SmartLink
Distribution and supply

Distribution ERP in Pakistan: cost, timelines and stock accuracy

Distribution ERP in Pakistan earns its keep in fractions of a percent, because the margin on a case of goods leaves very little room for a stock record that disagrees with the shelf. SmartLink Services works from Karachi with wholesalers and distributors on warehouse operations, landed cost, contract pricing and van sales across Sindh and Punjab. What follows is the part most vendor pages skip: market rates in PKR, how long a rollout runs across several warehouses and a van fleet, and what sales tax invoicing requires of the system once the goods leave the door. Everything priced below is a published market range rather than a SmartLink quotation.

Focus
Stock accuracy
Costing
Landed
Field
Van and route
Pricing
Per customer
Overview

Margin in distribution is thin enough to be lost in the detail

Distribution runs on volume and thin margin, which makes small inaccuracies expensive. A stock figure that is two percent wrong shows up as failed picks, emergency purchases and credit notes long before anyone calls it a data problem.

Most of the work is therefore about accuracy at the point of movement: scanning rather than typing, cycle counting rather than an annual shutdown, and put away rules that keep the warehouse map honest.

The other half is knowing what the margin really is. Landed cost has to include freight, duty and handling, or the price list is built on a number that flatters every imported line. Once landed cost is right, customer and product profitability finally mean something, including for the accounts that look big but earn little.

The work

Where wholesale distribution systems break

Drawn from the problems that come up repeatedly in this sector rather than from a generic capability list.

Where it usually hurts

  • Stock records drifting from the shelf, discovered at annual count
  • Landed cost missing freight, duty and handling, so margin is overstated
  • Customer specific pricing maintained by hand and inconsistently applied
  • Van sales and route deliveries reconciled on paper
  • Backorders and part shipments handled outside the system
  • No visibility of which customers and lines actually earn money

What the work covers

  • Warehouse operations with scanning: receipt, put away, pick, pack and despatch
  • Cycle counting and stock accuracy reporting rather than an annual shutdown
  • Landed cost including freight, duty, clearing and handling
  • Customer and price list management with contract pricing
  • Van sales, route delivery and proof of delivery capture
  • Customer and product profitability reporting on true cost

Typically involves

ERP WMS Barcode and scanning Mobile apps EDI and APIs
Discuss your systems
01

Cycle counting works when it is small, frequent and unpredictable

Annual counts are a compliance ritual with management pretensions. They close the warehouse, exhaust the team, and produce one large adjustment that says nothing about when or where the drift began. Cycle counting replaces that with a rota: a short list of locations counted every working day, chosen by value and movement, with the fastest turning lines coming round most often. The count is small enough to be done properly and frequent enough that a variance is days old when it is found rather than months old.

Two design choices decide whether it works at all. The counter should not see the expected quantity, because a number on a screen is a powerful suggestion to a tired person at the end of a shift. Blind counts, with a recount on any variance outside tolerance, cost a little more time and produce data worth acting on. Second, a variance has to trigger an investigation rather than only an adjustment. An adjustment posted with no cause teaches a warehouse that the numbers are approximate.

Causes are usually few and repetitive. A location shared by two similar products. A case pack quantity entered wrongly on the supplier record. A pick shortage resolved on the floor without any system transaction. A customer return put away before it was ever received. Each is fixable once named, and naming requires a reason field somebody is expected to complete. Accuracy improves in steps rather than smoothly, and every step follows one of those causes finally being closed off.

  • Daily rota weighted by value and movement, not by aisle order
  • Blind counts, with a recount on any variance outside tolerance
  • A cause recorded before the adjustment is posted
  • Accuracy trended by location and product group over time
  • Repeat offending locations escalated rather than simply recounted
02

Landed cost and apportionment

Everyone agrees landed cost should include freight, duty, clearing and handling. Disagreement starts at apportionment. Spread freight by value and a container of heavy, cheap goods is subsidised by the light, expensive ones. Spread it by weight and the reverse happens. Spread by volume and you need cube data the supplier may not provide reliably. There is no universally correct basis, only one that suits your particular mix, and that decision belongs to finance and commercial together rather than to whoever happens to configure the system.

