Why Exporters Lose Days Before Their Cargo Even Moves

Executive Summary

When an export shipment is late, the post-mortem almost always points to the visible part of the journey: a vessel that rolled, congestion at a transshipment hub, a storm in the Pacific. Yet for a large share of shipments, most of the lost time accrues long before the cargo leaves the seller's premises. The order is confirmed, the goods are ready, and still the container sits — waiting on a document, an approval, a signature, a booking confirmation, an inspection slot. This is the invisible pre-shipment lead time, and it is where exporters routinely lose days they never account for.

This article decomposes that pre-departure window stage by stage: order confirmation, documentation preparation, compliance and licence checks, inspection and certification, booking and space allocation, customs export clearance, consolidation and stuffing, and the final gate-in before the cut-off. At each stage we separate the time spent doing actual work (processing time) from the time spent waiting (dwell time), because the distinction matters enormously. The uncomfortable finding, consistent across trade-facilitation research, is that the majority of pre-shipment lead time is not physical work at all. It is coordination overhead — sequential hand-offs, rework loops, approval queues, and time-zone latency between parties who each hold one piece of the puzzle.

We quantify typical lead-time ranges, explain why emerging-market corridors lose more days than developed ones, and show, without overselling, how a unified Trade Operating System compresses this window by parallelizing steps and pre-validating data rather than re-keying it across a dozen disconnected systems.

The Shipment That Was "On Time" — Until It Wasn't

Consider a mid-sized exporter of processed agricultural goods shipping a full container load from an inland location to an overseas buyer. The purchase order is signed on the first of the month. The goods are physically packed and palletised within five working days. By any reasonable internal definition, the order is "ready."

The vessel the exporter is targeting has a port cut-off roughly three weeks out — generous, comfortable, no reason for concern. And yet the container does not gate in until the day before that cut-off. Eighteen working days passed between "goods ready" and "cargo moving," during which the product sat in a warehouse, fully manufactured, generating storage cost and tying up working capital, while paperwork and approvals circulated.

Nothing in that story is exceptional. There was no crisis, no missing raw material, no labour dispute. The delay was entirely administrative and coordinative — the slow accretion of small waits across a chain of dependent steps, each handled by a different party using a different system. This is the pattern that trade-operations professionals recognise instantly and that outsiders to the industry rarely see, because it leaves no dramatic footprint. There is no single villain. There is only the cumulative drag of a process that was never designed as a process.

The purpose of this article is to make that invisible window visible: to name each stage, to estimate how long it takes, to identify what actually causes the wait, and to distinguish the time that represents genuine work from the time that represents pure friction.

Two Different Clocks: When the Lead-Time Count Begins

Before decomposing the timeline, it is worth being precise about what we are measuring, because much of the confusion around export delay comes from measuring the wrong interval.

There are at least three plausible "start" points for an export shipment, and exporters frequently conflate them:

• Order confirmation. The commercial moment — the contract is binding, terms and Incoterms are agreed.
• Goods ready. The operational moment — production is complete and the cargo is physically packable.
• Booking confirmed. The logistics moment — space on a specific vessel or flight is secured.

The interval that this article concerns itself with runs roughly from order confirmation to the moment cargo physically departs the origin gate. Within that interval, the most damaging and least examined sub-window is the gap between goods ready and cargo moving — because by then the exporter has incurred almost all of its production cost and is simply waiting. Capital is locked in finished inventory, and every additional day is pure carrying cost with no offsetting progress.

The reason this distinction matters is that it changes where you look for improvement. If you only measure transit time — port to port — you will optimise the part of the journey you control least and ignore the part you control most. Trade-facilitation bodies have made this point repeatedly: the time a shipment spends in border and pre-border procedures is frequently comparable to, or greater than, the time it spends in actual carriage on shorter corridors. The lever with the most slack is the one nearest the exporter's own desk.

Decomposing the Pre-Shipment Critical Path

The pre-shipment phase is not a single block of time. It is a sequence of distinct stages, each with its own owner, its own inputs, and its own failure modes. The following decomposition reflects the dominant pattern for containerised ocean exports; air and groupage shipments compress some stages but rarely eliminate the underlying dependencies.

