US Customs Entry Automation for Freight Forwarders and Customs Brokers
Most of the time and risk in a US customs entry sits in preparation. This guide covers how to automate it, from inbox to broker-ready entry data, while the licensed broker keeps control of what is filed.

Ripple automates US customs entry preparation before the licensed broker files. It processes incoming commercial invoices, packing lists and bills of lading, extracts and validates every line, applies importer-specific business rules, keeps each importer’s product catalogue current as HTS codes and tariff measures change, assigns 10-digit HTSUS codes with confidence scoring, assesses the applicable duty stack, requests missing data automatically, and writes structured, broker-ready entry data into CargoWise or your downstream system. Ripple is secure and engineered for production, with access controls, monitoring and a full audit record on every line.
The human-control boundary is deliberate. Under Section 484 of the Tariff Act of 1930 (19 U.S.C. 1484), only the importer of record, owner, purchaser or a licensed customs broker may make entry, and CBP’s Right to Make Entry Directive 3530-002A sets out how that works in practice. Ripple prepares, validates, records and routes. Licensed teams review and submit. Teams get automation and keep full control of what is filed.
Most of the time and risk in US customs operations sits in preparation. Forwarders and brokers can file quickly once the data is right. The cost and error risk build up upstream: reading variable-quality invoices, reconciling product descriptions against importer catalogues, applying the correct tariff measures to each line, spotting missing country-of-origin or valuation data, and routing genuine exceptions to a qualified reviewer. Ripple removes that preparation burden at scale.
This page covers Ripple’s full US customs entry preparation workflow. For 10-digit HTSUS classification, duty exposure modelling, ECCN and Schedule B scope, see the HTS classification capability page.
Where US Entry Preparation Breaks at Scale
High-volume US import operations face a preparation problem that compounds as tariff policy changes and importer portfolios grow. The failure points are consistent across freight forwarders, customs brokers and 3PLs:
Variable document quality
Commercial invoices arrive from dozens of suppliers in different formats, languages and levels of detail. Product descriptions range from precise part numbers to vague generic terms. Unit prices, quantities and country-of-origin fields are frequently incomplete, inconsistent or contradictory across the invoice, packing list and bill of lading.
Importer-specific rules at scale
Each importer of record has its own product catalogue, classification history, preferred HTS positions and internal rules for how goods should be grouped, split or treated. Applying those rules manually across hundreds of shipments per day is the work that consumes experienced customs staff.
A shifting tariff stack
The US International Trade Commission has published 23 revisions to the Harmonized Tariff Schedule so far in 2026, the latest on 10 November. Alongside those revisions, Section 301 measures, Section 232 tariffs, Chapter 99 provisions, AD/CVD orders and other tariff rules change effective dates, rates and product scope. Every change can leave a code or duty position in an importer’s catalogue out of date. Manual workflows absorb those changes slowly and inconsistently, and a stale code reused on the next entry means the wrong duty, rework and penalty exposure.
Missing data that stalls entries
When a country of origin is absent, a product description is too vague to classify confidently, or a valuation input is missing, the entry stalls. In manual operations, that stall often goes undetected until a broker reviewer catches it late in the process.
The result: preparation queues grow, experienced staff spend their time on routine extraction and validation instead of genuine exceptions, and classification errors accumulate under time pressure. Automation addresses each of these failure points before the broker opens the file.
From Inbox to Broker-Ready Entry Data
Ripple runs the full preparation workflow, from document arrival to structured output in the downstream system, without per-customer templates and without rekeying.
Document intake and extraction
Ripple monitors your shipment inbox or document source, identifies relevant shipment documents and extracts structured data from commercial invoices, packing lists and bills of lading. It works across variable formats without a fixed template for each supplier or importer. Where documents conflict, Ripple applies your source precedence rules to decide which field value takes priority.
Importer catalogue matching and completeness validation
Each extracted line is matched against the importer’s existing product catalogue, and Ripple checks that the catalogue entry is still current. A code affected by an HTS revision or tariff change since it was confirmed is flagged before it is reused. Known products are classified using confirmed HTSUS codes and importer-specific rules. New or unrecognised products are flagged for broker review with the extracted description and supporting context. Missing fields, including country of origin, unit value, quantity basis and manufacturer data, are identified at this stage, and Ripple requests the required information automatically before the entry proceeds.
