# Huxberry Flagship Decision — Evidence Dossier

_Compiled 21 August 2026. This is the verification-grade record of the flagship-location analysis. It is written for a review process in which every statement will be independently re-verified: each claim carries a tag and, where applicable, a pointer sufficient to re-run the check. Verdicts rest only on the evidence cited — independent of any position previously held by any principal._

**Claim tags used throughout:**
- `[FACT-VERIFIED]` — externally checkable; verification pointer attached
- `[FACT-INTERNAL]` — true in company systems; query/source attached
- `[REPORTED]` — stated by a principal; externally unverifiable
- `[INFERENCE]` — analytical conclusion from tagged facts; premises listed
- `[JUDGMENT]` — opinion or management view; owner named
- `[HYPOTHESIS-UNTESTED]` — testable, not yet tested

---

## 1. Decision context

- Current flagship: ~5,000 sq ft on Sheikh Zayed Road, rent ~AED 700,000/yr (≈140/sq ft). `[REPORTED — principals; lease document would verify]`
- Lease expiring in ~60 days from mid-Aug 2026; landlord's renewal ask AED 2,000,000/yr (400/sq ft). `[REPORTED — Rajiv]`
- Working budget cap AED 1,200,000/yr. `[JUDGMENT — Rajiv; a policy choice, not a market fact]`
- Second store: Dubai Festival Plaza (Jebel Ali Village), ~3,000 sq ft, ground floor SU102. `[REPORTED + FACT-VERIFIED: huxberry.com store listing]`
- 12-month webstore-recorded revenue AED 7.577M; 93% entered as staff-keyed draft orders (showroom-led). `[FACT-INTERNAL — ShopifyQL: FROM sales SHOW total_sales GROUP BY sales_channel SINCE -12m]`

## 2. Scope & definitions — what was keyed in and why

These boundaries fix earlier errors and must be preserved by any successor analysis:

1. **"Umm Suqeim Road" = the inland Al Barsha 2 corridor** between Al Barsha and Al Quoz — NOT the coastal Umm Suqeim villa district. Early analysis and Property Finder's "Umm Suqeim Road" location node refer to the coastal area; those listings were discarded. Corridor anchors confirming the definition: The Mattress Store (1/11 Umm Suqeim Rd), MEROË, The Bedroom by Intercoil, Art of Living Mall, Marina Home, Pan Emirates. `[FACT-VERIFIED — brand store-locator pages; themattressstore.com, thebedroom.com, artoflivingmall.com]`
2. **"SZR zone" = Interchanges 2–4 frontage** (Umm Al Sheif, Al Safa edge, Al Quoz 1–3 frontage) — because that is where the sleep-brand cluster sits: Hästens (Eiffel 1, Umm Al Sheif), Bed Quarter/Tempur (725 SZR), Simmons (4th Interchange), Serta. `[FACT-VERIFIED — hastens.com, bedquarter.ae]`
3. **Al Shafar Investment Building was ruled OUT of the zone**: its own listing text says "direct access to Sheikh Zayed Road & Hessa Street" — the Hessa St junction is the Barsha Heights/TECOM office district (≈Interchange 5), outside the sleep-brand zone. `[FACT-VERIFIED — bayut.com/property/details-16195967.html, retrieved 21 Aug 2026]`
4. **"Inner Al Quoz"** = interior industrial streets (Al Quoz Ind. 1–4 away from arterials), distinguished from Al Quoz *frontage* on SZR or Umm Suqeim Rd — the rent tiers differ by 2–3×.
5. **Huxberry B2C retail in the ERP** = NetSuite sales orders with line-level subsidiary 13 ("Huxberry Trading LLC") AND `otherrefnum LIKE 'HUX-%'`. Every webstore order is keyed into NetSuite twice (retail SO + SpringTech intercompany SO under customer CUS-03146 shipping to Huxberry's own PO box); only the subsidiary-13 side counts, or revenue double-counts and geography is garbage. `[FACT-INTERNAL — SuiteQL, §15.2; verified against Shopify to the fils on 10 orders]`

## 3. Verdict summary

**Core finding:** the assumption that collapsed all options to ~3,000 sq ft — "every brand-safe corridor costs ≈AED 400/sq ft" — failed verification in all three corridors against 30+ live listings. `[FACT-VERIFIED — §4]` At actual asking rents, AED 1.2M buys a full-size (4,000–5,500+ sq ft) flagship, kitchens and sofas retained, in every candidate corridor. `[INFERENCE — from §4 listings + §5 mix]`