Stage 1 — Order Confirmation and Terms Lock

The clock starts when the buyer and seller agree commercial terms: price, quantity, Incoterms, payment instrument (open account, documentary collection, or letter of credit), and required documents. This stage is short in elapsed time but disproportionately important, because it sets every downstream requirement. If the letter of credit specifies a document the exporter cannot easily produce, or names a certificate that requires a multi-day lead time, the delay is effectively scheduled here — it simply manifests later. Errors locked in at terms-agreement propagate through the entire chain and are the most expensive to unwind.

Stage 2 — Documentation Preparation

This is the stage where days quietly disappear. A single containerised export can require a commercial invoice, packing list, certificate of origin, bill of lading instructions, export declaration data, insurance certificate, and — depending on the product and destination — a sheaf of regulatory documents. Each document draws on overlapping data: consignee details, HS classification, quantities, weights, values. In a disconnected workflow, that same data is re-entered into separate templates and systems multiple times, and every re-entry is an opportunity for a discrepancy.

The delay here is rarely the typing. It is the waiting and the rework. The invoice waits for the finance team to confirm the final value. The certificate of origin waits for a chamber of commerce to attest it. The bill of lading instructions wait for the forwarder to confirm container details. And when one document disagrees with another — a weight that does not match, a description that does not align with the HS code — the whole bundle loops back for correction, often after it has already been forwarded to the next party. Under documentary letters of credit, even a trivial discrepancy can trigger a rejection cycle that adds days while documents are re-presented.

Stage 3 — Compliance and Licence Checks

Before goods can be exported, the exporter must confirm that the transaction is permitted: that the product is not restricted or prohibited for the destination, that any required export licence or authorisation has been obtained, that the buyer is not a sanctioned or denied party, and that the correct HS classification has been applied. For controlled goods — dual-use items, certain chemicals, agricultural products, defence-adjacent technology — a licence application can itself take days or weeks and is often on the critical path.

The friction at this stage is twofold. First, the checks are frequently performed late, only once documentation is underway, rather than at order confirmation when they belong. A licence requirement discovered in week two is far more damaging than the same requirement flagged on day one. Second, the checks are manual and siloed: someone consults a list, someone emails a compliance officer, someone waits for a screening result. When compliance is a sequential gate rather than a parallel, pre-validated check, every shipment inherits its full latency.

Stage 4 — Inspection and Certification

Many exports cannot move until a third party has physically inspected the goods or certified an attribute of them. The common varieties include:

• Pre-shipment inspection (PSI), often mandated by the buyer or by the importing country's customs regime, verifying quantity, quality, and price before loading.
• Phytosanitary certificates for plant and plant-derived products, requiring inspection by an agricultural authority.
• Fumigation certificates for wood packaging and certain commodities, requiring a treatment that itself takes time and must cure before loading.
• Quality, lab, or conformity certificates for regulated product categories.

These steps share a defining characteristic: they depend on an external party's schedule, not the exporter's. An inspector must be booked, must travel, must be available. A fumigation treatment must be performed and then must elapse. A laboratory must run an assay and issue a result. Each introduces dwell time that the exporter cannot compress by working harder — only by booking earlier and in parallel, which a disconnected process rarely enables because the trigger to book is buried in a sequential checklist.

Stage 5 — Booking and Space Allocation

In parallel with documentation, the exporter (or its forwarder) must secure space: a vessel sailing, a container allocation, and an equipment release. On constrained corridors and during peak season, space is genuinely scarce, and the booking itself can take days to confirm. Even when space is available, the booking generates a cascade of deadlines — the documentation cut-off, the verified gross mass (VGM) submission deadline, and the physical gate-in cut-off at the terminal or container freight station.

The hidden delay here is the dependency loop between booking and documentation. The bill of lading needs the booking reference; the booking sometimes needs cargo details that depend on final documentation. When these are handled by different people exchanging emails across time zones, each round trip costs most of a working day, and a shipment can stall not because anyone is idle but because each party is waiting on the other.

Stage 6 — Customs Export Clearance

Before cargo can be loaded for export, the export customs declaration must be lodged and cleared, culminating in the authority's release — variously called the let-export order, the customs release, or the equivalent in the local regime. This requires accurate declaration data: classification, valuation, origin, quantities, and any licence references. Where the declaration depends on documents still being finalised upstream, clearance cannot begin, and the stage inherits all of the documentation delay that preceded it.

