Six months of disruption has produced a consensus that companies lack supply chain visibility, and the consensus is slightly wrong in a way that matters. Most organisations have plenty of data about their supply chain. What they discovered this year is that their ERP was designed to record transactions that have already happened, and every question that mattered between March and September was about something that had not happened yet. Where is the container now. Can I still promise this to the customer. What breaks if that factory stays shut for another month. Those are not transactional questions, and a system of record answers them only by accident.
Three questions that get called one thing
"Visibility" is used to mean at least three different capabilities, and conflating them is why so many projects this year delivered dashboards nobody used. Where is it. The physical status of goods already ordered: at the supplier, at the port, on the water, at customs, in a bonded warehouse. Standard ERP models this as a binary — the purchase order is open, or the goods receipt exists. Everything in between is a gap of six to ten weeks in which the system knows nothing. What can I promise. Availability to promise across every location, taking into account committed stock, realistic lead times and open commitments. Most implementations calculate this from lead times entered during the original configuration and never revisited, which in March meant the system was confidently promising dates based on a world that had ended. What happens if. Scenario and dependency questions. Which customers, revenues and contracts sit behind a single supplier, a single port, a single item. Almost no ERP answers this without external work, because the dependency information is spread across the item master, the vendor master, open orders and the sales pipeline, and nothing joins them. A programme that does not say which of the three it is buying will deliver the easiest one and call it done.
| Question | Information needed |
|---|---|
| Where is it? | Supplier confirmation and intermediate shipment, port and clearance states. |
| What can we promise? | Committed stock, current lead times and open customer commitments. |
| What happens if? | Dependencies joining items, sources, orders and the revenue they support. |
Qualitative summary of this article's source text, not a measured outcome or performance estimate.
The data model is the constraint
The reason the answers are poor is structural, not analytical, and it is worth being specific because each defect has a cheap remedy. Purchase orders carry an original promised date that is treated as the truth long after it has become fiction, because nobody has a field for the supplier's latest confirmation. In-transit goods have no intermediate state. The vendor master holds one address and no information about where the vendor actually manufactures, let alone who supplies them. The item master usually names one source of supply, so alternates exist only in a buyer's memory. And planning parameters — lead time, safety level, minimum order quantity — were set at go-live and have not been recalculated from actual performance since. Fix those five and you have most of what the visibility projects were trying to buy.
What worked this year, in order of cost
The organisations that coped best did unglamorous things quickly rather than starting a programme. They added a confirmed date alongside the promised date on every open purchase order, updated it whenever the supplier communicated, and reported the slippage between the two. That single field converts an inbox full of supplier emails into a measurable trend, and it is the highest-value change available to most companies. They created a light in-transit record with three or four states — produced, shipped, arrived at port, cleared — updated manually if necessary. Precision is not the point. Knowing that forty per cent of your inbound value is sitting at a state nobody has touched in three weeks is the point. They flagged single-sourced items and valued them by the revenue they support, which turns a procurement curiosity into a board-level list of maybe forty items that deserve a second supplier. They recalculated lead times from actual receipts rather than from the master data, which typically reveals that the system has been planning on numbers that were never accurate even before the pandemic. And they built one exception list: purchase orders whose confirmed arrival now falls after the date the goods are committed to a customer. Not a dashboard. A list, reviewed daily by someone with the authority to act on it.
Visibility that does not change an allocation is a dashboard
The harder half is what happens after the system tells you the truth. Scarcity forces three decisions that most companies had never written down. Who gets the item when there is not enough: the largest customer, the most profitable, the one with a contractual commitment, or the one you want to keep next year. Somebody is making that call weekly at the moment, usually a sales manager, usually without a rule. What may be substituted: approved alternates in the bill of materials or item master, with the quality and customer-approval constraints recorded, so a substitution does not require a three-day investigation. How you re-promise: the operational ability to give a customer a revised date, proactively, from the system rather than from a spreadsheet. This year that capability was worth more than the inventory itself, because customers forgive delay far more readily than they forgive discovering the delay themselves. Write those three down, connect them to the exception list, and the visibility work starts paying for itself.
Practical Guidance for Supply Chain ERP Strategy
- Add a supplier-confirmed date field and report slippage against the original promise. The cheapest meaningful improvement available, and it takes days rather than months.
- Model in-transit as three or four states, even if updated manually. The gap between despatch and receipt is where the entire year's uncertainty lived.
- Flag single-sourced items and rank them by revenue at risk. Dual sourcing everything is unaffordable; dual sourcing the top forty items is not.
- Recalculate lead times and planning parameters from actual receipt history. Go-live assumptions have quietly been driving your plan for years.
- Build one daily exception list, not a dashboard. Orders that will now arrive after they are committed elsewhere, reviewed by someone empowered to act.
- Write the allocation rule and the substitution rules into the system. Scarcity decisions made informally are inconsistent, unauditable and slow.