**Second core finding:** monthly revenue tracks the advertising budget (r = 0.87), not road visibility — including a natural experiment in April 2026. `[FACT-INTERNAL + INFERENCE — §9]`

### 3.1 Option ratings (asking prices, 21 Aug 2026; all pending broker ground-truth)

| # | Option | Size sq ft | Asking (per sq ft) | Rating | Rationale (tagged) |
|---|---|---|---|---|---|
| 1 | Umm Suqeim Rd frontage, Al Barsha 2 (avail. Sept) | 5,138 | 1.5M (292) | **9/10** | Full size keeps kitchens+sofas `[INFERENCE §5]`; on the mattress cluster `[FACT-VERIFIED §2.1]`; between both customer masses `[FACT-INTERNAL §8]`; needs ~20% negotiation to cap `[FACT — ask vs cap]` |
| 2 | Hessa St, Al Barsha 3 new strip | 4,002 | 1,201,200 (300) | **8/10** | At cap exactly `[FACT-VERIFIED]`; shell = fit-out capex `[FACT]`; 1 block off frontage `[FACT]` |
| 3 | Umm Suqeim Rd, Al Quoz Ind. 4, divisible | 8,500 | 1.275M (150) | **7.5/10** | Cheapest full-size on corridor `[FACT-VERIFIED]`; less-established end, heavy mezz `[JUDGMENT — analyst]` |
| 4 | In-mall Al Barsha 2 (plausibly Art of Living Mall) | 6,318 | 758k (120) | **7/10** | Most space/dirham `[FACT-VERIFIED]`; mall underletting signal: 35 stores vs "50+" positioning `[FACT-VERIFIED — Wikipedia/artoflivingmall.com]` |
| 5 | Al Joud Centre, on SZR nr Equiti Metro | 3,833 (combos to 7,000) | 680k (177) | **7/10** | Real SZR address at ~1/3 Eiffel rate `[FACT-VERIFIED]`; set-back, mezz-heavy shell `[FACT — listing]`; premium-read untested `[HYPOTHESIS-UNTESTED — needs visit]` |
| 6 | Jumeirah/Al Wasl (best units) | 3,100–3,250 | 487k–991k (150–316) | **6.5/10** | Coastal catchment + luxury-furniture neighbours `[FACT-VERIFIED — OBEGI, Maison Lamassu §4.4]`; no category cluster `[FACT-VERIFIED]`; roadworks ahead `[FACT-VERIFIED §14]` |
| 7 | Renewal, current store | 5,000 | 2M ask (400) | **F at 2M / 7 if ≤1.2M** | 400/sq ft exceeds every full-size comp found `[FACT-VERIFIED §4]`; 2M = 26% occupancy ratio `[FACT-INTERNAL]` |
| 8 | Eiffel 1 (Hästens bldg) | 2,400 | 1.2M per lease terms (500); possibly 900–950k `[REPORTED — Rajiv]` | **5/10 as flagship** | Best adjacency `[FACT]`; drops kitchens AED 1.06M `[FACT-INTERNAL §5]` + sofas `[FACT-INTERNAL §6]`; needs ≈2.2× sales density `[INFERENCE §10]`; smaller than Festival Plaza `[FACT]` |
| 9 | Al Shafar Bldg | 3,100 | 980k (316) | **3/10** | Outside the zone `[FACT-VERIFIED §2.3]`; dominated by rows 2 & 5 on price, size and location `[INFERENCE]` |
| 10 | Inner Al Quoz | 4,000–6,000+ | 88–200/sq ft | **5/10 — parked** | Ad-fed model makes it viable `[INFERENCE §9]`; excluded from the comparison circuit `[INFERENCE §11.4]`; saves only 0.3–0.7M vs corridor `[FACT-VERIFIED §4]` |

## 4. Rent evidence (live listings, retrieved 21 Aug 2026)

Method: direct fetches of Bayut, Property Finder and dubizzle search/listing pages plus brokerage stock. Every listing cited was individually observed with URL, agency and reference. Asking ≠ achieved; excludes service charges, DREC, fit-out. `[FACT-VERIFIED — all rows]`

