# Button & Bear: choose the reset, earn the bundle

**Decision owner:** Maya Chen, founder. **Decision date in the case:** September 30, 2026. **Forecast:** October–December 2026. USD; synthetic US domestic DTC business.

**Recommendation:** Pilot B against A at comparable acquisition spending. Use B as the preferred operating policy if the pilot confirms contribution gains. Test C against B with limited stock exposure before broader rollout. This reflects your selected preference to pilot before scaling.

The stronger model outcome comes from list pricing and charging $4.95 delivery below a $45 merchandise subtotal. The bundle’s incremental advantage is small, composition-dependent, and requires more inventory cash. Do not scale solely because C leads a spreadsheet forecast.

## 1. Agreed analysis brief

Audience: founder choosing offers before purchase commitments. Primary outcome: operating profit, defined as total revenue less product cost, shipping/fulfilment, marketing and fixed overhead. Guardrails: stock available when promised and a $15,000 cash reserve. Scope: April–September actuals; Q3 versus approved budget and Q2; October–December scenarios for A/B/C. Deliver a reconciled historical explanation, comparable customer offers, transparent demand/cost assumptions, inventory/cash implications and a practical pilot.

The founder audience, scope and accounting definitions come from the case. You explicitly selected pilot-before-scaling in the clarification. Remaining demand, basket-substitution and timing assumptions are analyst choices, not business facts.

## 2. Q3: volume masked weak economics

| Metric | Q2 actual | Q3 budget | Q3 actual |
|---|---:|---:|---:|
| Orders | 4,800 | 5,000 | 6,000 |
| Units | 6,500 | 6,500 | 7,202 |
| Total revenue | $208,640.00 | $175,000.00 | $185,998.50 |
| Operating profit | $55,360.00 | $31,400.00 | $5,536.50 |
| Operating margin | 26.53% | 17.94% | 2.98% |

Q3 exceeded planned orders by 20.0% and revenue by 6.28%, but missed planned profit by $25,863.50, or 82.37%. Versus Q2, orders increased 25.0%, revenue fell 10.85%, and profit fell 90.00%. Quarter comparisons have changing offers and basket/acquisition mix; no experiment isolates causality.

**Exact ordered profit bridge:** start $31,400; flex to actual order volume at budget economics +$9,880; gross merchandise basket mix −$3,335; product-cost basket mix +$258; discounts −$20,666.50; shipping/fulfilment rate −$9,000; marketing rate −$3,000; end $5,536.50. Fixed overhead has no variance. Volume, mix and rate effects depend on this stated ordering; they are not causal estimates.

The plan had $35 gross merchandise/order, $14.12 product cost/order, $5 delivery/fulfilment and $6 marketing. Q3 actuals were $34.4442, $14.0770, $6.50 and $6.50 respectively. Discounts were 10% of merchandise. Customer shipping was zero both in actual Q3 and the budget, so lost shipping revenue is **not** a Q3 budget driver. It belongs in the Q2 comparison, where $14,602.50 was collected.

Q3 paid checkout-source orders were 3,900 versus 2,100 organic/direct; marketing spend was $39,000. $10 per paid-attributed order is a descriptive spend ratio, not CAC, incremental ROAS, or lifetime value. Customer identities, visitor conversion denominators and experimental controls are absent.

Sources: `monthly_actuals.csv`, `budget.csv`, `budget_product_units.csv`, `budget_baskets.csv`, `orders.csv`, `order_items.csv`, `marketing_attribution.csv`, `historical_policies.csv`; workbook `Company_Financials.xlsx`, all four sheets. Detailed bridge: `bridge.csv`.

## 3. Customer offer comparison

| Delivered purchase | A: 10% off/free delivery | B: list price/threshold | C: B plus bundle |
|---|---:|---:|---:|
| Pocket, 15 cm | $18.00 | $24.95 | $24.95 |
| Classic, 28 cm | $31.50 | $39.95 | $39.95 |
| Bedtime, 40 cm | $45.00 | $50.00 | $50.00 |
| One Pocket + one Classic | $49.50 | $55.00 | $49.00 with sleeve |

B raises Pocket’s delivered price 38.61% and Classic’s 26.83% versus A. The model’s assumed demand change must cover the entire pricing/shipping reset, not just discount removal.

