Over the years, I have watched supply-chain teams chase more dashboards, real-time tracking and data feeds, convinced that the next integration would end the firefighting. The industry spent a decade solving visibility and largely succeeded. Most carriers now provide basic track and trace directly from their websites.
Teams reached for the next obvious lever: more alerts. If visibility alone was not enough, surely more notifications would help teams stay ahead of problems. In practice, it made things worse.
Why more alerts is the wrong answer
Flood an operations team with notifications on every shipment and you have not solved anything. You have moved the noise from a spreadsheet to a dashboard. Operators spend their time describing problems rather than resolving them, and the alerts that matter get buried by those that do not.
Not every delay carries the same consequence. A three-to-four-day delay on a fifty-day haul out of Europe barely registers. The same delay on an intra-Asia leg with a total transit time of three to four days doubles the journey. A single threshold applied across an entire network will either drown a team in low-priority alerts or cause it to miss the one that was urgent.
What a better approach looks like
The shift that changes outcomes is exception-based freight management: fewer signals, defined by lane and context, delivered early enough that someone can still act. Teams define what risk means for each corridor and let the system surface only shipments that breach those thresholds.
Nishant Kumar, who leads transportation for Mondelēz International across AMEA, described this during a fireside chat at LogiSYM Asia Pacific 2026. Each market sets its own lane-based risk threshold, grading exceptions as high, medium or low based on what is at stake. Attention goes where it counts instead of being spread thin across shipments that have always been fine.
What this makes possible
Transshipment dwell time. The team tracks the expected dwell window for each lane, typically three to seven days. When a container exceeds that window, it gets a specific alert—for example, that a next-leg vessel has not been assigned—so the team can contact the carrier before the delay compounds.
Detention and demurrage. The team catches containers approaching the free-time limit while there is still room to act, rather than receiving a charge and disputing it later.
Crisis response. During the Persian Gulf disruption, the team worked port by port, identifying containers in transshipment with no confirmed onward vessel, often before carriers had a plan.
Food products cannot simply be rerouted without consequence. Every region has different packaging requirements and shelf-life constraints due to local customs rules. Every day counts. Early visibility gives teams something concrete to take to carriers.
Where this is heading
Exceptions never disappear in logistics. What separates a well-run supply chain is how quickly they are caught and how much warning a team has before they turn expensive. An exception-first approach means fewer hours combing through updates and more time making decisions that change outcomes.
The next layer is cost reconciliation, where AI has a genuine near-term role. Freight costs are rarely what they appear to be on a rate card. Incidental charges accumulate across invoices, lanes and carriers in ways that are difficult to track manually. AI can aggregate this information and surface the gap between what was quoted and what was paid.
At Portcast, we have built that layer above visibility: one that turns information into action and moves teams from tracking shipments to intervening before outcomes are affected. The value was never in more data. It was always in better decisions, made earlier.
About the author
Nidhi Gupta
Founder and CEO · Portcast