Inventory Management Software for Africa: What Actually Matters
Most inventory advice written for African businesses is just inventory advice with a map on the cover. Four conditions genuinely change the priority order — long import legs, distance between locations, capture habits and custody — and they push the same three features to the top of every list, in every country.
A warehouse in Tema, a distributor in Lusaka and a retailer in Mombasa are told they need cycle counting, ABC analysis, reorder points and demand forecasting. All true, all available in a hundred products, and none of it explains why the same three problems keep appearing in businesses that already bought the software.
The honest position is that inventory management is not fundamentally different in Africa. What differs is the priority order — which features earn their cost first — and that ordering is driven by four conditions rather than by geography.
The four conditions that reorder the list
- A long, expensive leg between the supplier and your shelf. Ocean freight, customs, clearing, and often several hundred kilometres of road afterwards. The cost of getting goods in is a large and variable fraction of what they cost — far larger than in the markets most inventory software was designed for.
- Real distance between locations, with imperfect connectivity. A branch or site store that cannot reach the system at the moment of the transaction will keep its own book. Within a quarter you have two truths.
- Capture happening away from a desk. Receiving on a loading bay, issuing at a store hatch, counting in a yard, selling from a van. If capture requires walking to a computer, it happens later or not at all.
- Stock in the hands of people the system does not employ. Contractors, agents, van salespeople, field teams. Custody is a live question rather than an accounting one.
Those four are not universal across the continent and they are not exclusive to it — but where they hold, they change which capability you should insist on first, and they explain most of the disappointment with systems that were technically fine.
What actually matters, ranked
Given those conditions, this is the order. Anything below the line is worth having and is not what decides whether the implementation succeeds.
Inventory capabilities, in the order they earn their cost
Everything here should be demonstrated live, on your data. The top three are where implementations succeed or fail.
Every sale and issue moves stock in the same second
Make them prove it: Record a sale, watch the level drop, then run variance against a physical count of that item.
Landed cost folded into the unit
Make them prove it: Enter an import with freight, duty, clearing and inland transport. The receipt should hit stock at true cost, not invoice value.
Offline capture that syncs without duplicating
Make them prove it: Airplane-mode a device, record a receipt and a count, reconnect, and confirm nothing arrived twice.
Transfers that show goods in transit
Make them prove it: Move stock between two locations. It should belong to neither until confirmed received.
Counts and variance as a routine, not an investigation
Make them prove it: Run a count on a subset of items and produce the variance report without closing the business for a day.
Batch or lot tracking where the goods demand it
Make them prove it: Receive two batches of the same item at different costs and prove which one was sold.
Reorder points using real lead times
Make them prove it: Set a reorder point that reflects the customs and road leg, not the supplier's quoted lead time.
Barcode or QR capture at the point of movement
Make them prove it: Receive and issue by scan on a phone, not by typing a code from a printed list.
Stock in someone else's custody, visible
Make them prove it: Show what a named agent, van or contractor is holding right now.
Demand forecasting from your own history
Make them prove it: Worth having once the four above are true. Worthless before, because it will forecast from records that were estimates.
Forecasting built on records that were estimates does not produce a forecast. It produces a confident estimate.
Landed cost: the one that pays for the system
If only one thing changes, make it this. Pricing off a supplier invoice ignores everything between the factory and your shelf, and in an importing business that gap routinely runs into double-digit percentages. The margin you believe you are making is not the margin you are making, and no amount of good selling recovers a floor that was set too low.
What the gap looks like on one consignment
None of those lines are hidden — every one of them is on a document somebody in your business has seen. What is missing is the allocation onto the receipt, so the number reaches the unit rather than the profit and loss. The mechanics are in what landed cost actually means.
The second-book problem
The most common way an inventory implementation fails on this continent is not rejection. It is a branch or site store quietly keeping a parallel record because the system is unusable from where they stand, and everyone agreeing to reconcile monthly.
That reconciliation is not a control. It is a negotiation between two accounts of the same events, held weeks after either could be verified, and it always resolves in favour of whichever number is easier to explain.
-
Make the remote location a location
It holds its own stock position inside the same system, receives transfers, and issues. Not a spreadsheet that gets uploaded.