10-digit HTSUS classification with confidence scoring
Correct US tariff treatment requires two distinct steps. The first is assigning the right 10-digit HTSUS code from Chapters 1 through 97 of the Harmonized Tariff Schedule of the United States. The second is determining which Chapter 99 provisions, exclusions, trade-measure overlays and duty rates apply to that code for the specific country of origin and shipment date. A correct base classification does not by itself settle the total duty treatment.
For each line, Ripple assigns a 10-digit HTSUS code with a confidence score, then assesses the applicable Chapter 99 and trade-measure treatment as a separate, sequential step. Lines above your confidence threshold proceed to duty assessment. Lines below it are routed to a licensed broker reviewer with the extracted data, the recommended code, alternative headings considered and the reason for low confidence.
Duty stack assessment
Ripple assesses the duty inputs for each classified line:
- Base HTSUS duty rate
- Applicable Section 301 measures by country of origin and product
- Section 232 tariffs where in scope
- Chapter 99 provisions, inclusions and exclusions
- AD/CVD orders where applicable and in scope
- Merchandise Processing Fee (MPF) and Harbor Maintenance Fee (HMF) inputs for landed-cost readiness
Trade-programme eligibility (USMCA and other agreements) and partner government agency (PGA) requirements are assessed where scoped. Any line with ambiguous origin evidence, missing documentation or eligibility uncertainty is routed to the broker for a decision.
System-ready output
Validated classification and duty data is written into CargoWise or your downstream system. The output includes structured line data, tariff results, exception flags and the full audit record: the line-level data a broker needs to complete the entry (CBP Form 3461) and entry summary (CBP Form 7501) and transmit them through ACE. No rekeying by the broker team is required.
For a detailed look at the classification operating model, see how US logistics operators are automating 10-digit HTS classification.
The Rules Layer Behind Every US Customs Entry
Consistent output at scale depends on deterministic rules working alongside AI. Ripple applies a set of operational rules, configured with each customer, to every entry, producing the same result regardless of which staff member would otherwise have handled the queue.
The rules layer covers seven categories:
- Document precedence. When the same field appears in multiple documents, your source hierarchy determines which value is used. Commercial invoice, packing list and bill of lading conflicts are resolved by rule.
- Field-level fallback logic. Where a preferred source field is absent, Ripple falls back to the next permitted source. Where no fallback exists, the field is flagged as missing and a data request is generated.
- Importer-specific catalogue reuse. Confirmed HTSUS codes and classification decisions from prior entries are stored in the importer’s catalogue. Each new shipment draws on that history, so experienced decisions accumulate and are reused. Catalogue codes are kept current as HTS revisions and tariff changes are published.
- Line matching and duplicate control. Invoice lines are matched against packing list and bill of lading data. Duplicate line detection prevents the same goods from being classified or duty-assessed twice.
- Country-of-origin and importer-of-record validation. Country-of-origin data is validated against your entry rules. Importer-of-record data is checked for completeness before the entry proceeds.
- Grouping, split-line and item-combination rules. Where importer rules require goods to be grouped or split for entry purposes, those rules are applied consistently.
- Exception thresholds and broker-review triggers. Confidence thresholds, missing-data conditions and tariff-measure ambiguity each have defined routing rules. The broker receives only lines that genuinely require human judgement.
This rules layer is what turns HTS lookup into a production entry-preparation workflow. Classification accuracy matters, and so do the consistency, completeness and auditability of every entry that reaches the broker’s desk.
New Importer-of-Record Requirements from 5 January 2027
On 6 November 2026, CBP published an interim final rule, Heightened Requirements for Foreign Importers of Record, which takes effect on 5 January 2027. In summary:
- To qualify as a US importer of record, an entity must be organised and located in the US, and either meet a beneficial ownership or control test or hold significant, directly owned, unencumbered US real property. A US address alone is not enough.
- For the consumption entries the rule covers, foreign importers of record must be validated under CBP’s Customs Trade Partnership Against Terrorism (CTPAT) programme, or use a CTPAT-validated licensed customs broker.
- Entries must declare the importer’s US or foreign status, including on CBP Forms 3461 and 7501.
- Before declaring a client to be a US importer of record, a broker must verify that it qualifies, and keep the verification records for five years from verification. Inaccurate declarations carry penalty exposure under 19 U.S.C. 1592.
- Eligible brokers can request delayed compliance with the verification requirement by 5 January 2027. If CBP approves the request, full compliance is required by 6 May 2027. The other requirements take effect on 5 January 2027.
