Tyre supplier ordering, without the phone.

Where the time actually goes

Ask a workshop how long tyre supplier ordering takes and the answer is usually an underestimate, because it happens in fragments rather than in a block.

The fragments are checking availability across suppliers, comparing prices that are not directly comparable, placing the order, recording it somewhere so the job knows it is coming, and chasing when it does not arrive. Each is small and the total is not.

The other cost is the interruption. Ordering usually happens in the middle of something else, which is why it is both slow and error-prone.

One view of availability beats three portals

The single biggest saving is not ordering faster, it is not having to check in several places to know what is possible.

When availability and price across your suppliers appear in one view, the decision takes seconds and it is made on complete information rather than on whichever portal you checked first. It also stops the quiet habit of always ordering from the same supplier because their site is the least annoying.

That habit is expensive. Preferred suppliers earn their position on service and price, and neither can be assessed if you never see the alternatives at the moment of ordering.

Order against the job, not the size

An order placed as a size in a portal is disconnected from the reason it exists, and reconnecting them is manual work someone does later.

Placing the order against the booking means the job knows what is coming and when, the customer can be told honestly, and nobody has to reconcile a delivery note against a diary. When the delivery is late, the affected job is obvious rather than something you discover on the day.

It also makes the margin visible per job rather than per month, which is the level at which pricing decisions actually get made.

Keep the relationships, drop the phone calls

There is a real fear that automating ordering damages supplier relationships that took years to build, and it is worth taking seriously.

In practice the relationship is about credit terms, allocation when stock is tight, and someone answering when there is a problem. None of that depends on placing routine orders by phone, and all of it survives an automated order for a standard line.

Keep the calls for the things that need judgement: an unusual fitment, an urgent job, a delivery that has gone wrong. Those conversations are more valuable when they are not competing with forty routine ones a week.

Watch what arrives, not just what was ordered

The gap between the order and the delivery is where money leaks, and it is rarely checked properly.

• Short deliveries that are noticed on the day but never credited.
• Substituted lines accepted quietly because the job needs doing.
• Price differences between the quoted and invoiced figure on the same line.
• Repeated lateness from one supplier, which never gets quantified.
• Returns raised and then not followed through.

Each is small and none of them are memorable, which is why they persist. A record of ordered against received makes the pattern visible, and the pattern is what you take to the supplier.

Let an agent do the repetitive part

Ordering is a good candidate for automation precisely because it is repetitive, rule-driven and high-volume, and a poor candidate for judgement.

An agent can check availability across suppliers, apply your preference rules, place the routine orders and flag the exceptions that need a person. What it should not do is decide which supplier relationship matters to you, or accept a substitution on a job where the exact line was specified.

The test of whether it is set up correctly: routine ordering stops appearing in your day, while the unusual cases still arrive on someone's desk with enough context to decide.