Timing is the second problem and the more practical one. Goods arrive, get received, and are frequently sold before the clearing agent's invoice appears. If cost is finalised only when that invoice lands, early sales carry a wrong cost and the margin reporting for the period is fiction. The usual answer is a provisional landed cost applied at receipt from an agreed estimate, then a variance posted when actuals arrive. What matters is that the variance is visible and reviewed, not absorbed into a line nobody opens.

Exchange movement deserves separate treatment on imported lines. Purchase at one rate, pay at another, and the difference is a financing outcome rather than a product cost, yet many systems bury it in the item cost and quietly distort the margin on everything affected. Keeping rate variance out of item cost keeps product margin comparable from month to month. It also means that when margin does move on a line, the cause is commercial and worth investigating instead of being currency noise.

  • Apportionment basis agreed by finance and commercial before configuration
  • Provisional landed cost at receipt, with a variance when actuals arrive
  • Clearing, duty and handling captured against the shipment, not the invoice
  • Exchange differences kept out of item cost
  • Landed cost variance reviewed per shipment rather than netted at period end
03

Price hierarchy, contract pricing and the discount that outlived its reason

Distribution pricing is layered: a base list, a customer group, a customer specific contract, a promotional price, a volume break, and now and then a one off agreed by a sales manager on the phone. The system needs an explicit precedence order across those layers, and it needs to record which layer produced the price on any given order line. Without that, a credit query cannot be settled without a phone call, and a pricing review becomes an archaeology exercise across several spreadsheets and two people's memories.

Every price agreement should carry an expiry date, and expiry ought to be the default rather than the exception. Contract prices agreed for a season that never formally ended, discounts granted to win a customer who is now the largest and least profitable account, and volume breaks calibrated for a cost base that has since moved all persist because nothing forces a review. A dated agreement forces one. It also gives the sales team a natural moment to renegotiate that does not feel like an attack on the relationship.

Rebates and retrospective claims need modelling rather than remembering. Where a customer earns a rebate on volume across a period, the liability accrues as they buy, and reporting margin without it overstates every sale made in the meantime. Accruing against the agreement as orders are invoiced keeps reported margin honest and removes the year end surprise in which a large claim lands and the sales team is asked to explain a number they were never shown in the first place.

  • Explicit precedence across list, group, contract, promotion and volume price
  • The applied price layer recorded on the order line for later enquiry
  • Expiry dates mandatory on contract prices and special discounts
  • Rebate liability accrued as sales are invoiced rather than settled at year end
  • Approval limits on manual overrides, reported by salesperson
01

What a distribution system costs in Pakistan

Market bands map onto distribution more cleanly than onto most sectors. A single warehouse with purchase, stock and sales runs PKR 800,000 to 1,500,000 and goes live in two to three months. Five to seven modules with a mobile application sit at PKR 1,500,000 to 3,000,000 across three to four months, which covers most distributors with two or three locations. Once several warehouses, van sales, customer integration and real landed costing are in scope, published pricing starts at PKR 3,000,000 and runs four to six months or longer. Most distributors land in the middle band and get pulled upward by their customers rather than by their own ambitions.

Two add ons decide the shape of that number. Multi location capability is quoted in the market at PKR 200,000 to 500,000, and a mobile application at PKR 400,000 to 800,000, which in distribution is the van, the picker and the proof of delivery. Subscription platforms are commonly around PKR 2,500 per user per month, and warehouse operators counted as users can double the figure finance first calculated.

Scanning coverage is the largest variable, and it is a hardware question before it is a software one. A warehouse where every location is labelled and every case carries a barcode configures quickly. A warehouse where the pick face is identified by which supervisor is on shift needs a labelling exercise first, and that exercise is chargeable, physical and slower than anyone expects. Customer integration is next: a large retailer imposing its own file format sets both the specification and the deadline. Landed cost is the third. Agreeing how freight, duty, clearing and handling are apportioned across a mixed container is an argument worth having in week two rather than month six. All of these are market figures. A real figure follows discovery, once warehouse count, scanning coverage and the integration list are known.

