How to Reduce Excess and Obsolete Inventory Fast

To reduce excess and obsolete inventory fast, don't start with a fire sale. Start with a list. Pull on-hand value against trailing 12-month consumption for every SKU, sort the excess and obsolete by dollar value, and triage each item by what you can actually recover, fastest cash first. Then fix the root cause feeding it so it doesn't grow back. That sequence, find it, triage it, attack the cause, set a tripwire, moves the number faster than any liquidation event.

I ran planning at a $250M furniture manufacturer. We had eight figures in inventory, and somewhere between 12% and 18% of it was excess or obsolete, sitting in the dark, accruing carrying cost. We cut it hard. The playbook below is what worked.

First, define excess vs. obsolete precisely

Vague definitions let excess hide. "It feels like a lot" is not something you can act on. Pin the categories down in writing and apply them in code against your data.

• Excess inventory — on-hand that exceeds expected demand over a defined horizon. The clean rule: anything beyond 6–12 months of forward demand at the current run rate. You'll consume it eventually, but it's tying up cash now.
• Obsolete inventory (dead stock) — no demand in the last 12 months and none expected. Discontinued SKUs, superseded revisions, the custom run a customer cancelled. This isn't "slow," it's done.
• Slow-moving — the warning band in between. Catch parts here before they cross into obsolete.

These definitions aren't just operational. Under U.S. GAAP, FASB ASC 330 (ASU 2015-11, 2015) requires inventory to be carried at the lower of cost and net realizable value, with the write-down recognized as a loss in the period the value drops, including drops from obsolescence. The number you're hiding from is already a liability on the balance sheet.

Step 1 — Find it: build the E&O report

You can't manage what you haven't measured at the SKU level. Compute months of supply for every part: on-hand quantity divided by average monthly demand over the trailing 12 months. Then bucket the results.

Now sort the excess and obsolete list by dollar value, not unit count. This is the Pareto move. A handful of high-value SKUs almost always make up most of the trapped cash.

At our plant, the top 40 E&O items were over half the total dollars. Fix those 40 and you've moved the number more than chasing 800 low-value lines. If you've never run an ABC-XYZ inventory analysis, do it now. It tells you which SKUs deserve tight control and which can run on autopilot.

The scale of the problem is industry-wide, not just yours. McKinsey found that in aerospace and defense, as much as 60 to 80 percent of the value of parts on hand is not needed (2025), with industry inventories ballooning by more than $70 billion since 2016. Most mid-market manufacturers are carrying 10–20% E&O without a clean number for it.

Step 2 — Triage by recoverable value

Not all dead stock is equal. Sort the E&O list into recovery paths, fastest cash first. For each item, the discipline is the same: compute recoverable value minus the carrying cost of continuing to hold it.

• Re-sell at full or near-full price — there's a customer or channel you simply hadn't matched. Fastest, best margin. Check open quotes, other regions, and the sales pipeline first.
• Discount / promote — move it through normal channels at a markdown. Model the markdown against the carry you'll save. A 30% discount today often beats holding at 22% annual carry for two more years.
• Re-purpose / cannibalize — use the component in another BOM, kit it, or pull it for service and spares. Engineering can sometimes substitute an excess part into a current product.
• Liquidate / broker — secondary markets, liquidators, B-stock channels. Cents on the dollar, but cents beat a write-off, and you free the rack space.
• Scrap and write off — last resort, but do it. Holding obsolete stock to avoid the P&L hit is the most expensive form of denial there is.

That last point deserves a word on the tax side. The IRS won't let you just decide something is worthless on paper.

What the IRS actually requires for a write-down

Under 26 CFR § 1.471-2(c) (Cornell LII, current), goods that are "unsalable at normal prices or unusable in the normal way" may be valued below cost at "bona fide selling prices less direct cost of disposition." But there's a catch: a bona fide selling price means an actual offering of the goods within 30 days of the inventory date, and the burden of proof sits with you.

Translation: you can't claim the deduction by sentiment. Keep the count sheets, the disposal records, and the destruction certificates. The write-down is real money, so document it like real money.

Step 3 — Run the carrying-cost math out loud

People hold excess because the cost is invisible. Make it visible. Per APQC benchmarking data (current), carrying cost typically runs 20% to 30% of inventory value per year once you add up every component.

• Capital cost — the cash is locked, not earning. Often the biggest single piece.
• Warehousing, handling, and space — racks, labor, the building itself.
• Insurance and taxes — they scale with what you hold.
• Shrinkage, damage, and obsolescence risk — the longer it sits, the more of it goes bad.

So $2M in excess inventory burns roughly $400,000 to $600,000 a year just to sit there. Put that number in front of the team and the "let's hold it" arguments get a lot shorter. This is the same release the CFO cares about. McKinsey makes the point that optimizing working capital early in a transformation (2024) builds momentum precisely because the cash is real and the wins are fast.