### 4.1 Umm Suqeim Rd / Al Barsha 2 corridor
| Unit | Size | Ask AED/yr (psf) | Pointer |
|---|---|---|---|
| Al Barsha 2 frontage (GF 3,844+mezz), avail. Sept | 5,138 | 1,500,000 (292) | bayut.com/property/details-15708571.html — Blue Harbour, ref 10684-mse9nS |
| Al Barsha 2 frontage (GF 3,500+mezz 3,254) | 6,754 | 2,026,200 (300) | bayut.com/property/details-16206515.html — Engel & Völkers, ref G-04B7ED |
| Al Barsha 2 frontage (GF 7,723+mezz) | 10,416 | 3,000,000 (288) | bayut.com/property/details-15920025.html — Blue Harbour |
| Hessa St, Al Barsha 3 — 4 new shell units @300 | 4,002–4,736 | 1,201,200+ (300) | propertyfinder.ae …131120160.html — Arabian Sands, ref asre-626284 |
| Umm Suqeim Rd, Al Quoz Ind. 4, divisible G+M | 8,500 | 1,275,000 (150) | dubizzle …283106 — Superior Elite, ref STL-12 |
| In-mall Al Barsha 2, open layout | 6,318 | 758,160 (120) | bayut.com/property/details-15968586.html — Blue Harbour |
| On Umm Suqeim Main Rd, Al Quoz Ind. 3, semi-fitted | 2,500 | 490,000 (196) | propertyfinder.ae …120671263.html — Golf Hills |
| Al Quoz Ind. 4 main-road G+M, 50 kW | 3,012 | 903,600 (300) | bayut.com/property/details-16101616.html — D A W, ref DWD-3012 |
| Al Quoz Ind. 3 new-build scheme (mid/front units) | 2,354–4,883 | 250k–600k (106–150) | bayut.com/property/details-15878943 / -15878750 / -15917901 — Blue Harbour |

Corridor read: frontage 288–300; Ind. 3/4 side 100–220 (premium G+M to 300); in-mall ~120. Independent calibration: Engel & Völkers area guide "showroom and trade units — AED 297/sq ft" (engelvoelkers.com/ae/en/area-guide/al-quoz). `[FACT-VERIFIED]`

### 4.2 SZR Interchanges 2–4
| Unit | Size | Ask AED/yr (psf) | Pointer |
|---|---|---|---|
| Eiffel 1 (Hästens bldg), fitted, SZR-facing — headline 1.0M but lease-terms section states **1.2M** | 2,400 | 1,000,000 headline / 1,200,000 terms (417–500) | bayut.com/property/details-16169619.html — Nine Village RE (Ahmed Aly) |
| Eiffel 1, corner, shell&core — "450/sq ft + VAT" | 2,368–2,373 | 1,065,600 (450) | bayut.com/property/details-16125320.html — Colliers (Simon Whittle); PF dup by JLL …110960780 |
| Eiffel 1, fitted, high ceilings | 2,368 | 1,184,030 (500) | bayut.com/property/details-12896479.html — Chestertons (Nasry Isa) |
| Rawabeh Bldg, Al Quoz Ind. 1 SZR frontage, fitted | 2,956 | 1,034,737 (350) | propertyfinder.ae …132947700 — Colliers; CRC dup …112752539 |
| **Al Joud Centre, directly on SZR nr Equiti Metro, shell G+M, combos 3,833–7,000** | 3,833 | 680,000 (177) | propertyfinder.ae …133386148 — McCone (Eisa Furniturewala) |
| Al Joud internal units | 3,913–3,981 | 778k–861k (195–220) | PF …104118089 / …104117998 — CRC |
| Al Quoz Ind. 1, "direct connectivity to SZR", 66 kW | 3,899 | 905,000 (232) | bayut.com/property/details-16190993.html — Landwell |
| Al Wasl Center, 194 SZR collector (I2), direct frontage, vacant since Nov 2025 | 3,369–3,445 | 673,800–758,025 (200–225) | PF …110037347 (CRC) / …112753933 (Kennedy) / bayut 16168314 (AKAS) |
| Al Moosa Tower 1 (Trade Centre — outside zone, calibration) | 5,100 | 1,326,000–1,351,500 (260) | bayut 16220324/16236388 |

Zone read: **two tiers** — premium purpose-built (Eiffel class) 417–500; second-tier frontage/collector 177–240. `[FACT-VERIFIED]` The claim "no SZR 4,000–5,000 sq ft supply near 1.4M exists" is refuted by the Al Joud combination options and Al Wasl Center. `[FACT-VERIFIED]`