- Make proactive re-promising an operational routine. Customers tolerate delay; they do not tolerate finding out on the delivery date.
- Map tier-two exposure for critical items only, by asking suppliers. Where do you make it, where do your key inputs come from, what is your alternate. Three questions, forty suppliers, one afternoon each.
The Regional Angle
Four things have played out distinctively here. Start with air freight, because it has broken more landed-cost models in the Gulf than anywhere else. A large share of regional imports — spare parts, pharmaceuticals, electronics, fashion — moved in the cargo holds of passenger aircraft through the Dubai and Doha hubs. When the passenger fleets were grounded in the spring, that capacity vanished, rates multiplied several times over, and space was allocated to whoever had the relationship rather than whoever had the order. The ERP consequence is specific: landed-cost configurations and standard costs built on pre-pandemic freight assumptions are now producing margins that are simply wrong, and the variance is being absorbed quietly at period end. Update the freight elements in your landed-cost rules, and look at every product whose gross margin has moved more than a few points this year before you make a pricing decision based on it. Second, the re-export model carries a failure mode that standard configurations do not anticipate. A great deal of regional trade is entrepôt business into Africa, the CIS, the Levant and South Asia, and when destination markets closed their borders this year, goods were left in bonded and free zone warehouses — neither sold, nor available for domestic sale without duty and licensing consequences, nor returnable. Most systems show that stock as available inventory, which overstates what you can actually sell and understates the ageing problem sitting in the warehouse. Create a stock status for goods that are physically present and commercially blocked, and report it separately to the board. Third, watch the interaction between procurement slippage and contract penalties. Regional groups supplying government and semi-government projects operate under contracts with liquidated damages, milestone-linked payment and, occasionally, price escalation provisions. When a material delivery slips by eight weeks, the exposure is not a late delivery; it is a penalty calculated per day against a contract value. Almost no ERP project module links an open purchase order's confirmed date to the contractual milestone that depends on it, which means the commercial exposure surfaces when the claim arrives rather than when the delay happens. Even a manual link on the twenty largest contracts is worth building. Fourth, note that inventory policy in this region has been pulled in two directions at once. Working capital pressure says hold less; the food and medical security directives issued across the Gulf this year, and the general anxiety about import dependence, say hold considerably more, in some sectors as an expectation from the authorities rather than a commercial choice. Those two instructions cannot both be satisfied by one set of reorder parameters. Segment the item master explicitly: strategic cover items with mandated or policy-driven stock levels, and everything else on commercial rules. Otherwise the planner resolves the contradiction informally, and nobody can explain the inventory number at year end.
The objection worth taking seriously
The strongest objection is that visibility is a consultant's product being sold against a physical problem. Nothing on a screen reopens a factory in Ningbo, restores belly-hold capacity or gets a container off a quay in Jebel Ali. Companies spent the summer buying analytics to watch a shortage they could not fix, and the honest description of 2020 for many businesses is that the constraint was capacity and the correct response was patience and cash. That is true and incomplete. Where capacity was scarce, it was allocated, and it was allocated to the customers who confirmed fastest, ordered earliest and communicated credibly. Companies that could see their exposure by mid-March placed orders while others were still reconciling spreadsheets, and they kept their customers because they re-promised rather than went silent. Visibility did not create supply. It determined who got the supply that existed, which is a smaller claim and a defensible one. The practical implication is proportionality. Instrument the small fraction of items that carry the revenue, fix the five data defects, build one exception list and write down the allocation rule. That is a few months of unglamorous work. A control tower programme costing more than the shortage it monitors is how a genuine lesson from this year turns into next year's abandoned platform.
Common Questions
Do we need a specialist supply chain platform, or can ERP do this?
For most mid-sized groups, the ERP can do it once the master data and the confirmed-date discipline are in place. Specialist platforms are worth it when you have multi-tier manufacturing or genuinely global inbound flows.
How far down the supply chain should we map?
Two tiers, for critical items only. Beyond that the information decays faster than you can maintain it.
Is holding more inventory the answer?
Partly, and selectively. Buffer the items that stop revenue or that have long and unreliable lead times; buffering everything converts a supply problem into a cash problem.
What should we expect over the next twelve months?
Expect resilience budgets to be approved generously through 2021 and quietly trimmed once shortages ease, so secure the durable changes — data, dual sourcing, contracts — while the appetite lasts. Expect ERP vendors to push supplier collaboration portals hard, and expect a wave of funding into visibility platforms. Expect lead times and freight rates to stay volatile well into next year rather than snapping back. And expect the next squeeze to come from shipping and container capacity as demand recovers unevenly, which will catch the companies that have spent this year planning only for demand to stay weak.
Supply Chain ERP Strategy — we fix the five data defects that hide your exposure, build the exception list that drives allocation, and connect procurement slippage to the commitments it threatens.