Ship Agency and Port Call Software: Closing the Gap Between the Proforma and the Final Account

Why the argument is always about the difference, never the amount

An agent issues a proforma disbursement account for a bulk carrier calling to load. It covers pilotage in and out, two tugs, mooring gangs, tonnage and berth dues, garbage, the agency fee and a contingency. The principal transfers funds. The vessel arrives, the berth is occupied for eleven hours, a third tug is ordered by the pilot because of wind, the port applies an extended stay charge on a scale nobody outside the port authority has ever fully understood, and a launch boat is used twice for crew documents. The final account arrives forty per cent above the proforma. The principal's port cost controller queries it. The agent explains it in an email. Two months later the same conversation happens again on a different call at the same port.

Neither party is being unreasonable. The agent genuinely incurred the costs. The principal genuinely cannot approve a variance they cannot verify. The problem is that the proforma was built from a template someone copied from the last similar call, the tariff behind it is a PDF the port authority updated in January, and the final account is a stack of supplier invoices in three currencies with no structural link back to the lines that were estimated. There is nothing to compare, so the discussion becomes a negotiation about trust.

Marcura DA-Desk sits on the principal's side of this and does a genuinely useful job auditing disbursement accounts at scale. ShipNet and Softship provide broad agency and liner systems with real depth. What none of them removes is the agent's own operational reality: a specific set of ports with specific tariffs, a specific mix of principals with different approval thresholds and cost code requirements, funds held on behalf of others that must be accounted for exactly, and a port call process where twelve suppliers and three authorities have to be coordinated within a window measured in hours.

Problem 1: the tariff is the product and it lives in a PDF

Port charges are formulas. Tonnage dues on gross or net tonnage with a scale. Pilotage by length overall and draught, sometimes with a night or weekend factor. Towage by bollard pull and duration with a minimum. Berth dues per metre per period. Waste reception on a fixed and variable basis. Each port publishes its own tariff, revises it on its own schedule, and expresses it in its own language and structure. An experienced agent knows the local tariff intimately, and that knowledge is why they are worth their fee. It is also entirely undocumented in most agencies.

What a custom build does: model tariffs as versioned calculation rules with an effective date, so a proforma for a call in three weeks uses the tariff that will actually be in force, and a call being reconciled uses the one that was in force at the time. Each proforma line records the rule version and the inputs used. The immediate benefit is that a junior agent produces a proforma of the same quality as the veteran. The larger benefit is that variance analysis becomes possible, because you can compare the invoiced amount against a recomputed expectation rather than against a number typed into a template.

Problem 2: the proforma and the final account are not the same object

In most agencies the proforma is a document and the final account is a different document, assembled later from supplier invoices. Nothing links line to line. So a variance can only be explained narratively, and narrative explanations are exactly what principals dispute.

What a custom build does: make the disbursement a single object with a lifecycle. Every line starts as an estimate, becomes a commitment when the service is ordered, becomes an actual when the supplier invoice is matched, and carries a variance with a reason code. A principal reviewing the final account sees the estimate, the actual, the difference and the reason on each line, with the supporting invoice attached. Third tug ordered by pilot, wind above the port limit, becomes a line rather than a paragraph. The disputes that survive that treatment are the genuine ones and there are far fewer of them, which is worth more to an agency than any efficiency gain: unrecovered variance is the single largest silent cost in ship agency.

Problem 3: the money is not yours and the accounting has to prove it

Funds in advance sit in the agent's account and belong to the principal. Every call has its own position: funds received, disbursed, remaining, and any balance to be returned or carried. Multiple currencies are normal, exchange rates move between receipt and settlement, and the difference belongs to someone specific. Agencies frequently run this in a general accounting package that was not designed for client money, which makes reporting the position per principal per call an exercise in reconstruction.

What a custom build does: hold funds as a ledger per principal per call, with receipts, disbursements, exchange differences and returns each recorded against the call. The agency can see at any moment its aggregate position, which calls are underfunded before the vessel arrives rather than after, and which balances are owed back. Underfunding discovered at the quay is a genuine operational failure, because a supplier who is not paid does not attend, and the vessel is delayed for a reason that has nothing to do with shipping.

Problem 4: the port call itself is a coordination problem with a hard deadline

An appointment arrives, sometimes days ahead and sometimes hours. From that moment the agent must lodge arrival notifications and statutory declarations, book pilotage and towage, arrange mooring gangs, confirm the berth, handle crew changes and cash to master, manage stores deliveries, arrange bunkers if instructed, clear immigration and customs formalities, and keep the master and the principal informed. The whole sequence usually runs from a checklist in the agent's head and an email inbox.

