The dollar-store subscription that reads your pantry

The problem

Dollar stores are built for impulse visits. You go in for tape and leave with ten things, half of which you already had. The low prices hide a time cost: the trip itself, the mental load of scanning shelves, and the slow drain of running out of cling film on a Tuesday night.

What AI makes possible now

A vision model can now look at a photo of your open pantry or cleaning cupboard and parse what is running low. A small agent, running on cheap scheduled inference, watches your usage patterns and generates a restock list. Stripe subscriptions handle the prepaid drawdown. The product does three things: predicts what you will empty next week, bundles it at dollar-store margin, and ships it in a flat box on a rhythm you set. No browsing catalogues, no last-minute dashes to the strip mall.

How it works

1. You snap three photos of your storage areas during onboarding. The model identifies brands, pack sizes and fill levels from visual cues, then builds a baseline inventory.
2. You set a monthly prepaid drawdown amount and a delivery cadence. You approve or ignore the suggested restock list by a cutoff day.
3. The agent learns your household's depletion rate for each item. It queues refills before you register emptiness, not after.
4. You can text a photo of an unexpected empty. The system adds it to the next shipment if the credit pool allows, without forcing a separate order.

The first thirty days

Ship a single-box restock service for ten high-rotation dollar-store categories: bin bags, dish soap, sponges, batteries, tape, lightbulbs, cling film, foil, kitchen towels, envelopes. Onboarding is three photos and a prepaid balance. No customisation beyond quantity nudges. The 30-day scope is inventory inference, a fixed Tuesday packing run, and flat-rate local shipping.

How it earns

Prepaid drawdown accounts create a cash float. Margins come from the delta between dollar-store wholesale and a small convenience markup, plus shipping efficiency from consolidated boxes. Unused credit stays in the account, earning float interest.

Why now

Two things changed since 2017. Household vision models that can read a cleaning cupboard from a smartphone photo, and inference cheap enough to run usage prediction weekly per subscriber without eating margin. The rest is logistics, which Stripe and shipping APIs already solve.

First customers

Working parents who already shop dollar stores for staples, know their burn rate, and resent the trip. Also small office managers who restock kitchenettes and stationery cupboards on a recurring rhythm.

The hard part

Fulfilment is the honest hard part. Dollar-store assortments shift constantly and SKU churn is high. The agent must handle substitutions without breaking trust or requiring constant manual overrides.

Build this with us

Nexibeo can design the vision pipeline and the usage-prediction agent, build the Stripe prepaid logic, and run the full backend for you as a managed service. The same fixed monthly fee covers model choice, monitoring, infrastructure and ongoing optimisation. If a no-trip dollar-store restock sounds like something your customers would keep paying for, apply for a build slot.