Amazon DSP scorecard software starts with the record.The number tells you something moved. The record tells you why.

The scorecard is route allocation

It is graded weekly, and what makes it matter is not the grade but what the grade is attached to. Stations allocate routes on performance, so a poor week is not a poor week — it is a smaller next quarter. Enough of them and it becomes a conversation about the contract.

Which is why the scorecard gets read with more anxiety than understanding. By the time the weekly number arrives, the week that produced it is gone, and the only thing left to act on is a memory of what happened.

What we do, and what we do not

Worth being blunt, because this is where software in this category tends to be vague.

WaveGo does not import your scorecard. It does not read Amazon's metrics, does not receive per-driver scores, and does not raise alerts when a number moves. Those live in your partner portal, and anything claiming otherwise is worth asking hard questions about.

What WaveGo holds is the record underneath. Driver assignments by van and by day. The rota as it was actually worked, not as it was first published. Daily vehicle checks with the defect attached to the vehicle. Damage and incidents against the van that carried them.

That is the raw material of every scorecard diagnosis. Without it, a dropped metric is a rumour.

“Which routes, on which days, with whom”

That is the only question worth asking of a scorecard, and most depots cannot answer it a fortnight later.

The aggregate hides the shape of the problem. Five drivers at a comfortable standard and one having a genuinely bad month average out to something unremarkable — and the unremarkable average gets managed with a briefing that helps nobody, while the person who needed a conversation never has one.

When the assignment history is a lookup, the same week becomes three specific questions instead of one vague verdict. That is the difference between managing a number and managing an operation.

What is visible before the week closes

Not the metrics — we have said we do not have them. What is visible during the week is the operational side, and in practice that is where the week goes wrong long before a score records it.

Routes covered against routes offered. Drivers approaching hours you would rather not push. Vans booked into the garage and due back. Confirmations from drivers who have seen their shifts, and the silence of the ones who have not.

A rota that has been published is a plan; a rota that has been confirmed is a forecast. The gap between the two is where a missed route comes from, and it is visible the evening before rather than the Monday after.

From a number to a conversation

Feedback that improves performance and feedback that produces leavers differ in three ways, and all three need the record.

Timing. A conversation about last week is a conversation about something nobody can change. Specificity. “Your metrics are down” is a verdict; “these three stops on Tuesday” invites the explanation that might turn out to be a van problem or a route problem rather than a driver problem. Proportion. If the only individual attention a driver receives is corrective, the rational response is to keep their head down.

Which is also why this is a retention subject. Most scorecard problems that survive more than a month are retention problems nobody has named yet — drivers in their third week do some of the right things some of the time, and a depot with high turnover runs permanently with more of them on the rota.

About the metric definitions

You will not find a table of metrics on this page, and the reason is not coyness. Amazon does not publish them. They sit in the partner portal, behind a login, and they change.

Our guide to the Amazon DSP scorecard goes through what the published third-party sources say each metric measures, attributes every definition to the source it came from with a date, and shows where those sources contradict each other — including two current ones that define the same three-letter metric differently. It states no thresholds, because no public source states them.

Your own portal is the reference. Anyone selling you a definitive table is selling you someone else's guess.

Questions operators ask

Does WaveGo import my Amazon scorecard?

No. The scorecard lives in your Amazon portal and we do not read it. What WaveGo holds is the operational record the scorecard is produced from — who was assigned where, which van, which shift, what the daily check said. When a number moves, that record is what tells you why.

So what do I get that I do not have now?

The ability to answer 'which routes, on which days, with whom' in seconds rather than an afternoon. Most depots cannot answer it at all a fortnight later, which is why scorecard conversations end in a team-wide briefing instead of a specific fix.

Where do the metric definitions come from?

Amazon does not publish them — they sit in the partner portal. Our guide to the scorecard attributes each definition to the third-party source it came from, with dates, and shows where those sources contradict each other. Your own portal is the reference.

Can I track performance per driver rather than per DSP?

The assignment record is per driver, per van, per day, so you can look at any driver's week. What we do not do is attach Amazon's own per-driver metrics to that record, because we do not receive them.

Does it alert me before the week closes?

Not on scorecard metrics, because we do not have them. What is visible during the week is the operational side: routes covered against routes offered, drivers approaching hours you would rather not push, vans booked in and due back. That is usually where the week goes wrong.

What does it cost?

Five seats are free, permanently, then £10 per seat per month. A seat is any active user, whatever their role.