Customs clearance is where the dwell-versus-processing distinction is starkest. The actual assessment of a clean, low-risk declaration can be effectively instantaneous in a well-run electronic single-window environment. But declarations selected for examination — physical or documentary inspection — wait in a queue for an officer's attention, and that queue is dwell time, not work. Trade-facilitation research consistently finds that the variance in border-clearance time is driven far more by inspection rates and the responsiveness of inter-agency coordination than by the intrinsic difficulty of any single declaration.

Stage 7 — Consolidation, Stuffing, and the Container Freight Station

For less-than-container-load (LCL) cargo, or for full loads routed through an inland container depot (ICD) or container freight station (CFS), the goods must be physically consolidated and stuffed into the container. This stage has its own scheduling: a stuffing slot, dock labour, and equipment. For inland exporters, the cargo must also be trucked to the CFS or ICD, and that movement is itself subject to road conditions, vehicle availability, and the CFS's operating hours and cut-offs.

Each hand-off between the exporter's premises, the transport leg, and the CFS introduces a queue. Cargo arriving at a CFS does not stuff itself the moment it arrives; it joins the day's workload. The gap between physical arrival and actual stuffing is dwell time that rarely appears in any tracking system the exporter can see.

Stage 8 — Gate-In and the Cut-Off Cascade

The final pre-departure milestone is gate-in: the laden container physically entering the terminal before the gate cut-off, with the VGM submitted before its own deadline and the shipping instructions filed before the documentation cut-off. Miss any one of these and the container "rolls" to the next sailing — and on services that sail weekly, a missed cut-off does not cost hours, it costs days.

The cruelty of the cut-off cascade is that it is the point where all the upstream slippage finally becomes visible. A delay incurred during documentation in week one does not announce itself until the container fails to gate in on time in week three. By then the cause is invisible and the only remaining options — expedited trucking, a switch to a later or costlier sailing, air freight on perishables — are expensive remedies for a problem that was created weeks earlier and cheaper to prevent than to cure.

Where the Days Actually Go: Coordination Overhead, Not Physical Work

Having decomposed the timeline, a pattern emerges that is the central argument of this article. If you sum the *processing time* across all eight stages — the actual hours of human and machine work required to produce a document, screen a party, inspect a pallet, lodge a declaration, stuff a container — the total is modest. It is measured in hours, not days. The gap between that processing total and the real-world elapsed lead time is dwell: time the shipment spends waiting in a queue, waiting on a document, waiting on an approval, waiting on someone in another time zone to start their working day.

This dwell is not random. It has identifiable, recurring sources, and naming them is the first step to compressing them.

Sequential Hand-Offs

The single largest source of avoidable delay is the sequential structure of the process. Documentation completes, *then* compliance is checked, *then* the booking is confirmed, *then* customs is lodged. Each hand-off is a queue, and each party only begins its work when the previous party finishes and notifies it. Many of these steps have no genuine dependency on one another and could run concurrently — compliance screening does not require a finalised bill of lading, and an inspection can be booked the moment goods are ready rather than after documentation is complete. The sequence is an artefact of how the work is organised, not a requirement of the work itself.

Waiting on Documents

A great deal of dwell is simply one party waiting for a document held by another. The forwarder waits for the commercial invoice. Customs waits for the licence reference. The bank waits for the certificate of origin. Because the same underlying data — consignee, classification, quantities, values — lives in separate systems and is re-keyed at each stage, documents are produced late and inconsistently, and the waiting compounds.

Rework Loops

Discrepancies are the silent multiplier of pre-shipment delay. A weight that does not reconcile, a description inconsistent with the HS code, an address that does not match the letter of credit. Each discrepancy sends a document — sometimes an entire bundle — back through the chain for correction, often after it has already advanced two stages. A single rework loop can erase the time savings of an otherwise efficient process, and under documentary credits the cost of a discrepancy is not only time but the risk of non-payment.

Approval Queues and Single Points of Contact

Many stages funnel through a single approver or a single person who "owns" the export desk. When that person is travelling, on leave, or simply busy, the entire pipeline for every shipment waits. Concentrated knowledge and authority create fragile throughput: the process moves at the speed of its busiest bottleneck, and that bottleneck is frequently a human approval rather than a system constraint.

