Million + AI: the layer Million themselves will not ship
Million is the desktop accounting software thousands of Malaysian sole-traders and small shops picked because it was simple and affordable. The vendor will not ship an AI layer. Here is what one looks like, on the database file the bookkeeper has been running for years.
minidesk · Editorial team, minidesk
Million is the desktop accounting software thousands of Malaysian sole-traders and small shops picked because it was simple, affordable, and did the job. The customers are the small-business backbone of the Malaysian economy: the sundry shop, the tuition centre, the F&B outlet, the home-based distributor. The customers are loyal. The vendor's roadmap is the ledger, not the AI. Here is what the AI layer looks like, and what it does for a sundry-shop owner in Klang who has been running the same Million file since 2015.
What "AI on Million" actually means
Million is a desktop accounting suite. The chart of accounts, the general ledger, the AR, the AP, the bank reconciliation, the SST submission. The data lives in a local Microsoft Access .MDB file or a SQL Server database on the shop owner's PC. There is no public API. The product was built in an era when "integration" meant exporting a CSV and importing it into something else, and the vendor has not changed the shape.
When we say AI on top of Million, we do not mean a chatbot that lives in a tab next to the ledger. We mean an assistant that reads the same .MDB / SQL Server file the desktop software reads, answers your questions in plain language, and writes the answers into the WhatsApp thread your customers already use. The AI is not a new system. It is a new way to talk to the system your shop has been running for ten years.
Three concrete use cases, in plain language
1 · The sundry shop in Klang
Mr. Lee runs a sundry shop in Klang. He has been on Million since 2015, and the .MDB file on his shop PC has 480 active customers and 6 years of sales history. His two part-time staff handle the counter, the WhatsApp, and the monthly reconciliation. Every weekend, Mr. Lee spends his Saturday morning reviewing the AR aging, drafting WhatsApp messages to the 20-30 customers who owe him money, and sending the messages one at a time. By the time the last message is sent, the first customer has already replied with a question that needs another AR lookup.
With the AI layer, the Saturday morning cycle changes. The AI reads the Million ledger, finds every invoice over 30 days past due, drafts a WhatsApp message to each customer in Bahasa Malaysia or English (whichever the customer has been writing in), and shows Mr. Lee the queue of drafts before anything goes out. Mr. Lee reviews, approves, the messages go out in 2 minutes instead of 2 hours. The customers reply, the AI reads the reply, and the follow-up note is written back to the Million debtor record. The Saturday morning is over by 9:15am, not 11:30am.
2 · The tuition centre in PJ
Kak Ros runs a tuition centre in Petaling Jaya with 120 students. Every month, the centre sends a fee reminder WhatsApp to the parent of every student whose fee is unpaid. Today, the reminders are drafted by hand: Kak Ros opens Million, exports the unpaid list, opens WhatsApp, drafts each message individually, and sends. The whole cycle takes her Sunday afternoon. By Monday morning, half the parents have already replied with a question that needs another lookup.
With the AI layer, the monthly reminder cycle is automated. The AI reads the Million AR, drafts a WhatsApp for each parent in Bahasa Malaysia or English, and shows Kak Ros the queue. She reviews, approves, the messages go out. The parents reply, the AI reads the reply, and the follow-up note is written back to the Million record. The Sunday afternoon is free. The Monday morning starts with one less thread, not one more.
3 · The F&B outlet with weekend revenue spikes
Mr. Tan runs a small F&B outlet in Subang Jaya. Friday and Saturday nights are 4x the revenue of a weekday, and the monthly AR (mostly corporate catering customers who pay on 30-day terms) reflects that. Every Monday morning, Mr. Tan spends an hour pulling the weekend's catering invoices, drafting WhatsApp reminders for the corporate customers whose payment is due, and sending the messages. The hour is the difference between starting the week on the front foot and starting the week behind.
