What standardised delivery measures mean for sellers

Carriers report their own performance in their own terms. On-time rates calculated against their own service definitions, success rates counted their own way, transit times measured from a point they choose.

Every figure is defensible and none of them is comparable. That is the gap a standardised measurement framework across carriers is meant to close.

Why comparability is hard today

Four reasons the numbers do not line up.

Different definitions of on time. Measured against a published service level that each carrier sets, so a carrier promising a longer window has an easier target and can report a better rate.

Different start points. From collection, from arrival at a sorting facility, or from the order being placed. Transit time means different things depending on where the clock starts.

Different treatment of exceptions. Whether a failed attempt caused by the recipient counts against the carrier's success rate is a definitional choice, and carriers choose differently.

Different granularity. National averages hide regional variation, and regional variation is where the differences between carriers actually live — see choosing couriers by destination.

The consequence is that a seller comparing carriers on their published figures is comparing things that are not alike.

What standardisation would change

A common framework applied across participating carriers — as set out in the collaboration framework announced for Malaysian marketplace logistics — would in principle change three things — see the Shopee courier collaboration explained.

Comparison becomes possible. Same definitions, same measurement points, across carriers.

Competition moves beyond price. In a market that has competed largely on rate, a carrier that performs well gains something to compete on. That is better for carriers with genuine operational strength and better for sellers — see the economics of last-mile delivery in Malaysia.

Underperformance becomes visible. Measured consistently, a carrier falling behind in a region cannot be hidden inside a favourable national average.

Two caveats worth holding. It applies to participating carriers on that marketplace's volume, not to the market as a whole. And whether sellers ever see the resulting data is a separate question from whether it exists.

Build your own version now

The useful part: you do not have to wait for anyone. Everything needed to measure your carriers consistently is in data you already generate, provided consignment notes are captured against orders — see capturing the consignment note against the order.

Five measures, defined once and applied to every carrier identically.

First-attempt success rate. Delivered on the first attempt as a share of parcels shipped. The single most useful measure, because failed attempts drive the largest avoidable cost — see failed deliveries and what they cost.

Transit time from your dispatch, not from the carrier's collection scan. Your customer's experience starts when the parcel leaves you.

All-in cost per parcel, including surcharges and weight corrections rather than headline rates — see courier invoices and how to reconcile them.

Loss and damage rate, which is low for everyone and differs enough to matter on higher-value goods.

Exception rate — parcels needing intervention of any kind.

Segment all five by region, at minimum peninsular against East Malaysia. National averages hide exactly the differences you would act on.

Turning measurement into decisions

Three uses, in ascending order of value.

Route by strength. Where one carrier is measurably better in a region or a weight band, send that volume there. This requires no negotiation and produces an immediate improvement — see choosing couriers by destination.

Negotiate on evidence. A rate conversation supported by your actual volume and performance data by region is a different conversation from one based on a total parcel count.

Raise problems specifically. A carrier told that first-attempt success in a particular area has fallen over three months can investigate. A general complaint about service produces an apology and no change.

That third one is the most underused. Carriers generally have better regional data than their customers assume, and a specific, evidenced question gets a specific answer.

The honest caution

Measurement improves decisions and does not by itself improve delivery.

A carrier at capacity during a peak will be late whether or not you are measuring it. A region that is expensive to serve will remain expensive. Standardised measures, whoever produces them, make performance visible rather than better.

What changes outcomes is acting on what becomes visible: routing differently, packing better, capturing addresses more completely, offering collection where failures cluster. Those are within your control and they are where the return is — see what shared access points change.

Common questions

Why can carrier performance figures not be compared?

Because each carrier defines on-time against its own published service level, measures transit from a start point it chooses, decides for itself whether recipient-caused failures count against success, and reports national averages that hide regional variation. Every figure is defensible and none is comparable to another carrier's.

What would standardised measures change?

Comparison across participating carriers becomes possible, competition gains an axis beyond price in a market that has competed largely on rate, and underperformance in a region can no longer hide inside a favourable national average. The caveats are that it applies to participating carriers on that platform's volume, and that sellers seeing the data is a separate question.

Can a seller measure carrier performance themselves?

Yes, from data already generated if consignment notes are recorded against orders. Five measures applied identically to every carrier: first-attempt success rate, transit time from your dispatch rather than the carrier's collection scan, all-in cost per parcel including surcharges, loss and damage rate, and exception rate — each segmented by region.

Does measuring delivery performance improve it?

Not by itself. A carrier at capacity during a peak is late whether or not you are measuring, and an expensive region stays expensive. What improves outcomes is acting on what becomes visible — routing volume to whichever carrier is stronger in a region, packing better, capturing addresses more completely, and offering collection where failures cluster.

Related: choosing couriers by destination · the Shopee courier collaboration explained · failed deliveries and what they cost