Time-Zone Latency

For cross-border trade, the parties are by definition in different places, often many hours apart. An email sent at the end of one party's working day is not read until the start of another's, and a single clarification can consume a full calendar day even though it required two minutes of actual work. Across a chain of several such exchanges, time-zone latency alone can add several working days to a shipment that involved almost no real work — only the friction of asynchronous, manual communication.

Why the Same Shipment Takes Longer on Some Corridors

The pre-shipment window is not uniform across the world. The same product, packed identically, can clear the pre-departure phase in a fraction of the time on one trade corridor versus another. The variation is driven less by distance or product than by the maturity of the trade-facilitation environment at origin.

On corridors served by mature electronic single windows, well-staffed customs administrations, coordinated border agencies, and digitised certification, the processing steps are fast and the dwell between them is short. Documents are lodged electronically once and reused; risk-based inspection means low-risk consignments are released without examination; and inter-agency coordination happens in the background rather than through serial paper hand-offs.

On corridors where these facilitation measures are partial or absent, the same eight stages persist but each accrues more dwell. Paper documents are physically presented and stamped. Certificates require in-person attendance. Customs examination rates are higher and queues longer. The number of distinct agencies and signatures multiplies. The result is a pre-shipment lead time that can be several times longer for an identical shipment, with the difference concentrated almost entirely in dwell and coordination, not in any greater amount of real work. This is precisely why trade-facilitation reform — implementing the measures catalogued under the WTO Trade Facilitation Agreement and measured by the OECD's Trade Facilitation Indicators — yields disproportionate reductions in trade time and cost: it attacks dwell, which is the larger share.

For the individual exporter, the practical implication is sobering and empowering at once. Much of the corridor-level dwell is structural and beyond any single firm's control. But the coordination overhead *within the exporter's own process* — the sequential hand-offs, the re-keyed data, the rework loops, the single points of contact — is firmly within reach, and it is frequently the larger of the two contributors on the goods-ready-to-gate-in segment.

From Sequential to Parallel: What Compression Actually Looks Like

If most pre-shipment delay is dwell caused by sequential coordination rather than processing work, then the path to compression is not "work faster." It is "stop waiting." Concretely, that means three structural changes to how the pre-shipment process is organised.

First, pre-validation instead of late discovery. Compliance checks, document requirements, classification, and certification needs should be determined at order confirmation, when the cost of accommodating them is lowest, rather than discovered mid-process when they sit on the critical path. A licence requirement known on day one is a scheduling input; the same requirement discovered in week two is a crisis.

Second, parallelisation of independent steps. Inspection booking, space booking, certificate procurement, and document preparation do not genuinely depend on one another and should run concurrently from the moment goods readiness is forecast. The sequential default exists because disconnected systems offer no shared view that would let parties act in parallel with confidence.

Third, a single source of shipment data. When consignee details, classification, quantities, weights, and values are entered once and flow into every document and declaration automatically, the re-keying disappears, discrepancies fall toward zero, and the rework loops that silently consume days are largely eliminated. This is the difference between a chain of disconnected tools — email, PDFs, spreadsheets, and a dozen portals — and a unified system of record.

This is the thesis of a Trade Operating System: not a faster version of the existing fragmented workflow, but a structural reorganisation of it. By holding a single validated record of the shipment, screening and validating data the moment it is known, and allowing independent stages to proceed in parallel rather than in series, a unified platform attacks dwell directly. It does not make customs assess faster or make an inspector arrive sooner in isolation; it ensures those external steps are triggered as early as possible, fed with correct data the first time, and never blocked by an internal hand-off that should never have been sequential.

The compression that results is not marginal. When dwell is the majority of lead time and dwell is largely coordination overhead, removing the overhead removes the majority of the delay. The cargo that used to sit for eighteen working days between "ready" and "moving" can, with pre-validated and parallelised coordination, gate in days earlier — not because anyone worked harder, but because the shipment stopped waiting.

Key Statistics

The figures below are drawn from established trade-facilitation and logistics research. Trade time and cost vary widely by corridor, product, and year; the following should be read as directional reference points rather than guarantees for any specific lane.

