Module 4: Inventory KPIs: Sell-Through, Turn, GMROI

Why This Module Matters

Module 1 defined sell-through, turn, and GMROI as three related but distinct inventory KPIs. This module puts real jewelry-specific benchmark numbers behind each one, and covers the practical markdown discipline that actually moves aged inventory before it becomes dead capital — the single biggest inventory-management risk in a category where individual pieces can sit unsold for well over a year without anyone noticing until a manager finally runs the report.

Jewelry’s Inventory Turn Is Genuinely, Structurally Slow

Multiple independent sources converge on a consistent picture: jewelry retailers run among the slowest inventory turnover of any major retail category, typically 1 to 2 turns per year, with small boutique or luxury-focused stores often sitting at the lower end of that range, around 0.7 to 1.5 turns annually. This is not a symptom of poor management — it’s a structural feature of the category, driven by high per-unit price, deliberate and considered purchase decisions, and a customer base that browses and researches for months before committing. A jewelry manager who benchmarks their own turn rate against a general-retail apparel or grocery number, where turn can run 4 to 12 times a year or more, will conclude their inventory is badly mismanaged when it’s actually performing normally for the category.

GMROI in Jewelry Runs Lower Than General Retail, and That’s Expected Too

A general retail rule of thumb treats a GMROI of 2 to 3 as healthy and 3.5 or higher as excellent, but jewelry specifically runs lower than this general benchmark — often clustering near 1.0 to 1.1 for mass-market jewelry, though luxury boutiques with stronger margin structures can reach 3.0 to 3.4. The explanation ties directly back to the turn benchmark above: jewelry’s high margins are partially offset by its very slow turnover, since GMROI multiplies margin percentage by turn rate, and a category with structurally low turn will produce a lower GMROI even with genuinely strong per-unit margins. A manager should read a jewelry store’s GMROI in this specific context — a mass-market jewelry GMROI near 1.0 isn’t automatically alarming, but a GMROI meaningfully below even that already-modest category benchmark is worth investigating.

A Meaningful Discrepancy Worth Flagging

Not every source agrees on the exact turn range: most sources cluster around 1 to 2 turns annually for jewelry generally, with boutique stores lower still, but at least one source places jewelry retailers as high as 2.0 to 4.1 turns per year, notably higher than the rest. This module treats the 1-to-2 range as the primary benchmark, since it’s the range independently corroborated across the largest number of separate sources, but flags the outlier figure as worth checking against a store’s own specific category mix and price point before drawing firm conclusions from any single benchmark number.

What “Normal” Aging Looks Like, By Category

Aging tolerance varies meaningfully by category, and treating all jewelry inventory against a single aging threshold produces bad decisions in both directions. Engagement rings and bridal jewelry naturally take longer to sell, with 12 to 18 months considered a normal window given the category’s deliberate, milestone-driven purchase cycle. Fashion jewelry moves fast or not at all, with roughly six months as a practical maximum window before trend relevance fades. Classic, timeless fine jewelry designs (solitaire pendants, diamond studs, traditional pieces) can sit comfortably for 12 to 24 months without necessarily signaling a problem. Seasonal pieces (Valentine’s Day, Mother’s Day, holiday-themed items) need to clear before the season ends, since a piece that misses its season effectively ages a full year in a single missed window.

A Tiered Markdown Ladder That Removes the Emotion

A specific, widely recommended structure for jewelry markdowns: apply a first, modest markdown of 10 to 15 percent once a piece crosses 12 to 18 months of age, testing whether a small price nudge is enough to move it to the right buyer. If that doesn’t work, apply a second, more serious markdown of 20 to 30 percent at the 18-to-24-month mark. If the piece still hasn’t sold past 24 months, move to a final clearance markdown of 40 percent or more — at that point, the piece is effectively dead stock, and the capital tied up in it is doing the store more harm sitting on a shelf than it would recovering even at a steep discount. This ladder structure matters specifically because it removes the emotional decision-making that keeps aged pieces sitting indefinitely: the schedule decides when a markdown happens, not a manager’s reluctance to admit a piece isn’t moving.

Beyond Discounting: Bounties, Bundling, and Vendor Returns

Several specific, non-discount techniques appear consistently across jewelry-trade sources for moving aged stock. Placing a bounty or spiff — a bonus paid to whichever associate sells a specific aged piece — on items just entering the aging process gives staff a direct financial incentive to feature that piece in customer presentations before it needs a markdown at all. Bundling an aging piece with a strong seller (pairing a slow-moving wedding band with a popular engagement-ring style, for instance) moves the older item without a direct price cut on either piece. And negotiating consignment or return arrangements with vendors, when the vendor relationship allows it, moves aged stock off the store’s own books entirely rather than requiring an in-house markdown at all.

Putting the Vocabulary to Work

The specific manager habit this module recommends: run an aging report regularly (many jewelry-specific POS/ERP systems support this natively), bucket inventory into clear aging tiers (0-90 days, 91-180 days, 181-365 days, 365+ days), and pair that report with the category-specific aging tolerance from earlier in this module before deciding whether a given piece actually needs attention. A ring that’s 14 months old is a non-event in bridal; the same 14-month age on a fashion-jewelry piece is well past its practical window and belongs in this module’s markdown ladder immediately.

Practice Exercise

Run (or request from whoever manages your store’s inventory system) an aging report broken into the four tiers above. For every item in the 181-365-day and 365+-day buckets, note its category (bridal, fashion, classic fine jewelry, seasonal) and compare its actual age against this module’s category-specific aging tolerance. Flag any item that’s aged past its category’s normal window and hasn’t yet received a markdown, and apply this module’s tiered ladder (10-15% first markdown, 20-30% second, 40%+ final clearance) based on how far past that window it sits.

Key Takeaways

• Jewelry retailers run among the slowest inventory turnover of any major retail category, typically 1 to 2 turns per year (boutique/luxury stores often 0.7-1.5), which is a structural feature of the category rather than a sign of poor management — benchmarking against general-retail turn rates produces a misleadingly alarming conclusion.
• GMROI in jewelry runs lower than the general 2-3 “healthy” benchmark, often clustering near 1.0-1.1 for mass-market jewelry, because GMROI multiplies margin by turn rate and jewelry’s high margins are partially offset by structurally low turnover.
• Aging tolerance varies significantly by category: bridal/engagement pieces normally take 12-18 months to sell, fashion jewelry has roughly a six-month practical window, and classic timeless designs can sit 12-24 months without necessarily signaling a problem.
• A tiered markdown ladder (10-15% at 12-18 months, 20-30% at 18-24 months, 40%+ after 24 months) removes emotional decision-making from aged-inventory management by letting a schedule, not a manager’s reluctance, decide when a markdown happens.
• Beyond discounting, bounties/spiffs on newly-aging items, bundling slow movers with bestsellers, and negotiating vendor consignment or return arrangements are specific, non-discount techniques for moving aged stock before or alongside a markdown.