When lead times become unpredictable, the natural reaction is to add stock — a bit more here, a larger buffer there, a round-up on the reorder point. It feels prudent. It also quietly locks up working capital, fills warehouse space and, most damaging, hides the problems that caused the volatility in the first place.

The question is not whether to hold safety stock. It is how much, where, and on what basis.

What safety stock is actually for

Safety stock exists to absorb variability — variability in demand, and variability in supply and lead time. It is not there to compensate for a biased forecast, an unreliable supplier or a broken replenishment process. If those are the real issues, more stock only delays the reckoning.

The cost of "just add more"

  • Working capital — every extra week of cover is cash that cannot be used elsewhere.
  • Obsolescence and waste — especially in food, cosmetics and anything with a shelf life.
  • Space and handling — buffer stock still has to be stored, counted and moved.
  • Masked root causes — a large buffer makes a poor forecast or a shaky supplier invisible until it fails badly.

Sizing it properly

A sound safety-stock policy rests on a few decisions:

  • Set a service-level target by segment, not one number for the whole catalogue. A core, high-rotation product does not need the same protection as a long-tail item.
  • Use both demand variability and lead-time variability. In the GCC the second usually dominates — a supplier who ships in 30 days give or take 10 drives far more buffer than day-to-day demand noise.
  • Understand the shape of the maths. Safety stock grows with the square root of lead time and in proportion to demand variability — so shortening or stabilising lead time reduces stock faster than squeezing the forecast.
  • Segment the catalogue. A simple ABC (value) by XYZ (predictability) grid tells you where to spend analytical effort and where a basic rule is enough.

Make it a routine, not a one-off

Lead times shift as carriers, suppliers and routes change. A safety-stock number calculated once and left alone is wrong within a quarter. Recalculate on a fixed cadence, compare achieved service against target, and adjust as conditions normalise or deteriorate. Tie the review to your S&OP cycle so it has an owner and an audience.

Practical starting points

  • Pull the last 12 months of lead-time data by supplier and look at the spread, not just the average.
  • Take your top 20 SKUs by value and check whether their current cover matches a deliberate service target.
  • Identify any SKU where more than half the stock is "just in case" and ask what would have to be true to reduce it.

May Advisory helps teams set safety stock on a clear basis, segment the catalogue, and turn the calculation into a routine the planning team can run themselves.