• The WTO Trade Facilitation Agreement (TFA), in force since February 2017, was projected by the WTO to reduce average trade costs by roughly 14% or more for members implementing the full set of measures, with the largest gains for developing economies — most of which target border and documentary friction rather than physical transport. *(WTO, World Trade Report on the TFA.)*
• The OECD Trade Facilitation Indicators estimate that comprehensive implementation of trade-facilitation measures can reduce trade costs by 12.5% to 17.5%, with measures targeting formalities, automation, and inter-agency cooperation among the highest-impact. *(OECD Trade Facilitation Indicators.)*
• The World Bank's "Trading Across Borders" indicator (Doing Business series) measured border and documentary compliance time separately from transport, frequently recording documentary compliance alone at tens of hours to several days per shipment in lower-performing economies and only a few hours in top performers — a gap concentrated in paperwork, not carriage. *(World Bank, Doing Business / Trading Across Borders.)*
• The World Bank Logistics Performance Index (LPI) consistently shows that the largest performance gaps between top- and bottom-quartile countries appear in customs and "timeliness" dimensions rather than in physical infrastructure, underscoring coordination as the binding constraint. *(World Bank, Logistics Performance Index.)*
• UNCTAD has reported that trade-documentation and procedural requirements can involve dozens of distinct documents and signatures for a single international transaction in less-digitised environments, each an opportunity for delay. *(UNCTAD, trade-facilitation analyses; illustrative range.)*
• UNESCAP digital trade-facilitation surveys link higher adoption of paperless and cross-border paperless trade to materially shorter time-to-export, with the Asia-Pacific region showing wide dispersion in implementation. *(UN Global Survey on Digital and Sustainable Trade Facilitation, UNESCAP.)*
• Container dwell time at ports in efficient gateways is often under 3–4 days, while congested or less-automated ports can record dwell of a week or more — and pre-vessel dwell at origin (CFS/ICD and terminal) adds to the pre-shipment window before any sailing. *(Port-dwell-time studies; illustrative estimate.)*
• The WCO identifies risk management and coordinated border management as primary levers for reducing release times, with its Time Release Study (TRS) methodology repeatedly showing that examination and inter-agency referral, not core declaration processing, drive most clearance-time variance. *(World Customs Organization, Time Release Study guidance.)*
• The Digital Container Shipping Association (DCSA) estimates that industry-wide adoption of the electronic bill of lading (eBL) could save the container-shipping ecosystem on the order of billions of US dollars annually and remove days of documentary turnaround, against single-digit-percent eBL adoption historically. *(DCSA, electronic bill of lading initiative; illustrative.)*
• McKinsey and other analysts have estimated that digitisation of trade documentation and processes can reduce processing times and costs by double-digit percentages, with much of the benefit arising from eliminating manual re-keying and reconciliation. *(McKinsey, trade-digitisation analyses; illustrative estimate.)*
• Across multiple WCO Time Release Studies, a recurring finding is that the majority of elapsed clearance time is dwell — time between process steps — rather than the active processing performed by customs at each step. *(World Customs Organization, Time Release Study findings; directional.)*

Industry Analysis

The Anatomy of Pre-Shipment Lead Time

Pre-shipment lead time is best understood as a layered structure rather than a single number. At the base is irreducible processing time — the genuine work of producing accurate documents, performing a physical inspection, treating a wood pallet, assessing a declaration. This layer is small and not the primary target for improvement; an inspection takes as long as it takes, and a fumigation treatment must cure.

On top of that sits dwell time — the queues, waits, and hand-offs between processing steps. This layer is large, and it is where the leverage lives. The defining insight of decades of trade-facilitation measurement, from the WCO's Time Release Studies to the World Bank's Trading Across Borders methodology, is that dwell dominates elapsed time. A shipment is rarely late because the work was slow; it is late because it spent most of the interval waiting to be worked on.

Sequential Versus Parallel Processing

The conventional export process is overwhelmingly sequential. Each party completes its task and passes a baton to the next, and the total lead time is the *sum* of every stage's elapsed duration plus the latency of every hand-off. In a parallel model, independent stages overlap, and total lead time approaches the duration of the *longest single critical-path stage* rather than the sum of all stages. The arithmetic difference is decisive: a process with eight sequential stages averaging a day and a half each runs to roughly twelve days; the same stages, with the independent ones overlapped, can collapse the critical path to a fraction of that. The constraint on parallelisation is rarely the work itself — it is the absence of a shared, trusted view of the shipment that would let parties act before their predecessor formally "finishes."