Statutory filing has moved substantially towards electronic single window reporting, with the European maritime single window environment consolidating reporting formalities for calls at member state ports, and many other jurisdictions running their own systems. That is a genuine improvement and it also means the agent now has another system to feed, with structured data that must be correct.

What a custom build does: generate the task set from the call characteristics, so a call with a crew change and a bunker delivery creates the right tasks and a routine call does not, each with a responsible person, a deadline derived from the port's notice requirements, and an escalation. Vessel and voyage data is entered once and feeds the filings, the supplier orders and the disbursement. The most valuable output is the one nobody asks for in a requirements workshop: a live status the principal can see without telephoning, which quietly removes a large share of the agency's inbound calls.

Problem 5: every principal wants a different shape of the same information

What a custom build does: separate the underlying disbursement from its presentation, so the same call can produce the agent's internal view, each principal's cost code mapping, an owner and charterer split, and a formatted account without anyone retyping it. Approval thresholds and named approvers are configuration per principal, and the approval itself is captured in the system rather than in an email chain that nobody can find in six months. For an agency chasing a new principal, being able to demonstrate this on day one is a genuine commercial advantage, because the principal's cost controller is usually the person who feels the pain most acutely.

What this costs and how long it takes

Across the 2,000-plus projects Digital Heroes has delivered, here is the honest shape for port call and disbursement platforms. A first release covering appointment through to a tariff driven proforma, supplier invoice capture with matching, and final account reconciliation with line level variance runs $70,000 to $160,000 and ships in 12 to 18 weeks. Adding funds in advance ledgers, principal approval workflows, multi currency settlement with exchange difference handling, statutory filing integration and a principal portal takes the total to $200,000 to $500,000 across 6 to 12 months.

What drives the number up in agency work specifically: the number of ports, because each tariff is its own modelling exercise and complex ports take days rather than hours. Statutory filing integration, since each single window implementation has its own interface and validation rules. Accounting integration, which is unavoidable because client money must reconcile exactly. The number of principals with genuinely bespoke requirements. And the least visible driver, which is invoice capture quality, because supplier invoices arrive as PDFs in dozens of layouts and matching them to commitments is where the operational time actually goes.

Build versus buy, and the honest small agency answer

Do not build if you are a single port agency handling a couple of hundred calls a year with a handful of principals. A disciplined proforma template, a good accounting package and one experienced agent will beat any system you could afford, and we would tell you to spend the money on staff instead. This is a genuinely small business and software will not change that.

Build the tariff and disbursement layer only, if your problem is variance rather than coordination. Many agencies find that modelling tariffs properly and making the disbursement a single object with line level variance resolves most of the pain, at the lower end of the first release band, without touching the operational side at all.

Build properly when you operate across several ports with different tariff regimes, when you handle enough principals that re-keying into their formats is a real cost, or when client money is large enough that your funds position needs to be knowable rather than reconstructable. Shipping lines with an internal port cost function have a related but distinct case: their build is about challenging disbursements consistently across many agents, which requires recomputing expectations from tariffs rather than accepting what arrives. Either way the trigger is the same, which is that the difference between what should have been charged and what was charged is currently invisible.

How to choose a developer for port call software

Ask them to model a disbursement on a whiteboard. A team that has done this will draw a single object whose lines move through estimate, commitment and actual with variance and reason, and will ask immediately whether funds are client money. A team that draws quotes and invoices has built an accounting tool and does not understand that the proforma and the final account are the same thing at different times.

Ask how they would model a specific port tariff. Give them a real one with a scale and a night factor and see whether they treat it as a versioned calculation with an effective date. If they propose a rate table with a single amount per service, the system will be wrong within a year of the first tariff revision and every historical reconciliation will silently change.

Ask how supplier invoices get matched. The honest answer involves document extraction to propose a match against commitments with human confirmation, and a clear exception queue. Anything claiming full automation has not seen a launch boat invoice from a small port.

Ask who owns the code and the tariff library, and settle it in writing before kickoff. You should own the repository, the cloud accounts and the right to move to another firm. The tariff models are your accumulated local knowledge and are the most valuable thing in the system. At Digital Heroes the client owns the code and the data from the first commit, and a developer wanting to keep your tariff library as their platform content is taking the one asset that makes your agency worth appointing.