Six products. One data spine.

On screenTonight’s run proposes 142 moves (1,880 units across 22 doors); 86% of past runs accepted; 1 move held. Rules: min cover per door 2.0 weeks, run integrity at sender, max 8 moves per door per night, transfer cost ceiling $2.50/unit.

Iris saysThe run is weighted by conversion, not depth — Lille holds the units but Paris sells M and L twice as fast. One move is held back because it would open an S gap at Rennes to close an M gap at Bordeaux; that trade is not worth making.

On screen6 of 22 doors breaking on M and L; 84 units to move from 3 slow doors; 11 days full-price runway if moved today; 4.2 pts margin at risk if marked down instead.

Iris saysThe break concentrates in the four doors that took the first allocation and were never topped up — that says allocation, not demand. Moving 84 units from the three slowest doors holds full price 11 more days; markdown instead costs 4.2 points.

On screen1,840 versioned client profiles; 2.1 alterations per order (−0.4 vs last year); 64% fit first time (+9 pts); 3 files stale >24 months. Each alteration round costs ~6 atelier hours.

Iris saysA client’s file at v4 carries shoulder and fall adjustments from 2023. Applying it to the new order removes an expected fitting round — six atelier hours in a week when zardozi is already over-committed.

On screen4 orders at risk in week 34; highest slip probability 0.72; zardozi stage committed 118 of 80 available hours (over by 38).

Iris saysAll four at-risk orders route through one stage and one pair of hands — zardozi. This is not four problems, it is one. Moving the embellishment on two orders to the contract karigar clears the stage and every promise date holds.

On screenSeason readiness 92.4% vs 98% target; 3 schools 19 days behind last year; 5 weeks to a fixed intake date; intake week alone is 46% of the season.

Iris saysTwo schools are three weeks behind on a date that cannot move. There is no recovery window after intake week — last year the same gap cost 14% of intake footfall to queue walkouts. Pull both reorders forward now and it clears without airfreight.

On screenModel accuracy 90.6% by band; age 9–10 is −8% vs plan while 11–14 over-sells by +16%; 4,600 families moved up a band.

Iris saysAge 9–10 is not losing customers, it is losing them upward — the two variances are the same children. Rebalancing the buy between the bands fixes both numbers at once.

On screenReturn rate 16.8% vs 14% target; 58% of returns cite fit; guided silhouettes return at 10.1% vs 24.6% un-guided; compression base layer (31% returns) has no guidance.

Iris saysThe evidence is already in your own data — guidance halves the return rate where it exists. Compression is the worst line and has no guidance at all. This is two silhouettes missing a rule, not a modelling problem.

On screenProjected sell-out 89.3% (+11 pts); 140 pairs to move from 2 flagship doors; 48-hour clearance window across 92 doors.

Iris saysThe drop is allocated by store size, not by who buys this product. Two flagships hold 41% of units on 18% of conversion while streetwear doors clear everything in hours. Moving 140 pairs lifts sell-out to 89% inside 48 hours.

On screen9 articles broken on UK 8 and 9 across 23 doors; runs 91.8% complete vs 96% target; UK 8 + 9 are 46% of all pairs sold; £38k weekly sale at risk.

Iris saysThe run looks 92% complete because the count is dominated by sizes nobody buys. Filling these nine runs first is worth more than this week’s entire new-arrival intake — and the KPI should be re-weighted by demand.

On screen8-week projection 68.2% (±2.8 pts); model accuracy 92.2%; −6.4% vs buy plan; formal & dress projects 44% while every other category is at or ahead of plan.

Iris saysThe shortfall is one category, not one article — office footfall has not returned to these doors. Cutting the formal balance and moving it into boot, which is running ahead, lands the season on plan.

On screenShortest cover left 38 days on 2 batches; expiry write-off 0.9% vs 2.4% last year; £46k of stock inside 90 days across 63 counters.

Iris saysTwo batches have 38 days of life against 91 days of stock, and they sit in the three slowest counters. Moved to the counters that turn skincare twice as fast, both clear at full price instead of becoming a write-off.

On screenShade availability 94.2% vs 97% target; 6 deep shades broken; fastest shade sells 3.1×; only 31 of 63 counters carry the full deep range.

Iris saysThe deep range is not under-selling, it is under-distributed — where it is stocked it hits plan. One deep shade is out in 14 counters while a light shade sits on 9.8 weeks of cover. That is an allocation decision made on the wrong data.

On screenTrace time 0.8s roll to dye lot; 1,240 live rolls across 38 branches, 100% scanned on receipt; 3 cuts today split a lot; tolerance ΔE 1.5 (suiting lots measured at ΔE 3.1).

Iris saysTwo cuts span lots with shade drift above tolerance — a suiting order at ΔE 3.1 will show as two different fabrics once made up. The customer will not see it in the shop; they will see it at the first fitting, and by then the cloth is cut. Hold the cuts and source a single lot.

On screenCut yield 96.4% vs 97.5% target; wastage 3.6%; £8,400 recoverable remnant value; one branch loses 11.4% on cuts under 2 metres.

Iris saysThe branch is not careless across the board — it loses yield only on short cuts. One table, one habit, one cut length. A minimum-cut rule and priced offcuts recover most of the loss without retraining anyone.

On screen68 orders in pipeline; on-time install 93.8% vs 96% target; 2 orders promised the same day to the same fitter; 19 days measure to install; Friday booked 16 of 8 hours while Thursday has 3 free.

Iris saysThe diary accepted both orders because it books by day, not by hours. Moving the single-room curtain to Thursday keeps both promise dates — and the diary should switch to hours.

On screenSet attach 37% (+3 pts); 2.4 pieces per basket; £64 uplift when a set is completed; 82% of attaches happen in the same basket; rug underlay attaches at only 18%.

Iris saysAn attach not made at the till is usually never made. Rug underlay is not optional, yet no prompt was ever written for it — add the underlay prompt and bundle it at point of sale.

On screen0 oversells this period across 9 channels; pool reconciles 1.4s after each sale; before the single pool: 34 oversells and 3 account warnings in 12 months; one marketplace showing 46 units behind the pool.

Iris saysThe 46-unit gap is a manual safety buffer someone set and never removed. It is not an oversell risk — it is 46 units you are refusing to sell on your largest channel. Remove the buffer and audit all channels.

On screenMargin after fees 31.4% vs 46% gross; one channel negative; ad spend 6.2% of revenue; on the weakest channel every £100 of gross loses £14 commission, £22 returns and £11 ads.

Iris saysThe biggest marketplace is the best channel on revenue and the only negative one on contribution — and returns, not commission, do the damage. Fix the return reasons, cut ad spend there and reprice, or the volume is costing money.