FMCG distribution · Solution
Soluvide builds a delivery-confirmation and reconciliation system for UAE and GCC FMCG and beverage distributors running field van-sales teams. It captures an un-skippable confirmation at every handover, matches dispatched stock and cash collected against confirmed deliveries, and flags mismatches to a supervisor the same day — so leakage and short-deliveries surface in hours, not weeks later at month-end reconciliation.
The problem
What the day actually looks like — described as an operator would, not as a brochure.
A van loads at the depot before dawn. The load sheet says what went on board — cases, crates, mixed SKUs — and from that moment the stock, and by mid-morning a growing pile of cash, are in the hands of one driver moving stop after stop across a full route. Deliveries get made, short-deliveries get promised, returns get taken back, and cash gets collected. The record of all of it is a sheet filled in by hand, against memory, somewhere between stops.
At the end of the route the van comes back and someone reconciles that sheet against what the ERP expected to happen. But the sheet is the driver's account of the day, not an independent one. If a case went missing, if a customer paid for a full case and was billed for a partial one, if cash came up short — the paperwork still balances, because the paperwork is written by the same person the numbers depend on.
The gaps don't show up that evening. They surface weeks later, at month-end reconciliation, as a variance nobody can now explain — the delivery was three weeks ago, the customer remembers it differently, and the driver has run many more routes since. By the time the number is visible, the trail is cold and the money is gone.
Multiply one uncertain van by a fleet, across branches, every working day, and the leakage isn't a rounding error — it is a standing cost of doing business that leadership has quietly priced in, because nobody can see where it happens. The problem was never that people are dishonest. It is that the last mile has no independent record, so honesty and error look identical on paper.
Why the obvious fixes fail
None of them are stupid. They fail because the work itself stays manual — they just move who does it.
Your ERP knows exactly what left the warehouse and what got invoiced. It has no eyes on the stretch in between — the handover at the customer's door. The most valuable, least controlled part of the day is the one part your system of record never sees, so its numbers are only ever as good as the sheet someone types back in.
Checking the books weeks after the fact tells you a variance exists; it can't tell you which delivery, which stop, or which day produced it. By then the evidence is memory and the cash is spent. After-the-fact reconciliation is an autopsy — useful for the record, useless for stopping the loss while it's still happening.
Putting a checker on the van, or more supervisors over the routes, adds cost to every route without adding a record to any of them. Trust-based controls fail the same way — a step a tired driver can skip at the last stop of a long day will get skipped. You scale the wage bill and the routes still go dark.
What we build
The honey step is where a person stays in control. Everything else is the typing and checking the system does for you.
Engineering decision. The confirmation step was built to be hard to skip — a delivery is not complete in the system until the OTP or signature exists, because a control only works if it can't be bypassed. Mismatches surface the same day without stopping the route, and the system attributes, it never accuses.
The system starts from what your ERP already knows: the stock and cash-on-delivery load assigned to each van and route for the day. That dispatch record is the baseline every later step is measured against — no re-keying, no second inventory to maintain.
At each stop the delivery is confirmed at the point of handover — a per-delivery OTP sent to the customer, or a signed confirmation captured on the driver's handset. It's built to be hard to bypass: the delivery isn't complete in the system until the confirmation exists, so the record is created by the transaction, not reconstructed from memory afterward.
A reconciliation layer continuously matches three things that used to only meet at month-end: stock dispatched, deliveries confirmed, and cash collected. Where they agree, the route is clean and nobody is bothered. Where they don't, the mismatch is isolated to a single stop and a single moment.
Mismatches are pushed to a supervisor the same day, not held for month-end — a short delivery, an unconfirmed drop, a cash figure that doesn't tie out. By design the system surfaces exceptions rather than blocking the route: the field team keeps moving, and the flag lands with the person who can act on it while the day is still fresh.
Every handover is written to an audit log, attributable to a person, a stop, and a time — one durable record of who confirmed what, when. Reconciled, settled figures sync back to your ERP, so your system of record finally reflects what actually happened in the field, not just what was expected to.
We confirm exactly which of your systems connect during scoping. The custody-control layer commonly touches:
Engineering decisions
What the fixed price covers, and the things we will not build. Both are written into the scope before we start.
In scope
Not in scope
What you get
Most custody-control builds go live in about three to six weeks. What moves it inside that range is the number of routes and branches, how confirmation is captured at the door, and which ERP it reconciles against. No hourly billing, no surprises.
A short call, then we document how cash and stock move through your field team today — including the exceptions and the workarounds people built around it. You get a written scope, acceptance criteria and a fixed price before any build begins.
You get: Scope + fixed quote
WhatsApp Cloud API, SMS / OTP gateway, Odoo and the rest of your stack — no rip-and-replace. Outputs are structured and checked against your own data, not trusted blindly.
You get: Working system on your stack
The human step is tested with the people who will own it before anything goes out. Logging and alerts so problems surface early, documentation your team can read, and you own the code and the accounts.
You get: Approval gate, monitoring, full ownership
Tell us how cash and stock move through your field team today. We reply within one business day, and the price is fixed after a free scoping conversation.
Questions operations managers ask
If yours isn't here, send it on WhatsApp and you will get a straight reply.
Prefer a form? Contact pageAt each stop the delivery is confirmed at the point of handover — either a one-time code (OTP) sent to the customer and entered on the spot, or a signed confirmation captured on the driver's handset. The delivery is not marked complete in the system until that confirmation exists, which is what makes it hard to skip or backfill later.
Related solutions
Services behind this build
Each service page lists exactly what is included and the questions owners ask before they start.