If you want to translate this into a board-ready number, inventory turnover ratio benchmarks give you the language finance already speaks.

Step 4 — Attack the root causes, or it grows back

Liquidating without fixing the cause is mowing weeds. Excess comes from a short list of repeat offenders. Fix the top one or two driving most of your E&O and you change the trajectory, not just the level.

• Stale reorder points and safety stock. Buffers set years ago, never recomputed, sized for demand that no longer exists. The most common cause by far. Recompute them with a real method, not gut feel: see how to calculate safety stock.
• MOQ over-buying. Supplier minimums force big buys on low-runners. Negotiate the MOQ, find an alternate supplier, or accept that the part shouldn't be stocked at all.
• Forecast bias. Sales sandbags high "to be safe," planning builds to it, demand doesn't show. Measure bias by SKU and the over-forecasters reveal themselves. Our guide on forecast bias: how to measure and eliminate it walks the math.
• Engineering changes with no sell-through plan. A revision obsoletes the old part overnight and nobody planned the runout. Build phase-out logic into the ECO process.
• Cancelled or shrunk customer programs. Built ahead for a program that pulled back. Tie build-ahead to firm commitments, not hopeful forecasts.

Step 5 — Put in a tripwire

The teams that stay lean don't run a heroic E&O cleanup once a year. They catch it monthly, while it's small and cheap to fix. The difference between a clean operation and a clogged one isn't smarter people. It's a process that fires monthly versus a cleanup that happens when someone finally complains about the warehouse.

Build a standing process with four parts:

1. Monthly E&O report, reviewed by planning and finance together, sorted by dollar value.
2. Aging alerts that flag any SKU crossing into the slow-moving band (6+ months supply), so you act before it goes obsolete.
3. An accountable owner with authority to discount, return, or scrap. Not a committee that defers every quarter.
4. A forward-looking view, not just rear-view. Tie the E&O signal to a forward demand forecast so you catch parts trending toward excess before you've over-bought.

Where AI earns its keep

That fourth point is where modern tooling changes the game. A monthly rear-view report tells you what already went wrong. A forward demand model flags a SKU whose forward demand is collapsing months before the warehouse fills.

The results are real, not vendor hype. McKinsey reports companies achieving a 15 to 20 percent improvement in inventory turns (2023) through better demand-and-supply alignment. Gartner predicts that 70% of large organizations will adopt AI-based supply chain forecasting by 2030 (2025). The early movers are already pulling carrying cost out of the building. If you're weighing the investment, start with AI inventory optimization for mid-market manufacturers.

A 30-day plan to move the number

You don't need a year. Here's the compressed version that works for a mid-market plant.

Run that loop once and you'll free meaningful cash. Run the tripwire forever and you keep it freed.

Frequently asked questions

What counts as excess inventory vs. obsolete inventory?

Excess inventory is on-hand stock that exceeds expected demand over a set horizon, typically anything beyond 6–12 months of forward demand at the current run rate. You'll likely consume it, but it's tying up cash now. Obsolete inventory has had no demand in the past 12 months and none expected, such as discontinued SKUs or superseded revisions, and should be written down to net realizable value under GAAP.

How fast can a manufacturer actually reduce excess inventory?

A focused team can move the number meaningfully in 30 days. Week one builds the E&O report, week two triages the top dollar-value items by recovery path, week three executes fast wins like re-selling and discounting, and week four recomputes reorder points and stands up a monthly tripwire. The heavy lifting is the top 40 high-value SKUs, which usually represent over half the trapped cash.

What is inventory carrying cost and why does it matter?

Carrying cost is the total annual cost of holding inventory, including capital cost, warehousing, insurance, taxes, and obsolescence risk. Per APQC benchmarking, it typically runs 20% to 30% of inventory value per year. That means $2M in excess inventory quietly burns $400,000 to $600,000 annually, which is why making the cost visible is the fastest way to win the "let's just hold it" argument.

Can I write off obsolete inventory for tax purposes?

Yes, but the IRS sets a bar. Under 26 CFR § 1.471-2(c), unsalable or unusable goods can be valued below cost at bona fide selling price less disposal cost, but only if you made an actual offering of those goods within 30 days of the inventory date. The burden of proof is on you, so keep count sheets, disposal records, and destruction certificates.

How does AI demand forecasting help prevent future excess?

Traditional E&O reports are rear-view: they tell you what already went wrong. AI demand forecasting is forward-looking, flagging SKUs whose forward demand is collapsing months before the warehouse fills, so you can cut purchase orders before over-buying. McKinsey has documented 15 to 20 percent improvements in inventory turns from better demand-and-supply alignment, and Gartner expects 70% of large organizations to adopt AI-based forecasting by 2030.