The aftermarket pays the bills for most equipment makers. It is also where capital goes to sit: parts that move once a year but have to be there the day a customer's line is down.
Source: McKinsey, Turn slow-moving inventory into fast profits (2019).
Service and spare parts are where equipment makers make their margin. They are also where a large chunk of working capital quietly goes to sleep. McKinsey's work on the sector found that somewhere between 10 and 40 percent of an OEM's aftermarket inventory is slow-moving, yet it cannot simply be cleared, because a customer with a stopped line expects the part today. And when a slow part does finally turn obsolete, recovery is brutal: scrapping it typically returns 2 to 5 percent of its cost. Everything else was spent.
Three forces push the stock up. Machines live for decades, so you support parts long after the unit stopped selling. Demand for any single spare is intermittent and lumpy, a few pulls a year, impossible to read with a simple average. And a stockout on a critical component is not a lost sale; it is a customer whose production is down and who remembers it at renewal. Faced with that, the safe answer always feels like 'hold more,' and over a long catalogue that reflex compounds into dead capital.
On the build side, finished machines and long-lead components make the problem worse. You commit to castings, electronics and sub-assemblies months ahead, against an order book that shifts. Configure-to-order variety multiplies it: each option is its own forecast, each one thin. Plan it roughly and you end up with the wrong components staged for the wrong configurations while the line waits on the one part you under-ordered.
The temptation is to chase near-perfect availability on everything. That is how the bins overflow. The better question is which parts genuinely need high protection because a stockout stops a customer, and which can run leaner because a short wait is acceptable. Set that per part, on its real criticality and its real demand pattern, and you protect uptime where it counts without funding a decade of coverage on parts nobody will pull.
The makers who keep the aftermarket profitable rather than bloated forecast intermittent spare-parts demand with a method built for it, size each part's buffer on its real variability and lead time, and flag the pre-obsolescent stock early, while it can still be redeployed or sold, instead of writing it off for scrap later. Long-lead components and finished units get planned on the real order horizon, not a padded guess. That is the deterministic, auditable approach DemandIntel brings to equipment and machinery manufacturers, on your own data, beside the ERP you already run.
Machines are supported for decades, demand for any one part is intermittent, and a stockout stops a customer's line. That combination pushes 10 to 40 percent of aftermarket stock into the slow-moving bucket that still has to be held.
With a method built for intermittent demand rather than a moving average, which rounds sporadic pulls toward zero. Each slow part is then sized deliberately on its real pattern and lead time instead of guessed or over-held.
Catch pre-obsolescent stock early, while it can still be redeployed to another site or sold, rather than at the scrap stage where recovery is only 2 to 5 percent of cost.
Yes, by setting service level per part on its real criticality: high protection where a stockout stops a customer, leaner where a short wait is acceptable. That holds uptime where it matters without funding coverage on parts nobody pulls.
In the aftermarket, the expensive mistake is not the stockout you avoided. It is the decade of coverage you funded to avoid it on a part nobody pulled.
Picture your next S&OP meeting: one number on screen that nobody argues with. See it free, on your own data, in 45 minutes; and if it does not hold up, we part as friends.