Spare capacity is production that could be brought online quickly and sustained. The conventional definition requires availability within around thirty days and sustainability for a period of months, which excludes both a brief surge and a field that would take years to develop.
It matters because it is what determines how a supply disruption prices. Lose a million barrels a day of production from a market with five million barrels a day of spare capacity and the physical problem is solvable: somebody turns a valve. Lose the same million from a market with almost no spare capacity and there is no valve, so the adjustment has to come from price, until enough demand is destroyed or enough inventory released to close the gap.
Spare capacity is concentrated, which is the uncomfortable part. In practice most of the world's usable spare capacity sits with a small number of Middle Eastern producers, principally Saudi Arabia and to a lesser extent the UAE. That means the market's shock absorber is also a strategic asset held by a few sovereign owners, and the willingness to deploy it is a policy question rather than an economic one.
The relationship to inventory is worth being precise about, because the two are substitutes for the same job. Spare capacity is unproduced oil that can be produced; inventory is produced oil that can be released. Both cushion a disruption. A market with low inventory and high spare capacity is not obviously tighter than the reverse, and reading either in isolation as 'tightness' loses that trade-off.