# Landed cost exception alerts: the handful of warnings an operations team will actually act on | LandedSight

Source: https://landedsight.com/blog/2026-10-24-landed-cost-exception-alerts-ops/

Cross-border operations guide · October 24, 2026


Most cost dashboards fail by alerting on everything. Operations teams learn to ignore them within a month. This guide picks the few landed cost exceptions worth alerting on, sets thresholds that reflect money rather than noise, and shows how to route each alert to an owner with a next action.

## Alert on money, not on movement

Set thresholds in currency per unit or margin points, not in percentages alone. A twenty percent rise in a tiny surcharge is noise. A two percent rise in ocean freight on a high volume line is a pricing problem. Tie each alert to the volume it affects, so the team sees the estimated monthly impact in the alert itself and can rank it against everything else competing for their attention that week.

Suppress repeat alerts intelligently. If freight is elevated for a season, one standing alert with weekly updates serves better than a daily alarm that trains people to mute the channel. Alert fatigue is a design failure, not a staff attitude problem, and the fix is editorial: fewer messages, each one earning its interruption.

## The five alerts worth building first

Duty change alerts compare the rate applied on new entries with the rate in the product cost, catching classification changes, expired preferences, and measures that took effect while stock was on the water. Freight band alerts compare actual cost per unit or per kilo with the costing assumption for that route. Purchase price variance alerts compare supplier invoices with purchase orders before payment, when a conversation is still easy.

Exam and storage charge alerts flag abnormal port or warehouse charges on a shipment, while they can still be queried. Return rate alerts watch cross border channels for a spike that changes true margin. Together these five cover the places landed cost actually leaks: the state, the carrier, the supplier, the port, and the customer. Other reports can be interesting. These five pay.

## Every alert needs an owner and a verb

An alert sent to a group inbox with no named owner is a wish. Route duty alerts to whoever owns classifications, freight alerts to logistics, price variance to purchasing, and return spikes to the channel owner, with the expected action stated in the alert: confirm, correct the cost file, renegotiate, or reprice. Close the loop by recording the action taken, so the next review can tell whether alerts are changing outcomes or just documenting them.

Review alert outcomes monthly. An alert that never changes a decision should be retuned or retired. An alert that repeatedly finds the same root cause should become a process fix, not a permanent alarm. The goal is a shrinking set of surprises, because the system keeps converting exceptions into standards.

## Build trust with a clean first month

Launch alerts in shadow mode for a few weeks: generate them, review them with the owners, and tune thresholds before they go live to wider channels. Early false alarms are remembered long after they are fixed, and a team that decides a system cries wolf will ignore the one alert that mattered. A short, honest tuning period is the cheapest credibility an operations tool can buy.

Start where the data is cleanest, usually duty and supplier price, and add freight and returns when their data is reliable enough to defend. An alert is a claim that something is wrong. The team will check the first few by hand. Make sure those checks succeed, and the system will still be listened to when it finds the expensive problem in month six.

Keep reading: [Landed cost for subscription and replenishment products: when the same box ships every month](https://landedsight.com/blog/2026-10-24-landed-cost-subscription-replenishment), [Reading a freight invoice for landed cost leaks: the lines that quietly inflate every shipment](https://landedsight.com/blog/2026-10-24-freight-invoice-landed-cost-leaks), and [Landed cost variance: comparing what you estimated against what the shipment really cost](https://landedsight.com/blog/2026-10-22-landed-cost-variance-estimated-vs-actual) cover the nearby cases in this series.

## Frequently asked questions

## How many landed cost alerts should a team have?

Start with about five that reflect real money: duty rate changes, freight cost outside its band, supplier price variance, abnormal exam or storage charges, and cross border return spikes. Route the rest to a weekly report.

## How should alert thresholds be set?

In money per unit or margin impact, weighted by volume, not in percentages alone. The alert itself should show the estimated monthly impact so the team can rank it.

## What makes an alert actionable?

A named owner, the cause, the products affected, and the expected next action, with the outcome recorded. Alerts sent to a group with no owner are rarely acted on.

## Why launch alerts in shadow mode first?

To tune thresholds against real data before interrupting people. Early false alarms destroy credibility, and a system the team trusts will still be heard when it finds a costly problem.