Synthetic competitor snapshots: Little Paws Tiny Pal, 14 cm polyester, $21.95 delivered; Comfort Bear, 24 cm polyester, $34.95 delivered; Willow & Wool Heirloom, 30 cm organic-cotton outer/polyester fill, $43.95 delivered; Nightlight Goodnight Duo, 26 cm + 14 cm polyester bears plus sleeve, $48 delivered. C has larger written dimensions, 28 cm + 15 cm, at $49; sizes, materials and contents prevent a like-for-like quality conclusion. These offers do not prove demand or willingness to pay.

Sources: `competitor_snapshots.csv`, `products.csv`, learner PDF page 1, `planning_assumptions.csv`. No live shop claims are made.

## 4. Forecast and assumptions

September calibration uses 2,200 actual orders: gross merchandise/order $34.647727; inventory product cost/order $14.160909; 74.7727% of baskets below $45, implying B shipping revenue $3.70125/order. September units/order: Pocket 0.627273, Classic 0.415909, Bedtime 0.150909. Actual remote share was 27.2727%, close to management’s 27% base assumption.

| Assumption | Downside | Base | Upside |
|---|---:|---:|---:|
| A monthly demand vs 2,200 / 2,400 / 2,600 | 80% | 100% | 120% |
| B/C orders as fraction of A | 75% | 90% | 105% |
| C bundle share | 0% | 15% | 35% |
| Carrier surcharge/order | $2.00 | $1.50 | $1.00 |
| Remote-zone share | 35% | 27% | 20% |

All offers: carrier base $4.50, normal fulfilment $0.70, remote extra $1 when applicable; marketing $15,000/month and overhead $6,000/month. Fixed payments are not scaled with orders. SKU costs remain $8/$14/$22. C charges $49 and adds $0.50 sleeve to product cost. B/C shipping is $4.95 on regular baskets below $45 and free otherwise; the bundle ships free. Forecast quantities can be fractional expected demand; purchases are whole MOQ-compliant units.

The base uses management estimates; downside/upside combine supplied exploration bounds. These are joint stress tests, not estimated likelihoods or statistical bounds. All hold the non-bundle September basket mix constant, without elasticity, customer switching, or seasonality estimation. No evidence justifies treating management demand estimates as validated forecasts.

| Q4 operating profit | A | B | C |
|---|---:|---:|---:|
| Downside | −$8,441.02 | $8,876.45 | $8,876.45 |
| Base | $9,374.73 | $48,573.08 | $50,820.28 |
| Upside, **unconstrained demand** | $28,774.47 | $98,373.35 | $105,714.20 |

Base A has 7,200 orders; B/C 6,480. Contribution before marketing and overhead is $10.0520/order for A, $17.2181 for B, and $17.5649 for C. Q4 profit is orders × contribution − $63,000. Contribution includes revenue less product cost and delivery/fulfilment; it is not net profit or cash flow.

**Adversarial tests that matter:**

- At fixed base basket mix/costs, B matches A at 58.38% of A’s orders; this is a scenario threshold, not measured retention or a license to tolerate that decline.
- C exceeds B by only $2,247.20 and is overtaken if it loses a further 1.974% of orders relative to B, holding its assumed mix.
- Proportional substitution assumes bundles replace a representative slice of all baskets. Existing Pocket+Classic baskets were 182/2,200 orders, or 8.273%. If all of them switch first, and remaining bundles replace proportional other baskets, C profit becomes $46,442.89, **$2,130.19 below B**. Replacing a $55 existing pair with a $49 sleeve bundle loses $6.50 contribution.
- An independent agent recalibrated to Q3 rather than September basket mix: base A/B/C approximately $8,660/$47,881/$50,232. The pricing reset remains attractive, but this does not validate demand or bundle substitution.

Sources: `planning_assumptions.csv`, `Planning_Inputs.xlsx` Assumptions, September `orders.csv` and `order_items.csv`. Reproduction: `build_analysis.py`; detailed `scenarios.csv`, `forecastMonthly.csv`, `analysis_results.json`. Independent checks under `independent_audit/` and `forecast_challenge/`.

## 5. Inventory and illustrative cash

Opening stock: 2,500 Pocket, 1,800 Classic, 650 Bedtime; no inbound. Place a PO on hypothetical October 1 to receive November 15 after 45 calendar days. Minimum 500 units/SKU, increments of 100, half deposit and half at dispatch. Forecast is anchored to the case clock, not the actual date this assessment was prepared.

| Base case | A | B | C |
|---|---:|---:|---:|
| Pocket PO units | 2,100 | 1,600 | 2,000 |
| Classic PO units | 1,200 | 900 | 1,500 |
| Bedtime PO units | 500 | 500 | 500 |
| Inventory purchases | $44,600 | $36,400 | $48,000 |
| Initial 50% deposit | $22,300 | $18,200 | $24,000 |
| Illustrative minimum cash | $21,700 | $25,800 | $20,000 |

These purchases just cover modeled quarter demand with rounding and no safety stock. Before November 15, base demand fits opening inventory under even daily demand. MOQ creates excess ending Bedtime inventory, especially in lower-demand scenarios; a minimum rounded order is not necessarily the best commercial purchase plan.