### 4.3 Jumeirah / Al Wasl (3,000 sq ft band)
| Unit | Size | Ask (psf) | Pointer |
|---|---|---|---|
| Umm Suqeim 2, Beach Rd area, indoor+outdoor | 3,139 | 991,025 (316) | bayut.com/to-rent/shops/dubai/umm-suqeim/ (21 Aug) |
| Jumeirah 3 community mall | 3,084 | 1,079,400 (350) | bayut.com/to-rent/shops/dubai/jumeirah/ |
| Al Wasl Rd (Jumeirah 1), FITTED showroom | 3,247 | 487,050 (150) | bayut.com/to-rent/shops/dubai/al-wasl/ — verify the catch |
| Small prime Beach Rd units (calibration) | 530–1,384 | 325–446/sq ft | same indexes |
| Canal Front Residences new stock (calibration) | 1,032–4,307 | uniformly 500/sq ft | same indexes |

Band read: 150–350 at ~3,000 sq ft; the planning figure "≈400" was top-of-range, not the centre. `[FACT-VERIFIED]`

### 4.4 Comparable-brand map (why the corridors matter)
- Umm Suqeim Rd corridor is Dubai's de-facto bed & mattress mile (§2.1 anchors). `[FACT-VERIFIED]`
- Coastal Jumeirah holds the luxury-furniture set: OBEGI Home 3,000 m² flagship (Jumeirah Beach Rd), Maison Lamassu (Fendi Casa/Baxter/Living Divani — Umm Suqeim 3, opposite Jumeirah Beach Hotel; maisonlamassu.com/contact), The One, BoConcept. `[FACT-VERIFIED]`
- Inner Al Quoz has NO premium sleep flagship on interior streets; only Heveya (appointment-led "Sleep Studio", Al Quoz Ind. 3; heveya.ae) and Intercoil (factory-tied). `[FACT-VERIFIED]`
- CarpetLand is NOT on Umm Suqeim Rd (sole showroom: W/H 64, 23rd St, Al Quoz Ind. 3; carpetland.ae/contact). Corridor flooring anchors are Carpet Centre, Floorworld, The Red Carpet. `[FACT-VERIFIED — corrects a principal's recollection]`
- Al Quoz industrial rents rose from AED 33/sq ft (Q2 2021) to ~85–100 (2025) to ~90 (H1 2026, +6% YoY) — a 4.5-year climb, not a recent tripling. `[FACT-VERIFIED — Knight Frank H1-2024 PDF, H1-2025 & H2-2025 reviews, H1-2026 release]`

## 5. Product mix — the definitive classification

**Method** `[FACT-INTERNAL]`: all 736 retail SOs (12m to 21 Aug 2026; AED 7,042,133 inc-VAT headers = 93% of webstore total) pulled via SuiteQL; 4,675 lines; revenue per line = −netamount (ex-VAT, GL-signed — naive abs() would inflate by AED 2.71M via advance-payment pairs); 2,748 null-netamount BOM component lines excluded; reconciles to headers 736/736; 0% of line value unclassifiable. Queries in §15.2.

| Category | All orders (Cut A) | Excl. 82 closed (Cut B) |
|---|---|---|
| Mattresses | **53.4%** (AED 3,584,676) | 62.5% |
| Kitchens | **15.7%** (AED 1,056,476) | 8.7% (AED 423,762) |
| Beds & headboards | 12.3% (822,223) | 8.2% |
| Bedding & accessories | 8.8% (590,927) | 11.1% |
| Sofas & living | 4.0% (268,100) | 2.6% |
| Other furniture / wardrobes / services | 5.8% | 6.9% |

- The previously believed mix ("mattresses 76.2%, other <1%") was computed on only the AED 3.56M of webstore revenue with products attached — 47% of the total. The missing 53% contained the entire kitchens/wardrobes/custom business. Within its own four categories the old figure was roughly right (like-for-like: 74/10/13/3 on non-closed orders) — the error was treating a 47%-coverage statistic as the whole business. `[FACT-INTERNAL + INFERENCE]`
- **Closed-order check:** 82 SOs (27% of gross value, incl. 60% of kitchen revenue) are status "Closed" — potentially abandoned sales. All three closed sofa orders inside the webstore API window (HUX-3057, HUX-3064, HUX-3083) are **PAID, AED 0 refunded, not cancelled** in Shopify. `[FACT-INTERNAL — Shopify Admin GraphQL, §15.1]` → closure is a fulfilment-workflow artifact; **Cut A is the right read**. `[INFERENCE — 3-order sample; full workflow confirmation with sales team still recommended]`

