 8 outlets, or 20 outlets, without the enterprise price. Here is what an AI layer on top of it looks like — the one that answers the per-outlet breakdown, the promo comparison, the live menu reply, without the morning spreadsheet.

What "AI" means here — the short version

When we say AI on top of Slurp, we do not mean a chatbot that lives in a tab next to your POS. We mean an assistant that reads the same Slurp database the cloud POS reads — sales, menu, inventory, customers, the lot — answers your questions in plain language, and writes back the answers into the same WhatsApp your customers already use. The AI is not a new system. It is a new way to talk to the POS you have been running for two years — the one you chose because it was the cloud POS that actually fit a small Malaysian chain.

What it actually does for a 6-outlet bubble tea chain

Imagine a bubble tea chain Mr Goh runs, 6 outlets across the Klang Valley, the kind of business that runs a new promo every month and needs to know which outlet the promo actually worked at. The Slurp POS holds every transaction, every menu item, every inventory level, every customer order, every outlet's till count. The operations manager lives in the Slurp dashboard. The customers live in WhatsApp. The Monday morning meeting starts with someone copying the per-outlet sales out of Slurp into a spreadsheet, then into a WhatsApp group, then into the promo debrief.

The minidesk AI reads the same Slurp data and does four things Mr Goh used to do himself. (1) At 8am every Monday, the AI sends a WhatsApp to the operations group with last week's sales, broken down by outlet, in Bahasa Malaysia. (2) When a customer in Subang Jaya sends a WhatsApp asking for the drink menu, the AI reads the Slurp menu, the live inventory, and the Subang outlet's specific availability, and replies in Bahasa with the menu and the pickup ETA. (3) At the end of every month, the AI drafts the per-outlet promo comparison — which outlet the brown sugar promo worked at, which outlet the new fruit tea line did not, ranked in one short WhatsApp. (4) When the AI notices a low-water mark in the Subang outlet's tapioca pearls inventory, it sends a WhatsApp to the purchasing manager with the reorder list, with the supplier contacts attached.

None of that is a different product. It is the same Slurp Mr Goh has been paying for. The AI is the new interface — the layer that lets the business ask the POS a question in Bahasa and get the real answer back, in the channel the customers already use, in the language the staff already speak.

Why Slurp will not ship this themselves

Slurp is a POS. Their product is the till, the menu, the inventory, the receipt printer, the e-invoice. An AI assistant that reads all of those at once and answers the question in WhatsApp is a different product category — and a plugin that does it is the modern version of the same wedge the POS vendor will not ship because it is not the POS vendor's job to ship it. Slurp's job is to be the best POS for the small-chain F&B segment in Malaysia. The AI's job is to be the layer on top.

What it looks like across 6 outlets on a promo Monday

The honest test of an AI on a multi-outlet POS is what happens when the same AI meets 6 outlets, 3 menu categories, 1 promo line, and 1 Monday morning in the same ask. Here is the test. Mr Goh asks at 8am on Monday, "Macam mana promo brown sugar minggu lepas, per outlet?" The AI reads all 6 outlets, slices the brown sugar promo line, ranks the outlets by promo sales, flags the 2 outlets that ran out of stock mid-promo, and writes the WhatsApp in Bahasa Malaysia, with the per-outlet breakdown, the top 3 promo items, and the reorder list — all in one short message the operations manager reads on the way to the first outlet.

“We chose Slurp because it was the cloud POS that actually fit a 6-outlet bubble tea chain. The AI is the part that makes the per-outlet view actually usable — it answers the WhatsApp in Bahasa, on Monday morning, with the promo comparison the operations manager is actually asking for.”

— Mr Goh, who runs a 6-outlet bubble tea chain in the Klang Valley

The honest limits of the AI

Three limits worth naming, because the limits are the part nobody else puts in the brochure. (1) The AI reads the Slurp data you have connected. If the menu is a mess, the answers will be longer than they need to be. (2) The AI does not write back to Slurp. Every order and every inventory adjustment stays in the POS, exactly where your cashier made it — the AI is a reader and a writer of WhatsApp, not a writer of POS. (3) The AI is not a replacement for an operations manager. It is the layer that lets them stop copying numbers into spreadsheets. The operations manager still owns the promo debrief. The purchasing manager still owns the reorder decision.