**Illustrative cash schedule:** $65,000 opening; deposit October 1; assume dispatch/balance payment November 1 and receipt November 15 (dispatch date is not supplied); pay $21,000 marketing/overhead first of each month; collect revenue at each day’s end; pay shipping and sleeves daily. Model minimum includes cash before same-day receipts. Sold inventory cost is not paid again: ending cash = opening cash + operating profit + inventory COGS − inventory purchases. Sleeves are classified in product cost but paid daily and excluded from the inventory-COGS add-back.

Base ending cash is approximately $131,733 for A, $168,936 B, $167,203 C, reflecting the drawdown of already-paid opening inventory. This does **not** mean profit equals cash or that the business can sustainably repeat that working-capital release. No additional unknown receivables/payables, taxes, refunds, safety stock or settlement delays are modeled. Supplier dates, collections and intraday payments require confirmation. Prompt customer collections and uniform daily demand are assumptions.

Upside C requires $100,800 inventory purchases and a $50,400 deposit: $14,600 remains, already below reserve before operating payments. With first-day operating payments, the modeled pre-receipt minimum is −$6,400. Under even demand, Classic is exhausted around November 2 and Pocket around November 6. About 795 Classic and 670 Pocket units are needed before the November 15 receipt beyond opening stock. Therefore upside C profit and cash curves are **unconstrained opportunity figures**; actual sales cannot be assumed without earlier stock, phased procurement, better terms or funding. All upside offers breach the $15,000 reserve under the illustrative timing: A minimum $11,700, B $8,400, C −$6,400. Upside B also has pre-receipt shortages; inspect its SKU schedule before execution.

Sources: `inventory.csv`, `supplier_terms.csv`, `Planning_Inputs.xlsx`, `planning_assumptions.csv`. Schedules: `inventory.csv` and `cashDaily.csv` **inside Deliverables**, distinct from same-named source inventory CSV.

## 6. Pilot and scale decision

1. Randomize eligible visitors A versus B with stable assignments and comparable acquisition mix/spend. Record visitor denominator, which is absent from the case. Cap exposure according to available stock. Follow with C versus B after B’s response is understood.
2. Primary metric: contribution per eligible visitor, including discounts, collected shipping, product costs, delivery/fulfilment and sleeves. At comparable spend, assess total operating profit too. Guardrails: conversion, customer complaints/refunds, SKU availability, dispatch promises and at least $15,000 cash.
3. Start with two complete weeks, not a promise of statistical power. Use observed visitor conversion and contribution variance to predefine the sample size, minimum detectable effect and stopping rule; extend if evidence remains inconclusive. None of those denominators/variances is supplied.
4. Scale only when the pre-agreed lower confidence bound of the contribution-per-visitor improvement exceeds zero and stock/cash checks pass. Do not mistake positive modeled profit for validated uplift. Track full-price-pair displacement separately from bundle adoption.

## 7. Verification and evidence limitations

Independent checks found unique order and item keys, no orphan joins, no missing transaction fields, exact line/order/month reconciliation, correct policy equations and attribution totals, correct plan SKU/basket totals, and workbook cached values consistent with CSV counterparts. Source ZIP and extracted files byte-match. Review records provide the checks; no synthetic 'accuracy score' is asserted.

The original folder contains starting DOCX, a two-page learner PDF, a ZIP and an extracted folder with 17 source files: 14 CSVs, two workbooks and README. All seven workbook sheets were inspected, including formula expressions and caches. The promised local website/founder note is missing. We did not claim to inspect missing web pages or infer product quality from photos. Competitor evidence is the supplied frozen synthetic snapshot. Reconciliation across CSV/workbook copies is a consistency check, not proof that fictional observations represent a real business.

AI handled file inspection, reconciliation, scenario construction, challenge calculations and dashboard implementation. Human judgment supplied the decision posture and should approve demand, substitution, supplier timing and pilot criteria. The final recommendation deliberately favors a testable reset over automatically scaling the highest modeled figure.