## 6. Sofa trajectory

- Webstore classified sofa sales: May–Jul 2026 = AED 101.4k vs AED 0 in the same months of 2025; the entire prior year (Aug 2024–Jul 2025) ≈ AED 5.0k. `[FACT-INTERNAL — ShopifyQL: FROM sales SHOW total_sales WHERE product_type = 'Sofa' TIMESERIES month SINCE -26m]`
- ERP incl. custom lines: Jun 55.3k / Jul 56.9k / Aug-MTD 12.5k vs ~13.3k/month average over the preceding 8 months — a ≈4× step-change, ≈6.5% of last-3-month revenue. `[FACT-INTERNAL — §5 dataset]`
- Hero product: Kate Modular Sofa family ≈ AED 170k of the year's 268k. `[FACT-INTERNAL]`
- Festival Plaza attribution: **no field in Shopify or NetSuite records which store keyed an order** — the FP explanation is plausible on timing but unprovable in systems today. `[FACT-INTERNAL — absence verified]` Fix: tag draft orders by store. `[JUDGMENT — analyst recommendation]`

## 7. Kitchens economics

- Revenue is real: AED 1.056M/yr (Cut A; 0.42M on Cut B) across 12 kitchen-project SOs, worktops, and built-in appliances (Miele, Liebherr, Siemens, Bosch, Franke, Quooker). `[FACT-INTERNAL]`
- It is episodic: 72% of kitchen revenue landed in 3 of 12 months (Aug–Oct 2025: 750k). `[FACT-INTERNAL]`
- It is ad-fed: ~49% of recent Meta spend is kitchen lead-gen (§9.2). `[FACT-INTERNAL]`
- "Kitchens need in-store designers/reps and profitability is unproven." `[JUDGMENT — Nish; consistent with the episodic pattern but not measurable in sales data]`
- Consequence: since full-size space is affordable (§4), kitchens is now a **staffing/P&L decision, not a property decision**. `[INFERENCE]`

## 8. Customer geography (full year)

Method `[FACT-INTERNAL]`: shipping-address text of all 736 retail SOs, regex-classified by district; 91.6% of revenue classified, 5.5% blank, 2.9% residual. Query in §15.2.

| Cluster | Share |
|---|---|
| Southern/inland villa mass (Arabian Ranches 11.5% — #1 community; Emirates Living 7.3%; Victory Heights 4.4%; Dubai Hills 4.0%; MBR City 3.7%; Tilal Al Ghaf 2.5%; others) | **~31%** |
| Coastal mass (Jumeirah 1–3 7.6%; Palm 4.2%; Umm Suqeim 3.0%; Al Safa/Al Wasl 2.0%; Beachfront 1.5%) | **~18%** |
| Downtown / Business Bay / City Walk / DIFC | 8.4% |
| Abu Dhabi + Al Ain | **10.7%** |
| Sharjah + northern emirates | ~3% |

- Cross-check: ShopifyQL full-year GROUP BY shipping_city gives the same emirate split (Dubai ≈81%, Abu Dhabi ≈8.7% of total incl. blanks). `[FACT-INTERNAL]`
- Inference: buyers are villa families making planned trips; a tenth of revenue drives in from Abu Dhabi — destination-purchase behaviour is real. Every candidate corridor sits between the two masses → **drive-time does not crown any candidate**. `[INFERENCE — held by the prior analysis too; it survives]`

## 9. Advertising evidence — the demand engine

Data pulled 21 Aug 2026 from both ad accounts (currency AED on both; Aug 2026 partial to the 20th). Access pointers in §15.3.