-
Give it capture that works without a connection
Receipts, issues and counts on a phone, queued locally, synced when signal returns — with each operation carrying its own identifier so nothing arrives twice.
-
Show transfers in transit
Stock that has left one location and not been confirmed at the other belongs to neither. This single behaviour removes most of the monthly argument.
-
Count on a calendar, not on suspicion
A count triggered by a missing item is an accusation. A count on a rhythm is a control, and it is the one that gets cooperation.
-
Give the store keeper the report first
If variance reaches head office before it reaches the person who can explain it, you have built a surveillance system rather than an inventory system, and it will be defeated.
What does not vary, whatever the map says
It is worth stating the other half plainly, because it is where a lot of regional marketing quietly overclaims.
Genuinely different, or just inventory?
| Claim | Genuinely regional | Just good practice |
|---|---|---|
| Landed cost matters more because import legs are long | Yes | No |
| Offline capture is a requirement, not a nice-to-have | Yes | No |
| Custody by non-employees needs first-class handling | Yes | No |
| Reorder points must use real, not quoted, lead times | Partly — configurable by you | Yes |
| Stock should move when the transaction happens | No | Yes |
| Counts should be routine and reconciled | No | Yes |
| Batches should be traceable where goods require it | No | Yes |
| Forecasting needs clean history first | No | Yes |
Built and maintained Configurable by you, not maintained by us Not built
Read the right-hand column as reassurance rather than criticism. Most of what makes inventory work is universal, well understood and available — which means a vendor cannot justify a weak product by pointing at local conditions, and you should not accept it when they try.
The straight answer
What AWRA OpsHub does today
- Multi-location stock with governed transfers and in-transit visibility between warehouses, branches and remote site stores.
- Landed cost from freight, duty, clearing and handling, allocated onto the receipt so the unit carries its true cost.
- Batches and lots, with cost held per batch, so which consignment was sold is a fact rather than an assumption.
- Counts, blind counts and variance reporting against physical stock.
- Reorder points, item classification and stock valuation reporting.
- Barcode and QR labelling with capture by scan on mobile.
- Offline capture with a device register, queued operations carrying their own identifiers, duplicate-safe sync and a conflict view for administrators.
- Demand forecasting from your own transaction history, once that history is worth forecasting from.
What it does not do
- No warehouse execution or automation. No pick-path optimisation, no conveyor or sorter control, no automated storage and retrieval.
- No manufacturing execution. Bills of material and production exist; shop-floor control, machine scheduling and line-side systems do not.
- No customs, clearing or freight forwarding. You record the costs you incur; classification, duty determination and declarations belong to your clearing agent and the authorities.
- No revenue-authority transmission outside Kenya. eTIMS is our only fiscal e-invoicing integration anywhere, so a sale in another country does not reach a tax authority from here.
- No automatic supplier catalogue or EDI integration. Supplier data is maintained by you or imported.
The first two lines are worth a moment if you are large. There is a real point where a distribution business needs warehouse execution and a manufacturer needs shop-floor control, and at that point the correct answer is a specialist system alongside this one, not a longer feature list on ours.
What is not built for your market today can still be built for you
Anything described above as not built is a statement about what ships in the standard product today — not a limit on what AWRA OpsHub can do in your market. Kenya's eTIMS integration and its maintained payroll engine exist because Kenyan clients needed them and commissioned them; neither appeared by itself. The same door is open here. If a revenue authority pipeline, a bank or mobile money feed, a statutory return format or a link to a system you already run is what stands between you and a decision, tell us and we will scope it as a build — written spec, timeline and price — before you commit to anything.
Tax and e-invoicing pipelines
Electronic invoicing against your revenue authority's published interface, with the parts vendors gloss over — retries, a failure queue and a daily report of sales carrying no fiscal reference.
Banks, payments and mobile money
Statement feeds, payment gateways, bulk-payment files and collection accounts wired into the Payments Register so money in and out reconciles without re-keying.
Payroll and statutory returns
Payroll and social security schedules produced in the layout your filing body expects, generated from live payroll records rather than rebuilt each month.
