Selling in ZiG and US dollars: what a Zimbabwean shop needs from its till

By Grace Wanjiru · Updated August 15, 2026 · 10 min read · Practical guide · Zimbabwe

A customer buys three items, hands you a twenty dollar note, and you owe them change you do not have in the right currency. You do the sum in your head at a rate you half remember, give back a mix of ZiG and coins, write something on the back of a receipt, and move on because there are four people waiting. Multiply that by ninety customers and you have the reason your cash never balances at closing. Trading in two currencies is not an accounting problem. It is a counter problem, and it needs to be solved at the counter.

The rate, and where it comes from

Zimbabwe has traded in more than one currency for years, and since the ZiG arrived in April 2024 the pairing that matters for most shops is ZiG against the US dollar. The Reserve Bank of Zimbabwe publishes rates every working day. To give you a sense of scale rather than a number to reuse: on 14 August 2026 the published USD to ZWG average sat at 26.5673, with a bid of 25.9031 and an ask of 27.2315. By the time you read this it will have moved, which is exactly the point.

Whether you may price above the official rate, and by how much, has been the most contested question in Zimbabwean retail since the ZiG launched. Formal retailers have argued publicly and repeatedly that being held to the official rate while suppliers price differently squeezes them against informal traders who face no such constraint. Statutory instruments on the subject have been introduced and later repealed. We are not going to tell you where the line sits today, because it moves and because getting it wrong is expensive. Ask your accountant.

What your till owes you is narrower and does not depend on the policy debate. It has to hold a rate you set, apply that same rate to every transaction until you change it, and record on each sale which rate was used. Software that lets each cashier convert in their head is not a dual-currency system. It is a single-currency system with extra arguments.

Split payments: the feature everyone skips

Here is the test that separates a till that genuinely handles two currencies from one that merely displays them. Ring up a sale. Take part of it in US dollars and the rest in ZiG, on the same receipt. Can the system do it, convert at your stored rate, and record both parts separately?

A surprising number cannot. They will let you choose a currency for the sale, which is not the same thing at all. And the moment your system forces one currency per transaction, half of every mixed payment migrates to a notebook, and everything downstream, your stock value, your margins, your day-end, quietly stops being true.

Three more things worth demanding in a demo, in this order:

1. A price shown in both currencies on screen and on the receipt, so the customer sees the conversion rather than trusting it.
2. Change calculated in whichever currency you have, at the same stored rate, without mental arithmetic.
3. A record of the rate applied attached to the sale itself, so a dispute three weeks later has an answer.

The change problem, and the honest fix

Small denominations are the daily headache. You have twenties and no ones, or ZiG notes and a customer who paid in dollars. Shops across the region have improvised the same three answers for years: give change in the other currency, round the amount, or promise it next time.

The first two are fine as long as they pass through the till at your stored rate, so the conversion is recorded rather than absorbed. The third is where money disappears. A balance owed to a customer, scribbled on a receipt or remembered by one member of staff, is a liability you cannot see and cannot count.

The clean version of the same practice is a customer credit feature: the balance is recorded against that person's name, they can see it, you can see it, and it draws them back. This is the same mechanism shops elsewhere use for buying on account, and we cover the discipline around it in our guide to customer tabs and staff discount limits. In a two-currency shop it has a second job: it absorbs the change you cannot physically give, without either of you losing out.

Change you owe and never recorded is not a rounding error. It is an unpriced loan from your customer, and you will repay it in goodwill.

Why your day-end stops balancing

Almost every shop that starts taking two currencies watches its closing count drift. The causes are usually the same three, and none of them is theft.

First, conversions happen at slightly different rates through the day because they happen in people's heads. Second, change given across currencies never gets recorded as a conversion, so the cash moves between drawers without a trace. Third, closing is done on one converted total rather than a separate count per currency, which hides the drift instead of exposing it.

The fix is procedural more than technical. Set the rate in one place, at one time of day. Force every conversion through the till. Count each currency separately at close, and compare each against what the system says it should be. If a gap appears, you now know which currency it is in, which is most of the work of finding it.

Where fiscalisation fits

This is separate from the currency question and you should not let a vendor blur the two. ZIMRA states that all VAT-registered operators must fiscalise as outlined in SI 104 of 2010, and that all taxpayers are required to fiscalise in terms of Section 90 of the Income Tax Act [CAP 23:06], including those below the USD 25,000 VAT registration threshold. Fiscal devices record and transmit sales and other tax information at the time of sale to the ZIMRA servers, and hold that information in memory that cannot be altered afterwards.

So the practical question when you shortlist software is blunt: does this product have working fiscal device integration in Zimbabwe today, with a shop you can call? A general statement about being ready for electronic invoicing is not an answer to that question, and treating it as one is how businesses end up buying twice. Several vendors in the Zimbabwean market specialise in exactly this integration, and if fiscalisation is your binding constraint, that is where to start your shortlist rather than with features.

Power cuts and the offline question

Load shedding decides more about your choice of till than any feature list. A cloud system that stops selling when the connection drops is not usable here, and most vendors know it, so almost everyone advertises an offline mode.

The gap is in what offline actually covers. Selling offline is common. Being able to look up stock levels, apply your stored exchange rate, take a split payment and check a customer's balance while offline is much less common, and those are exactly the things you need when the power is out and the shop is busy. Test it rather than reading about it: switch off data and Wi-Fi, then try a two-currency sale with a stock lookup and a customer balance check. Our guide to running a POS through power cuts goes through the full checklist.

What to look for, and who does it

Broadly you are choosing between fiscalisation-first local vendors, general cloud tills, and systems built around multi-currency trading. Most shops in Zimbabwe end up needing something from at least two of those columns, so be clear with yourself about which constraint binds hardest before you shortlist.

At a glance

Compiled from public vendor documentation reviewed in August 2026 and features confirmed by the vendors. "Ask for a demo" does not mean absent: it means we could not confirm it publicly and you should have it shown to you on your own stock before deciding.

Frequently asked questions

Official sources

Currency and tax rules in Zimbabwe move quickly. Check them at source rather than taking our word for it.

• Reserve Bank of Zimbabwe: daily published exchange rates and monetary policy statements.
• ZIMRA, fiscalisation explained: who must fiscalise, what a fiscal device records and how the information reaches ZIMRA.