For brokers, this adds a verification step for every client declared as a US importer of record, on top of the preparation work described above. The rule is interim and open for comment until 5 January 2027, so check the Federal Register notice for the current position before relying on this summary.
Keeping Catalogues Current When HTS Codes and Tariffs Change
Every HTS revision or tariff change can leave codes in an importer’s catalogue out of date. If a stale code is reused, the error carries into the next entry, and it is often found only after filing, as rework or a penalty risk. Manual workflows track these changes slowly and inconsistently.
Ripple keeps importer catalogues up to date automatically, as each change is published, so the next entry starts from the current position:
- Change monitoring. Ripple tracks HTS revisions published by the USITC and changes to tariff measures, recording the effective date, scope and affected HTSUS subheadings.
- Automatic catalogue updates. Affected SKUs and product groups in each importer’s catalogue are identified and updated automatically. Changes that need judgement, such as a subheading split that could move a product to more than one code, are routed to a licensed broker before the code is used again.
- Open invoice prioritisation. Incoming invoices containing affected products are flagged before classification proceeds, so the broker can confirm the correct duty position before the entry is prepared.
- Landed-cost input update. Duty and fee inputs for affected lines are recalculated to reflect the new measure, keeping landed-cost data current without a manual recalculation exercise.
- Audit trail. The prior duty position, the change event, the effective date and the revised position are all recorded. The reason for any classification or duty change is preserved alongside the original decision.
This operational response to tariff changes is covered in depth in why classify once no longer works and keeping HTS classification accurate as US tariffs change. For the broader pressure on high-volume low-value shipments following de minimis rule changes, see the end of de minimis.
What the Broker Receives With an Exception
Every line routed to a licensed broker reviewer arrives with a complete evidence package. The broker does not need to re-extract data or reconstruct the classification rationale from the original documents.
The exception package contains:
- Source documents. The commercial invoice, packing list and bill of lading pages relevant to the flagged line, with the extracted fields highlighted.
- Extracted line data. Product description, quantity, unit value, country of origin, manufacturer and any other fields extracted from the documents.
- Importer catalogue context. Whether the product matched a prior entry, what HTSUS code was previously confirmed, and whether this is a new product with no catalogue history.
- HTSUS recommendation and confidence. The recommended 10-digit code, the confidence score, and the alternative headings that were considered and ruled out.
- Tariff measures assessed. The Section 301, Section 232, Chapter 99, AD/CVD and other measures evaluated for the line, with the applicable rates and effective dates.
- Missing-data request history. Any automated requests sent to the importer or supplier, and the responses received.
- Human decision and audit record. Once the broker makes a decision, the override reason, the reviewer identity and the timestamp are recorded alongside the original recommendation.
This evidence layer gives the broker everything needed to make a fast, accurate decision, and it supports the broker’s and importer’s own recordkeeping for every entry.
CargoWise Integration
Ripple integrates with CargoWise through a connected, system-to-system workflow integration. The integration pulls relevant job data from CargoWise at the start of the preparation workflow and writes validated results back once preparation is complete.
The acceptance standard for a completed entry preparation is:
- The correct CargoWise job is identified and updated
- Structured line data, including product descriptions, quantities and values, is written to the correct fields
- 10-digit HTSUS codes and tariff results are populated against each line
- Exception flags and broker-review items are visible in the broker’s interface
- No rekeying is required by the broker or operations team
CargoWise stays the system of record. Ripple handles the preparation work that currently sits outside it: inbox monitoring, document extraction, catalogue matching, classification and duty assessment. The broker team works in the same CargoWise environment they already use, with preparation done before they open the job.
For teams evaluating US customs entry automation options more broadly, see the best HTS classification software for freight forwarders guide.
Operational Proof
Ripple is in production with US customs operators. US results will be published as customers approve them.
The same production workflow architecture (document extraction, importer catalogue management, tariff-measure assessment and exception routing) runs UK customs declaration preparation for Pentagon Freight Services, handling approximately 140 customs declarations per day.
The following results are from that UK deployment and should not be read as US outcomes:
- Up to 1,400 hours per week of manual customs preparation removed
- Up to 35 FTEs of capacity freed for exceptions and complex work
- 60% lower clearance costs
- 40% lower software costs
Frequently asked questions
Can AI prepare a US customs entry?