Systems you already run
The accounting package, CRM, online store or custom database you intend to keep — connected through our API so a fact is entered once and appears everywhere it is needed.
How it works: you describe the requirement, we return a written scope, timeline and cost, and once agreed it is built into your environment and maintained as part of the product. No roadmap slide, and no pretending in a demo that something exists when it does not.
Tell us what you need integratedWhere to go next
The continental framework this sits inside is the best ERP software for African businesses, and the connectivity half is in cloud, connectivity and data residency.
By market, the same problems in local detail: inventory and distribution in Lagos, retail and wholesale in Accra, distributors and retail in South Africa and Dakar and Abidjan. The foundational version, market-neutral, is the stock control guide.
Our take
Ignore the feature list and fix three things in order: make every sale and issue move stock at the moment it happens, fold landed cost into the unit so your pricing floor is real, and give remote locations capture that works without a connection so the second book never opens. Those three carry almost all of the return. Everything else — forecasting, ABC, optimisation — is worth buying afterwards and worth nothing before, because it will be computed from records that were guesses.
See inventory built for long legs and weak signal
One stock position across locations, in-transit transfers, landed cost folded into the unit, and offline capture that syncs without duplicating a single receipt.
Explore inventory managementFrequently asked questions
Is inventory management really different in Africa?
Less than marketing implies, and more than a global product assumes. The principles are universal — move stock when the transaction happens, count on a rhythm, reconcile variance, trace batches. What changes is the priority order, driven by four conditions: long expensive import legs, real distance between locations with imperfect connectivity, capture happening away from a desk, and stock held by people you do not employ. Those push landed cost, offline capture and custody to the top of a list where a global product would put forecasting and optimisation.
What is the single highest-return change we can make?
Fold landed cost into the unit. Pricing off a supplier invoice ignores freight, duty, clearing, port charges and the inland leg, which in an importing business routinely amounts to a double-digit percentage of the invoice value. Every one of those numbers is already on a document somebody in your business has seen — what is missing is allocating them onto the receipt so they reach the unit rather than disappearing into overheads. It is the change that most often pays for the system by itself.
Can a branch with unreliable internet use the system?
Yes, and this is the capability that decides whether a remote location stays inside the system or reverts to its own book. Receipts, issues, transfers and counts are captured on a phone or tablet offline and queued locally, syncing when the device reaches a signal. Each queued operation carries its own client-side identifier so the same receipt cannot arrive twice, conflicts are surfaced to an administrator rather than resolved silently, and there is a device register showing which devices are behind, what is queued and what failed.
Do you do demand forecasting?
Yes, from your own transaction history — and we would rather you did not switch it on first. A forecast is only as good as the records underneath it, and a business whose stock levels have been estimates for three years will get a very confident forecast of the wrong thing. Fix capture, landed cost and counts, run for two or three quarters on records that are actually true, then turn forecasting on. It will be worth considerably more and you will trust it, which matters as much.
Does it handle batches and expiry?
Yes. Items can be received and tracked in batches or lots with cost held per batch, so which consignment was sold is a recorded fact rather than an assumption — which is what makes both margin analysis and a recall workable. Where goods carry expiry, that is held on the batch. Whether you need this is a question about your goods rather than your country: pharmaceuticals, food and agricultural inputs generally do, hardware and general merchandise generally do not.
Can we see what a van salesperson or agent is holding?
Yes — treat them as a location or a custody relationship rather than as a customer. Stock issued to a van or an agent remains yours and remains visible, movements are recorded, and reconciliation at the end of a route is against a position the system already knows rather than a manual count against a memory. This is one of the areas where the four conditions genuinely bite, and where a global product designed around fixed warehouses tends to be weakest.
Is this a warehouse management system?
For most distribution and retail businesses, it does what you need: multi-location stock, governed transfers, batches, counts, scanning and reorder points. It is not warehouse execution — there is no pick-path optimisation, no conveyor or sorter control and no automated storage and retrieval — and there is a genuine size at which a business needs those. If you are at that point, the honest recommendation is a specialist warehouse system connected to this one, rather than a longer feature list from us.
