Every hardware team has had the same meeting. The design is finished, the review went well, and then the buyer comes back with a number: forty-two weeks on one line. Not an obscure line, either — usually something ordinary, chosen early, never questioned again.
Nobody did anything wrong. The engineer picked a part that met the specification. The buyer quoted what they were asked to quote. The problem is that those two decisions happened in sequence, weeks apart, and the first one was made without any of the information that would have changed it.
Availability is a specification line
The habit worth building is simple: treat availability as a requirement with the same standing as current draw or package size. When you write down that the converter must deliver 2 A at 92 % efficiency, write down that it must be obtainable in your build quantity within your schedule, from more than one source.
That reframing does most of the work. A requirement gets checked. A preference gets forgotten.
Three checks that cost minutes
Look at real stock, not the datasheet. Manufacturer pages tell you a part exists. Distributor stock pages tell you whether it can be bought this quarter. Those are different questions and only one of them affects your schedule.
Ask how old the part is. A component released eighteen months ago and one released eighteen years ago carry opposite risks. The new one may not survive to your production ramp; the old one may already be on a quiet path to end-of-life. Neither is disqualifying — but you should know which one you are holding.
Find the second source before you need it. Not a vaguely similar part: an actual alternate you have checked against your circuit, with the footprint and the control interface confirmed. Write it in the bill of materials as a note. The version of you that has to find it under deadline pressure will not have time to do it properly.
The footprint is where the real cost sits
A single-source part is survivable if the replacement drops into the same footprint. It becomes a redesign if it does not.
So when you have a genuine choice, prefer the package that several manufacturers build to. A common footprint with three suppliers behind it is worth a meaningful amount of performance, because it converts a potential board respin into a purchasing decision.
Where the value actually appears
None of this shows up in the first project review. It shows up eighteen months later, when a part goes on allocation and the response is a purchase order to a second supplier rather than four weeks of rework and a new fabrication run.
That is the whole argument for keeping the engineering and the buying in the same conversation. The information that makes a good component decision is mostly commercial. The person who needs it is the one drawing the schematic.