Yes, AI can prepare the entry data, but a person with the legal right to make entry must file it. AI can extract invoice, packing list and bill of lading data, assign 10-digit HTSUS codes, assess duty measures and flag missing information. The importer of record or a licensed customs broker then reviews and files through ACE. Ripple automates that preparation stage and routes uncertain lines to a licensed broker.
Who is legally allowed to file a US customs entry?
Under Section 484 of the Tariff Act of 1930 (19 U.S.C. 1484), entry may be made only by the importer of record, owner, purchaser or a licensed customs broker acting for them. From 5 January 2027, CBP’s interim final rule on importers of record adds a US-status test and broker verification duties. Ripple prepares entry data for licensed brokers and does not file.
How do Section 301, Section 232 and Chapter 99 apply to the same line?
The 10-digit HTSUS code sets the base duty rate. Chapter 99 provisions then add, modify or exclude duties for that code, depending on country of origin, product and entry date, and Section 301 and Section 232 measures are applied through those Chapter 99 headings. A correct base classification therefore does not settle the total duty on its own. Ripple assesses the base code and the Chapter 99 treatment as separate, sequential steps for each line.
What data does a broker need to prepare a US entry summary?
At line level, a broker typically needs a product description, 10-digit HTSUS code, country of origin, quantity and unit of measure, entered value, manufacturer identification and the applicable Chapter 99 or trade-programme treatment, plus importer-of-record details. Gaps in any of these stall the entry. Ripple identifies missing fields at extraction and requests them automatically before the entry proceeds.
Does Ripple submit entries to CBP?
No. Ripple prepares, validates and routes entry data. The licensed broker reviews the output, makes any required decisions and files through the Automated Commercial Environment (ACE). Ripple does not file in ACE.
How does Ripple handle Section 301, Section 232 and Chapter 99 measures?
Ripple assesses applicable Section 301 measures, Section 232 tariffs and Chapter 99 provisions for each classified line, using the country of origin and HTSUS code to determine which measures apply. Effective dates, rates and product-specific inclusions or exclusions are applied as part of the duty stack assessment. Lines with ambiguous applicability are routed to the broker with the relevant measure details and evidence.
Can Ripple process invoices with hundreds or thousands of lines?
Yes. Ripple processes commercial invoices at scale, including shipments with hundreds or thousands of line items. Each line is extracted, matched against the importer catalogue, classified and assessed for applicable tariff measures. High-line-count invoices do not require manual splitting or pre-processing.
Does Ripple use importer-specific HTS catalogues?
Yes. Confirmed HTSUS codes and classification decisions are stored in an importer-specific catalogue. Each new shipment draws on that catalogue first, so established decisions are reused consistently. New products are flagged for broker review and added once confirmed. Ripple also keeps each catalogue current, updating codes automatically when HTS revisions or tariff measures change, so stale codes are not reused.
Can Ripple prepare CargoWise-ready data?
Yes. Ripple writes validated line data, HTSUS codes and tariff results into CargoWise through a system-to-system integration. Exception items are visible in the broker’s interface. No rekeying is required.
What happens when a product description or country of origin is missing?
Ripple identifies missing or incomplete fields during extraction and validation. An automated data request is sent to the importer or supplier. The entry does not proceed until the information is received or the broker confirms how to handle the gap. Requests and responses are recorded in the audit trail.
Does Ripple support US exports, including Schedule B and ECCN?
Schedule B classification for the Automated Export System (AES) and Export Control Classification Number (ECCN) assignment are available where scoped as part of the deployment. See the HTS classification capability page for detail on export classification and ECCN screening.
Bring Your Live Invoices
The clearest way to assess US customs entry automation is against your actual work: your supplier invoices, your importer catalogues, your tariff exposure and your current exception rate.
Ripple works with freight forwarders, customs brokers and 3PLs across North America, the UK, Europe and the Middle East. Deployments go live in weeks, integrate securely with CargoWise and existing systems, and are operated and continuously improved by Ripple as part of the engagement.
Bring a sample of your most difficult import packs to a demonstration, and we’ll start with how your operation runs today.
Further reading:
- 10-digit HTS classification, duty exposure, ECCN and Schedule B
- How US logistics operators are automating HTS classification
- Keeping HTS classification accurate as US tariffs change
- Why classify once no longer works
- The end of de minimis: a guide for brokers and forwarders
- Best HTS classification software for freight forwarders