  • Warehouses, vans and pick faces counted before the module list is priced
  • Mobile application quoted in the market at PKR 400,000 to 800,000
  • Location labelling and barcoding scoped as physical work, not configuration
  • Each customer integration costed to the customer's format and deadline
  • Landed cost apportionment rules agreed before costing is configured
02

How long a distribution rollout takes

Reported timelines give six to twelve weeks for a focused single warehouse go live and three to six months for a distributor taking more modules and a second site. Van sales and customer integration usually push the figure to the top of that range, and occasionally beyond it, for reasons that have nothing to do with configuration speed.

Stock is the first gate. There is no honest go live without a count that both finance and the warehouse accept, and a warehouse that has not counted properly in two years needs the count planned as an event: a closed day, agreed cut off rules, and a reconciliation somebody signs. We would rather move a go live date than start on a stock figure nobody believes, because every variance afterwards gets blamed on the new system.

Customer deadlines are the second. When a large retailer requires an electronic order and invoice exchange, the conformance test happens on their calendar and their test slot, and missing it can mean waiting weeks for the next one. Route rollout is the third and it is genuinely incremental: one route runs on the new system for a full week including its collection day, the reconciliation is proved at the end of the round, and only then does the second van follow. Doing all the vans in one week produces a nightly argument nobody wins.

  • Focused single warehouse go live reported at six to twelve weeks
  • A full stock count planned as an event and signed before cutover
  • Customer conformance testing scheduled on the customer's calendar
  • Routes converted one at a time, each proved over a full week
  • Seasonal peaks treated as closed windows for change
03

Sales tax invoicing and the records a distributor keeps

Distribution sits closer to the tax machinery than most sectors realise, because almost every movement out of the warehouse generates a document with a tax consequence. Section 3(9A) of the Sales Tax Act requires Tier-1 retailers and other notified persons to integrate with the FBR computerised system for real time reporting, and distributors raising sales tax invoices meet the same requirement where they are notified. The engineering does not change: the invoice is posted, FBR returns an invoice reference number and a QR code, and both print on the document that travels with the goods.

Returns and credits are where distribution differs from a shop. A van comes back with two cases refused at the door, a customer claims a short delivery a week later, and both events have to produce a correcting document that ties to the original invoice rather than a manual adjustment somebody makes in the ledger. Build that badly and the sales tax position drifts from the stock position, quietly, until a reconciliation at year end takes three weeks. Non compliance can mean disallowance of a substantial share of input tax adjustment, currently sixty percent, which makes this a cash issue rather than an administrative one.

Import documentation belongs in the same conversation. The values your clearing agent files, the duty paid and the freight invoiced all feed landed cost, and if that flow is manual then margin is being reported on an estimate. Holding the import documents against the receipt, rather than in a folder beside it, means the cost of a case reflects what the case actually cost, and it means a query about an old shipment is a lookup rather than an archaeology exercise.

The boundary is the same one we state everywhere. SmartLink implements and integrates. Whether your business is notified, which supplies are taxable, how a credit note should be treated and what an unusual transaction requires are questions for your own tax adviser, and we configure to their written instruction.

  • FBR integration where the distributor is notified, tested in the sandbox first
  • Invoice reference number and QR code printed on the despatch document
  • Returns and credits raised against the original invoice rather than the ledger
  • Import documents held against the receipt so landed cost reflects reality
  • Treatment of credits and unusual supplies confirmed by your tax adviser
How we deliver

Delivering in wholesale distribution

Margin is thin enough that a small stock inaccuracy reaches the customer as a failed pick long before anyone calls it a data problem. Accuracy work therefore comes before reporting work.

  1. 01

    Discover

    Time is spent in goods-in, on the pick face and in the van yard, watching where a scan is skipped. Landed cost workings, the price file and last year's count variances are pulled apart alongside that.

  2. 02

    Blueprint

    Apportionment basis for freight, duty and clearing is settled by finance and commercial together, not by whoever configures the system. Price precedence, expiry rules and the van as a stock location are written down.

  3. 03

    Build

    Scanning is deployed where it buys the most accuracy first, usually goods receipt and the fastest moving pick lines. EDI maps for each large customer are built as monitored interfaces with their own state and alerting.

  4. 04

    Test

    Pickers and drivers test on live routes with the device offline, because that is the condition it will meet. Each EDI trading partner runs its own cycle and signs off before a single message goes live.