With the AI layer, the Monday morning cycle is one prompt. Mr. Tan opens the WhatsApp chat: "draft reminders for every corporate customer whose catering invoice is due this week." The AI reads the Million AR, drafts each message in the customer's language, and shows Mr. Tan the queue. He approves, the messages go out, the follow-up notes are written back to Million. The Monday morning is done by 8:30am, the corporate customers get their reminder on the same day every week, and Mr. Tan starts the week on the front foot.
Why Million themselves will not ship the AI
Million is a small-team Malaysian accounting product. The product has been in the market for two decades, and the customer base is loyal. The vendor's investment is in keeping the desktop current, not in shipping a category of feature the desktop was never designed to support. The AI layer is a different product category from the ledger — the ledger is the chart of accounts and the sales history, the AI is the thing that reads all of it and answers the questions the chart of accounts cannot answer in a single screen. The vendor will not ship it, because the vendor's roadmap is the ledger. The AI is the layer the customer adds, and the layer is what we are writing about.
“I have been on Million for ten years. I am not migrating. I just want my customers' WhatsApp to know what my ledger says. That is what the AI gives me, and it gives me that without changing a single chart-of-accounts line.”
— Mr. Lee, who runs a sundry shop in Klang
What a normal Tuesday looks like with the AI on
It is 7:42am on a Tuesday. Mr. Lee is at the counter, the staff are opening the shop, the WhatsApp is pinging with overnight messages from customers. Before the AI, the morning was the same every day: open the ledger, check the bank balance, open WhatsApp, start answering. After the AI, the morning is one prompt and a coffee. The AI reads the Million ledger, drafts a 3-point summary (who paid overnight, who is past due, which customer follow-ups are queued), and shows Mr. Lee the queue. He approves, the WhatsApp is up to date, the ledger is clean, the bank feed is reconciled. The morning is done by 7:46am.
What it costs, in real money and in real time
The AI plan is a flat monthly fee — the number is on the pricing page, the quote is the quote, no per-message markup, no per-prompt surcharge, no enterprise tier. The translator for Million is included. The WhatsApp Business API access is included. The Meta conversation fees are the same fees every platform pays, and we do not add a per-message markup on top. The only other line item is the AI usage, which is metered in the same units every modern AI tool uses, and which your bookkeeper will not notice unless she is doing something unusual.
The time saving is harder to put a number on, because the time is not a single block — it is a thousand two-minute savings across a month. The Saturday morning WhatsApp cycle that used to take 2 hours. The Monday morning catering reminder cycle that used to take an hour. The "any update on my invoice" reply that used to take fifteen minutes per customer. The end-of-day reconciliation that used to take 30 minutes. The AI takes each one of those from minutes to seconds, and the sum is the difference between a small shop that is drowning in WhatsApp and a small shop that is using WhatsApp to grow.
How to add the AI to your Million
The install takes about twenty minutes. The translator — a small Windows program — sits next to your existing Million install on the same PC. It reads the same .MDB / SQL Server file the desktop software reads. Nothing is uploaded to the cloud. minidesk holds the AI and the WhatsApp layer. You sign in, the translator talks to the AI, the AI reads your ledger. You see the answers in the minidesk console and in your WhatsApp. That is the whole shape.
- 1Sign up at minidesk.co and pick Million as your connector.
- 2Install the translator on the PC that already runs Million (the install is signed, the size is small, the firewall prompt is one click).
- 3Connect your business WhatsApp number through Meta Embedded Signup — minidesk holds the Meta Business verification, so the typical 2–4 week wait is gone.
- 4Ask the AI a question in plain language. The first answer takes a minute because the translator is mapping your chart of accounts. After that, every answer is under three seconds.
- 5Hand the AI a workflow ("text every customer over 30 days every Friday at 4pm") and it does it, shows you the queue, and asks before anything goes out.
If your shop is already running on Million, the AI layer is the smallest change you can make to the largest difference in how the day feels. The ledger stays the ledger. The bookkeeper stays the bookkeeper. The Saturday morning WhatsApp cycle is over.
→ See the Million integration: /integrations/million