### 9.1 Spend & output, trailing 12 months (Sep 2025 – 20 Aug 2026)
| | Google Ads | Meta | Combined |
|---|---|---|---|
| Spend | **468,610** (73%) | **173,519** (27%) | **~642,129 = 8.5% of revenue** |
| Impressions | 18.45M | 9.39M | ~27.8M |
| Clicks | 220,863 | 121,747 | ~343k |
| Store signals | ~26k directions-clicks/yr; ~1,015 measured store visits since May 2025 (a modeled floor); 2,304 calls/6m | 1,336 WhatsApp conversations; 344 leads | — |

24-month totals: Google 869,850 · Meta 285,905 · combined ≈ 1.156M. `[FACT-INTERNAL]`

### 9.2 Composition
- **~35% of recent Google spend is explicit local/store-visit campaigning**: Pmax_Local_Dubai (47,107, 6m), Arabic_Local (31,466), plus seasonal in-store PMax pushes; PMax = 71% of Google spend overall. `[FACT-INTERNAL — GAQL campaign segmentation]`
- **~49% of recent Meta spend is kitchen lead-gen**: Kitchen_Leads (25,039; 92 leads) + wa_kitchens WhatsApp (9,517; 393 conversations). `[FACT-INTERNAL — Graph API campaign insights]`
- Meta monthly spend stepped from ~8–10k (2024–mid 2025) to ~15–19k from Sep 2025. `[FACT-INTERNAL]`

### 9.3 Monthly spend vs revenue (full months; AED k)
| Month | Ad spend (G+M) | Webstore revenue | Month | Ad spend | Revenue |
|---|---|---|---|---|---|
| 2025-09 | 58.2 | 880.3 | 2026-03 | 32.3 | 253.3 |
| 2025-10 | 60.9 | 804.5 | 2026-04 | **30.2** (Meta 1.8) | **281.5** |
| 2025-11 | 64.0 | 631.3 | 2026-05 | 47.6 | 401.3 |
| 2025-12 | 62.8 | 631.6 | 2026-06 | 62.4 | 759.6 |
| 2026-01 | 63.6 | 677.6 | 2026-07 | 62.6 | 871.8 |
| 2026-02 | 55.6 | 634.9 | | | |

- Pearson r = **0.869** over the 11 full months; **0.595** excluding Mar+Apr. `[FACT-INTERNAL — computed from the two series above; reproducible]`

### 9.4 The natural experiment — stated with precision
During Ramadan (17 Feb–18 Mar 2026) ad spend was cut ~45% and revenue fell 60% — confounded, since both demand and spend drop in Ramadan. **April 2026 is the clean month**: no religious suppression, ads still near-dark (AED 30.2k; Meta 1,776), the store on the same SZR frontage with the same passing traffic — and revenue stayed at 281k (−60% vs trend). Ads restored May→Jun (48k→62k); revenue followed (401k→760k→872k, the two best months of the year). `[FACT-INTERNAL — §9.3]`

**Conclusion, precisely bounded:** this is strong evidence that **supports** "advertising, not road visibility, fills the showroom" — the store's own frontage held constant while revenue tracked the media budget. It is not controlled proof (n = 1 dark month; a demand-side April shock cannot be fully excluded); the geo-targeted demand test remains the controlled confirmation. `[INFERENCE — premises: §9.3 correlation, April observation, constant frontage]`

**What was deliberately NOT used:** Meta's attributed purchase value (AED 5.04M on 174k spend, a 29× ROAS) and Google's "conversions" column (29,936 — dominated by local-action and page-view-type events, only ~129 tracked purchases/6m). Platform attribution is inflated and appears nowhere in the conclusions. `[JUDGMENT — analyst method choice, stated for transparency]`

### 9.5 Implication for the SZR premium
The premium-tier SZR ask (Eiffel class 417–500/sq ft) exceeds corridor frontage (288–300) by ~AED 250–500k/yr at 2,400–5,000 sq ft — i.e. **40–80% of the entire annual ad budget** — to buy drive-by visibility that §9.4 indicates does not drive revenue. The address premium and the Pmax-Local campaigns purchase the same customer trip; one is measurable and adjustable monthly, the other is a multi-year lease. `[INFERENCE — premises: §4.2 rents, §9.1 spend, §9.4]`

Side finding: the Aug 2026 webstore anomaly (632k sessions vs ~7–15k/month baseline) is NOT paid traffic — combined paid clicks in August are ~23k — it is bot/scraper traffic polluting analytics, worth blocking, not budget burn. `[FACT-INTERNAL — ShopifyQL sessions series vs platform click counts]`

## 10. Small-format economics

| | Current store | Small-SZR (Eiffel @950k `[REPORTED]`) | Full-size second-tier (§4 opts 1–5) |
|---|---|---|---|
| Size | 5,000 | 2,400 | 3,833–5,138 |
| Rent (psf) | 700k (140) | 950k (396) | 680k–1.5M-ask (150–300) |
| Rent change | — | **+36% rent for −52% space** | ≈flat to +71% for ≈full space |
| Categories carried | All | Mattresses (+token) — kitchens & sofas out | All |
| Sales density needed for 15% occupancy | 1,520 AED/sq ft (actual) | **~2,640 = 1.74× actual, with ~20% of categories cut** | ≈at or below actual |

