Why the phone is the primary device here, not the second screen
In most software markets the mobile app is a companion — a way to check something between meetings. In Tanzania it is frequently the only device the business owns. The shop has a phone, the delivery rider has a phone, the site supervisor has a phone, and the "computer" is a laptop that lives at home and comes out for month-end. Designing the app as a cut-down version of the web product would therefore be designing for the wrong user.
Two realities shape everything about how the app behaves: connectivity is intermittent rather than absent, and mobile data is a real, felt cost that people ration. Those two constraints explain most of the design decisions below.
How offline mode actually works
"Works offline" is a phrase used loosely, so here is the precise behaviour. The app keeps a local copy of the data you need to trade — your product list with prices, your customer list, and your recent transactions. When you make a sale, record a delivery or capture a timesheet without a connection, the record is written to the device immediately and queued. When a connection returns, the queue is sent in order and the server confirms each item.
Three things follow that are worth understanding before you rely on it:
- Stock levels shown offline are as at last sync. If two tills sell the same last item while both are offline, both sales succeed on the device and the discrepancy surfaces at sync. This is unavoidable in any offline system; the practical mitigation is that tills sync as soon as signal exists, so the window is minutes rather than hours.
- Fiscal receipts are transmitted when connectivity returns, not at the moment of sale, because TRA transmission requires the network. The sale is recorded and receipted locally, and the fiscal transmission completes on reconnection. Check that your pending queue is clearing before you close for the day rather than discovering a backlog at month-end.
- Sync is not backup. A phone that is lost before it syncs takes its unsynced queue with it. Sync at least whenever you have signal, and never leave a device unsynced overnight.
What each role actually does on the phone
The app is not one screen for everyone. What a user sees follows their role, which matters both for usability and for control.
The owner wants three numbers on opening the app: today's sales, how that compares to a normal day, and anything that needs a decision — a stock-out, an unusually large transaction, an overdue invoice. Anything beyond that belongs on a bigger screen. The cashier or attendant needs a fast sell flow and nothing else; every extra option on that screen is a mistake waiting to happen during a queue. The field user — driver, technician, agent — needs to capture: a signature, a photo, a reading, a timesheet, a delivery confirmation. The store keeper needs stock lookup and receiving.
Restricting each role to what they need is the main defence against error, and a secondary defence against fraud. A cashier who cannot access cost prices cannot leak your margin; an attendant who cannot edit a completed sale cannot quietly reverse one.
Data cost, battery and low-end devices
The app is built to work on entry-level Android hardware, because that is what most Tanzanian teams actually carry. Sync transfers changes rather than whole datasets, which keeps ordinary daily use in the range of a few megabytes rather than tens. The heaviest consumers are photographs — PODs, site evidence, product images — so where bandwidth is scarce it is worth agreeing internally how many photos a job really needs. Two good photos settle a dispute; twelve just cost money.
Practical field advice: a phone used as a till all day will not survive on its own battery. Budget for a power bank per device. It is a small cost that prevents the specific failure of a till going dark at 4pm on the busiest day of the week.
Security when the device is shared
The realistic threat to a business phone in Tanzania is not remote attack. It is the device being lost, stolen, or handed between shifts without anyone logging out. Three habits cover most of the risk:
- Every person has their own login, even on a shared handset. Shared credentials destroy the audit trail, which is the whole point of having one.
- Log out at handover, and make it part of the closing routine rather than a policy nobody follows.
- Revoke access the day someone leaves. Access is managed centrally, so a departing staff member's account can be disabled from the office without touching the phone — but only if someone actually does it.
Business data lives on the server, not only on the handset, so a lost phone is an inconvenience rather than a data loss — provided it was synced. Our broader controls are described on the security page and in the privacy policy.
What the phone is genuinely not good for
Being honest about limits saves everyone time. Bulk data entry — loading a thousand products, importing an opening stock count — is miserable on a phone and should be done on a computer, or handled by our team during onboarding. Detailed financial analysis, multi-column reports and reconciliation work want a wide screen. Payroll runs, chart-of-accounts configuration and permission design are office tasks. And initial setup, particularly VFD certificate configuration, is far easier on a laptop.
The right split for most businesses: the phone is for capturing what happens as it happens, and for the owner's daily glance. The browser is for configuring, analysing and closing the month. Trying to do everything on one or the other is where frustration comes from.
Getting set up
Install from the Google Play Store and sign in with your existing Tawala account — the app does not create a separate account or a separate dataset. On first sign-in, allow the initial sync to complete on Wi-Fi if you can; it is the largest single transfer the app will ever make. Then set each user's role before letting anyone trade, not after.
If you are new to Tawala entirely, the getting started guide covers account setup, products and your first sale, and the automation guide covers the alerts that will land on this app once you are running. The modules people use most from a phone are point of sale, inventory, CRM, wakala management and timesheets. Pricing, including which tiers include which modules, is on the pricing page, and there is a 14-day free trial. Questions about a specific device or a specific site's connectivity are best sent to us on WhatsApp.