  5. 05

    Go live

    Opening position is set by a full count over a quiet weekend, with open purchase orders, backorders and van stock loaded and reconciled. Deliveries go out on Monday morning on the new system, on time.

  6. 06

    Run

    Cycle counting starts in week one, blind, on a rota nobody can predict. We stay while landed cost variances and rebate accruals go through a first period end, then hand the routine to your team.

Working together

Where the system stops and the warehouse starts

Accuracy is a behaviour before it is a report. Scanning at the point of movement, counting blind, recording a return the moment it comes off the vehicle: these are floor level habits, and a system can make them fast and obvious but cannot perform them. Distributors who install the software and leave the habits alone get the same variances they had before, now with better graphs attached. The technology removes the excuses and it does not remove the work.

Costing carries a similar honesty. Landed cost, cost to serve and rebate accrual all rest on assumptions somebody in the business has to own and revisit as freight rates, duty and customer behaviour move. We will build the mechanism, document the basis and make the assumptions editable by the people who understand them. What we will not do is pretend an apportionment rule is a fact, because a costing model treated as gospel becomes wrong quietly and stays wrong for years.

What a good distribution system genuinely delivers is a shorter distance between a problem and the person who can fix it. A picking error found the same week. A contract price flagged before it expires rather than after. A customer whose margin has been drifting for two quarters appearing on a list rather than in a year end discussion. Those are modest sounding outcomes, and in a business running on thin margin and high volume they compound quickly.

Credentials

Compliance for wholesale distribution clients

Larger customers send a supplier assurance form before they will trade electronically, and these are the answers that go on it.

Client words

What wholesale distribution clients say

Comments from people running wholesale distribution systems day to day.

  • We had been through one failed implementation already, so we were sceptical of the whole category. The difference here was the migration work. Two full rehearsal loads before the real one, with a reconciliation pack we could check ourselves. Nobody had ever handed us evidence like that and asked us to sign it.
    Head of IT Wholesale distribution business
  • Our first concern with FBR integration was simple: what happens to the tills when the line drops. They built the queueing and retry before anything else and demonstrated it by pulling the connection in front of us. Trading carried on, and the invoices went up when the link came back.
    Operations Manager Retail chain, Pakistan
  • Every vendor we spoke to said they could handle style, colour and size. This team asked to see our order book first, then told us which of the shortlisted platforms would need thousands of item codes to do it. That one piece of advice probably saved us a year.
    General Manager Textile exporter
Questions

Questions about wholesale distribution systems

Market rates run PKR 800,000 to 1,500,000 for a single warehouse covering purchase, stock and sales, and PKR 1,500,000 to 3,000,000 for five to seven modules with a mobile application, which suits most distributors. Several warehouses with van sales and customer integration start at PKR 3,000,000. Those are market ranges rather than our price, and our figure follows discovery once warehouse count and scanning coverage are known.

Reported timelines put a focused single warehouse go live at six to twelve weeks and a wider rollout at three to six months. Van sales and customer integration push it to the top of that range. The stock count is the real gate, because there is no honest go live on a stock figure the warehouse and finance do not both accept.

Distributors raising sales tax invoices meet the same requirement as retailers where they are notified under section 3(9A). The system posts the invoice, stores the invoice reference number and QR code returned by FBR, and prints both on the despatch document. Whether your business is notified is a question for your tax adviser rather than for us.

Market pricing for a mobile application runs PKR 400,000 to 800,000, and in distribution it covers the van, the picker and proof of delivery rather than being an optional extra. Rugged devices for warehouse scanning are a separate hardware decision that we specify but do not supply, and they belong in the budget from the very start.

You need them before scanning goes live, not before the project starts. Labelling a warehouse is physical work with its own timetable, and it is usually slower than the software configuration running alongside it. We scope it as a separate work stream so it never becomes the reason a go live date has to move.

Corrections are possible and they are messy, which is the argument for agreeing the apportionment rules early. Where freight, duty and clearing arrive after the sale, the system can revalue and post the difference, but the reports for that period change afterwards. Better to agree how a mixed container is apportioned in week two than to correct it in month six.

Working with wholesale distribution systems?

Tell us what you run today and where it breaks. The first conversation is a consultation, not a pitch.