- Layout check at 2,400 sq ft: ~1,450–1,600 net display after BOH/circulation; ~90–110 sq ft per displayed bed → 14–16 beds with zero sofas/kitchens; keeping 2–3 sofa vignettes (150–250 sq ft each) drops beds to ~9–11 — fewer than the range itself (12+ models across Huxberry organic + Hypnos lines) — next door to Hästens' ~4,000 sq ft mattress-only flagship. `[INFERENCE — standard space-planning allowances applied to the verified range list; the physical layout exercise remains open]`
- The flagship would be smaller than Festival Plaza (3,000). `[FACT]`

## 11. Where data replaced intuition — the explicit record

1. **"SZR visibility drives our sales"** → the store kept its frontage all spring while revenue tracked the ad budget (r = 0.87; April experiment §9.4). Evidence supports ads, not address, as the demand engine. `[INFERENCE from FACT-INTERNAL]`
2. **"Every brand-safe corridor costs 400/sq ft"** (the premise that forced 3,000 sq ft) → 30+ live listings put corridor frontage at 288–300, SZR second-tier at 177–240, Jumeirah at 150–350; 400–500 is real only at the premium tier. `[FACT-VERIFIED §4]`
3. **"Sofas are only 4%, so a small store fits the mix"** → true trailing-12m, but the category ran ~0% a year ago and ≈6.5% in the last 3 months — a 4× step the annual average hides. `[FACT-INTERNAL §6]`
4. **"No one shops in Al Quoz" vs "location is irrelevant for destination retail"** → both feelings, both replaced: customers demonstrably travel (Abu Dhabi 10.7%), AND destination shoppers batch comparison visits, so cluster membership captures demand competitors generate — the measured rent gap for that membership is only 0.3–0.7M/yr. `[FACT-INTERNAL §8 + INFERENCE]`
5. **"Kitchens are a future option"** → AED 1.056M of existing, ad-fed revenue. `[FACT-INTERNAL §5, §7]`
6. **"Closed orders are lost sales"** → sampled closed orders are paid and unrefunded. `[FACT-INTERNAL §5]`
7. **"The landlord's 2M is market-consistent"** → above every observed full-size comp; it prices a 5,000 sq ft box at boutique-tier unit rates. `[FACT-VERIFIED §4]`

## 12. Fact-vs-conjecture ledger
The full 27-row ledger with per-claim evidence is maintained in the Evidence Update document, §2 (same folder; Google Doc review copy). It is normative: any successor analysis should extend that table rather than restate claims untagged.

## 13. Discussion record (positions, symmetrically stated)

- **Nish** `[JUDGMENT]`: destination-retail thesis; ≥4,000 sq ft; kitchens matter strategically; ads drive the showroom; skeptical of paying for SZR prestige. Evidence outcome: destination behaviour and ad-driven demand **supported** (§8, §9); full-size affordability **verified** (§4); kitchens revenue **verified** (§5); the strong destination claim (a non-retail address works) remains **untested**.
- **Rajiv** `[JUDGMENT]`: brand safety requires SZR-class adjacency; 1.2M hard cap; corridor rents ≈400; proposed Eiffel 1 at a believed 900–950k and Al Shafar. Evidence outcome: small-format pricing instinct **confirmed** at the premium tier (§4.2); the 400-everywhere generalization and "no full-size supply" **refuted** (§4); Eiffel proposal is best-in-class execution of a strategy (boutique-billboard) that the mix and ad evidence argue against (§10, §9); Al Shafar **ruled out** on location (§2.3). His cap, applied to verified rents, now *supports* a full-size corridor store.
- Convergence point: both principals want the brand protected and the cap respected. The evidence says both are achievable at full size on the Umm Suqeim Rd corridor; the residual disagreement is the value of premium-SZR adjacency, which §9 prices against the ad budget.