How to add the AI to your Slurp

Three steps, end to end in twenty minutes. (1) Sign up at minidesk.co, pick Slurp as the connector, and authorise the OAuth — the same flow you use to link Google Calendar to a third-party app. minidesk requests only the scopes the AI needs: read sales, read menu, read inventory, read customers. (2) Connect your business WhatsApp number through Meta Embedded Signup. minidesk holds the Meta Business verification, so the typical 2-week Meta wait is gone. (3) Hand the AI its first workflow — "every Monday at 8am, send me last week's per-outlet sales, in Bahasa" — and the first message lands that Monday. The cashier does not have to do anything different. The customer does not have to download an app. The message lands in the WhatsApp they already have open, in the language they already speak, in the menu the Slurp POS already holds.

What the first 30 days actually look like

Most F&B operators that turn the AI on for the first time expect the AI to be the whole answer on day one. The honest answer is that the first 30 days are the calibration period — the AI is learning the per-outlet sales pattern, the promo response by outlet, the Bahasa-vs-English preference for each outlet, the menu items that only run at certain outlets, the promo lines that only count at certain times. The operations manager is teaching the AI which promo lines should be on the Monday morning comparison, which suppliers the WhatsApp should go to, which low-water marks need the purchasing manager's attention. The purchasing manager is teaching the AI which inventory items have the longest lead times, which suppliers respond to WhatsApp fastest, which reorder lists can wait until Tuesday.

By day 7, the AI is sending the first 3 workflows — the Monday morning per-outlet sales comparison, the menu auto-reply, the reorder list — with no human in the loop. By day 14, the AI is handling the 80% of customer WhatsApp messages that are routine (menu enquiry, order confirmation, pickup update, opening hours) without a cashier touching the keyboard. By day 30, the AI is the layer that reads Slurp in Bahasa and English, writes the WhatsApp with the live menu, and drafts the per-outlet comparison for the Monday morning meeting — and the operations meeting starts with a WhatsApp the AI drafted at 8am, not with a spreadsheet someone built at 8:45am.

The number that matters most at the 30-day mark is the same one Mr Goh watches in his own chain: how many of last week's customer WhatsApp messages were handled by the AI, and how many needed a human. The honest target, for a Malaysian bubble tea chain running on Slurp, is 70% handled by the AI at 30 days, 85% at 60 days, 90% at 90 days. The remaining 10–15% is the part that genuinely needs a human — the promo debrief, the supplier negotiation, the new menu line decision. The AI is not trying to replace those. The AI is trying to clear the 85% so the operations manager and the purchasing manager have the time for the 15%.

Common questions F&B operators ask before they start

1. 1Does the AI replace the cashier? No. The AI is the layer that lets the cashier stop typing the same drink menu 30 times a day. The cashier still owns the till, the cash, the discount decisions. The AI is the new chat surface, not the new cashier.
2. 2Does the AI write to Slurp? No. Every order and every inventory adjustment stays in the Slurp POS, exactly where the cashier made it. The AI is a reader of Slurp and a writer of WhatsApp, never a writer of POS. The data isolation is the product (see the security page) — the wall is in the database, not in a prompt.
3. 3Does the AI draft the per-outlet promo comparison? Yes. The AI reads all 6 outlets (or however many you have), slices the promo line, ranks the outlets by promo sales, flags the outlets that ran out of stock mid-promo, and drafts the Monday morning comparison in Bahasa. The operations manager reviews, adjusts for the local factors, and sends.
4. 4Does the AI work in Bahasa Malaysia and English? Yes. The AI matches the language the customer wrote in. The reply to a Subang Jaya customer is in Bahasa. The reply to a Singapore customer is in English. The menu items are in the language the Slurp menu holds them in.
5. 5What if I am not on Slurp yet? Sign up for Slurp first (slurp.com.my), connect your outlets, and let the till settle for a week. Then sign up for minidesk, pick Slurp as the connector, and the AI starts reading the POS the moment the OAuth is authorised.

→ See the Slurp integration: /integrations/slurp