This is a composite case study: a typical scenario we hear from small online shop owners more than any other. The name, brand, and some details have been changed, but the automation mechanics and the scale of the numbers are real, drawn from similar projects.
The situation before automation
A small clothing shop: around 15–20 orders a day through the website, Instagram, and WhatsApp. One person runs everything alone, from talking to customers to tracking stock. The same questions came up several times a day: “where’s my order,” “do you have size M,” “when will it arrive.” That ate up about 1.5 hours a day. Another 30 minutes a day went into manually copying each new order into the inventory spreadsheet and updating stock. Once a week, another 2 hours went into pulling together a report: what sold, what’s running low, what isn’t moving at all.
Altogether, about 12 hours a week on tasks that need no creative judgment, just constant attention.
What got automated
- An AI agent on WhatsApp and the website answers the routine questions — order status, size availability, delivery time — and only brings the owner in for anything unusual.
- Every new order’s data flows straight into the inventory spreadsheet and updates stock automatically, with no manual copying.
- Once a week, the AI agent puts together its own sales and stock report and delivers it already organized, with actual conclusions instead of a raw table.
The result: nearly 10 hours a week
| Task | Before automation | After | Time saved |
|---|---|---|---|
| Answering routine customer questions | ~7.5 hrs/week | ~1.5–2 hrs/week (edge cases only) | ~5.5–6 hrs |
| Copying orders into inventory | ~2.5 hrs/week | ~10 min/week (spot checks) | ~2.3 hrs |
| Putting together the weekly report | ~2 hrs/week | ~15 min/week (just reading it) | ~1.8 hrs |
| Total | ~12 hrs/week | ~2–2.5 hrs/week | ~9.5–10 hrs |
The owner didn’t spend the freed-up time resting, though that would have been a fair choice too. Instead it went toward what had been on hold for months: refreshing the product range and negotiating with a new supplier.
What it cost
Setting up an AI agent like this isn’t a one-time purchase of an off-the-shelf bot — it’s work built around the shop’s actual processes: connecting WhatsApp and the inventory sheet, training the agent to sound right, testing it against real customer questions. What that costs, and how to even think about that cost, is its own conversation, and there’s an honest answer in “how much AI automation costs for a small business.”
When this works, and when it doesn’t
This case worked because the customer questions were predictable and the inventory process was clear and consistent day to day. If every order in that shop were handled differently, and replies to customers depended on the seller’s mood, there would be nothing to automate: the process itself would need sorting out first, before anyone could speed it up.
If your situation looks similar — routine questions eating up hours, stock tracked by hand in a spreadsheet — the simplest way to find out what’s actually worth automating is to message me on Telegram or WhatsApp. I’ll give you an honest answer, even if it turns out you don’t need this yet.
Leave a Reply