## 14. Open questions & next actions

1. Ground-truth calls (rent, service charges, DREC, parking, handover): options 1–5. Blue Harbour holds #1, #4 and the Ind. 3 scheme; Art of Living Mall leasing = CRC (exclusive), +971 52 143 7581 / info@artoflivingmall.com. `[FACT-VERIFIED — contacts from listings/mall site]`
2. Written Eiffel counter — validate the 900–950k `[REPORTED]` and resolve the 1.0M-vs-1.2M listing discrepancy.
3. Closed-order workflow: one question to the sales team (may double as Festival Plaza attribution). Start tagging draft orders by store.
4. Kitchens: staffing/P&L decision (designer + rep cost vs margin on ~1M revenue).
5. Renewal counter at 1.2M using §4 comps.
6. Geo-targeted demand test per shortlisted corridor — the controlled confirmation of §9.
7. Roadworks watch: Jumeirah St↔Al Khail project (incl. SZR interchange) awarded 26 Apr 2026, construction ahead, completion unpublished; the Al Barsha 2 stretch's own upgrade is complete (tunnel opened 3 Aug 2025). `[FACT-VERIFIED — Dubai Media Office releases 25/31 May 2025, 3 Aug 2025, 26 Apr 2026]`

## 15. Appendix — data access & re-verification guide

### 15.1 Webstore (Shopify)
Store "huxberry-dev.myshopify.com" (storefront huxberry.com), AED. Key queries:
- Mix belief source: `FROM sales SHOW total_sales, net_sales, orders GROUP BY product_type SINCE -12m`
- Channel: `... GROUP BY sales_channel SINCE -12m` → Draft Orders 93%
- Geography cross-check: `... GROUP BY shipping_city SINCE -12m`
- Sofa series: `FROM sales SHOW total_sales, orders WHERE product_type = 'Sofa' TIMESERIES month SINCE -26m`
- Sessions: `FROM sessions SHOW sessions GROUP BY referrer_source SINCE -12m UNTIL 2026-07-31` (Aug excluded — bot anomaly)
- Closed-order check: Admin GraphQL `orders(query:"name:HUX-3057 OR name:HUX-3064 OR name:HUX-3083"){... displayFinancialStatus totalRefundedSet cancelledAt}`
- Limitation: order-level API returns only ~5 weeks of history; aggregates go back years.

### 15.2 ERP (NetSuite) — SuiteQL
Retail definition and joins per §2.5. Warehouse note: as of 21 Aug 2026 the Metabase "NetSuite Warehouse" carries `otherrefnum`, `shipaddress`, `subsidiary_id/name` (ETL extended same day), so these pulls are re-runnable as plain SQL.
- Headers: `SELECT t.id, t.tranid, t.otherrefnum, t.status, t.trandate, t.foreigntotal FROM transaction t JOIN transactionline tl ON tl.transaction=t.id WHERE t.type='SalesOrd' AND tl.mainline='T' AND tl.subsidiary=13 AND t.otherrefnum LIKE 'HUX-%' AND t.trandate >= ADD_MONTHS(SYSDATE,-12)` → 736 rows, Σ 7,042,132.97
- Lines: same predicate with `tl.mainline='F' AND tl.taxline='F'`, fields item/BUILTIN.DF(tl.item)/quantity/rate/netamount → 4,675 rows
- Geography adds `t.shipaddress`. Raw pulls archived internally.

### 15.3 Ad platforms (credentials by location only — never in documents)
- Meta: ad account **act_445735369557596** ("Huxberry", AED, Asia/Dubai). Access: via the company credential vault (internal — ask Nish). Call used: `GET act_…/insights?level=account&time_increment=monthly&time_range={2024-08-01..2026-08-20}&fields=spend,impressions,clicks,actions,action_values` (+ level=campaign for the 6-month composition).
- Google Ads: customer **8578064711** ("AK | Huxberry", AED). Access: via the company's read-only reporting client (internal — ask Nish). GAQL: monthly `metrics.cost_micros, impressions, clicks, conversions, conversions_value` segmented by `segments.month`; campaign breakdown by `campaign.advertising_channel_type`; store visits via conversion-action segmentation (series starts May 2025).
- Rule inherited by any successor project: platform-attributed revenue is not evidence; spend, clicks, and the revenue correlation are.

### 15.4 Documents
- Evidence Update (companion, incl. 27-row ledger & verdict tables): `huxberry-flagship-evidence-update-2026-08-21.md` + Google Doc review copy.
- Research archive with full listing tables and session log: `worklog-2026-08-21.md`.
- All in `~/Documents/Projects/Projects - retail store location /` (folder name ends with a space).