Coordination Overhead as the Dominant Cost

The economic framing matters for executives allocating improvement budget. Investment aimed at making any single processing step faster yields diminishing returns, because processing is already the small part. Investment aimed at coordination — eliminating re-keying, pre-validating data, enabling parallel action, removing single points of contact — attacks the large part. This is why the highest-ranked OECD Trade Facilitation Indicators cluster around automation, formalities simplification, and inter-agency cooperation rather than around physical capacity. The bottleneck is informational and organisational, not physical.

Corridor Variation: Developed Versus Emerging Markets

The same structural truth produces very different outcomes across corridors. In developed-market gateways with mature single windows, risk-based release, and digitised certification, dwell is compressed by design, and pre-shipment lead time is short and predictable. In emerging-market corridors, the identical eight stages accrue more dwell at every junction — more documents, more in-person attendance, higher examination rates, more agencies, more paper. The processing work is broadly similar; the coordination overhead is multiplied. For exporters operating across both kinds of corridors, this means lead-time planning cannot be a single global standard — and it means the return on internal process digitisation is highest precisely where the surrounding environment is least facilitated, because the exporter's own coordination overhead is then the controllable share of a much larger total.

Process Diagram (Text Description)

The two timelines below illustrate the same shipment under a conventional sequential workflow and under a parallelised, pre-validated workflow. Blocks marked `[WAIT]` are dwell; blocks marked `[WORK]` are active processing. Durations are illustrative working-day estimates for a containerised ocean export from an inland origin.

SEQUENTIAL PRE-SHIPMENT CRITICAL PATH (conventional, ~12-18 working days)

Day:  0    1    2    3    4    5    6    7    8    9   10   11   12   13   14
      |----|----|----|----|----|----|----|----|----|----|----|----|----|----|
Order [WORK]
 conf  v
Docs       [WORK][WAIT--- finance/CoO attestation -------][WORK rework loop]
prep        re-key data        waiting on chamber of commerce   discrepancy fix
Compl                                              [WAIT queue][WORK screen]
checks                                              started late, sequential
Inspct                                                        [WAIT slot][WORK]
/cert                                                          inspector booking
Bookng                                         [WAIT space confirm][WORK]
space                                           email round-trips, time zones
Cust                                                            [WAIT][WORK release]
clear                                                            exam queue
CFS /                                                                 [WAIT][WORK]
stuff                                                                  trucking + slot
Gate                                                                       [WORK gate-in]
-in                                                                         barely before cut-off
                                                                            |
                                                              CUT-OFF ------>+  CARGO MOVES (day ~14-18)

Observation: most blocks are [WAIT]. Active [WORK] totals a handful of days;
the rest is dwell from sequential hand-offs, re-keying, rework, and time-zone latency.

PARALLELISED, PRE-VALIDATED CRITICAL PATH (Trade OS model, ~4-6 working days)

Day:  0    1    2    3    4    5    6
      |----|----|----|----|----|----|
Order [WORK]  single validated record created; requirements known on day 0
 conf  v
                  +-- Docs generated from one data source --[WORK, no re-key]
                  |
                  +-- Compliance pre-validated at order conf --[WORK, parallel]
   PARALLEL  -----+
   (overlapped)   +-- Inspection booked day 0, performed -----[WORK, parallel]
                  |
                  +-- Space booked day 0, confirmed ----------[WORK, parallel]
                  |
                  +-- Export declaration pre-filled ----------[WORK, parallel]
                                                      |
Cust release (clean, low-risk, data correct first time) [WORK short]
CFS / stuff (slot pre-booked)                              [WORK]
Gate-in well before cut-off                                   [WORK]
                                                              |
                                                CUT-OFF ----->+  CARGO MOVES (day ~4-6)

Observation: independent stages overlap instead of queuing; data entered once;
dwell collapses because no stage waits on a predecessor's paper hand-off.

The contrast is not that any individual task is performed faster. The commercial invoice still takes the same minutes to generate; the inspection still takes the same hours. What changes is that the tasks no longer wait on one another, the data is entered once and validated early, and the rework loops that consumed days are designed out. The critical path shortens because dwell — not work — has been removed.

Pre-Shipment Stage Reference Table

Durations are illustrative working-day ranges for a containerised ocean export from an inland origin; they vary by corridor, product, regime, and season. The "Dwell vs. processing" column indicates where the time predominantly accrues.