As you can see, The Inputs and the Outputs are in spaghetti format, so here is a link to the Organized Output File and the Inputs to check for yourself: (https://www.4shared.com/folder/f0vTS32e/_online.html) _______ Sales Decline Intelligence Report Egyptian DTC modest-fashion brand | Assessment as of 17 September 2026 1. Executive diagnosis The clearest problem is declining conversion and order volume, concentrated in abayas, with disproportionate weakness in returning-customer orders. Higher average order values partially conceal the deterioration. What declined: May–August revenue fell approximately 25%; orders fell 29.8%, against a 6.9% traffic decline. Where concentrated: Abayas, popular size combinations, checkout completion, and returning-customer orders. Dominant origin: Provisionally internal, with possible external amplification. Leading contributors: Worsening availability and mobile purchasing friction. Pricing and reduced purchase reassurance are additional supported hypotheses. Evidence confidence: High for the arithmetic pattern, conditional on consistent reporting; medium for causal attribution. Largest uncertainty: How much conversion deterioration comes from website friction, unavailable sizes, pricing, customer mix, and external demand respectively. Management implication: Correct identifiable purchasing obstacles and validate pricing before committing additional acquisition budget. A single dominant root cause cannot yet be established. However, the evidence is sufficient to justify targeted operational corrections without waiting for a complete market study. Evidence convention: Internal observations below are user-supplied evidence, not independently audited facts. External findings are cited. Causal explanations remain hypotheses unless directly demonstrated. 2. Data quality assessment Overall: ADEQUATE for provisional diagnosis; LIMITED for causal attribution and profitability decisions. The table reconciles closely: May: 2,505 orders × EGP 1,490 = EGP 3,732,450. August: 1,758 × EGP 1,590 = EGP 2,795,220. Orders divided by sessions reproduce the reported conversion rates, allowing for rounding. Five limitations matter: Issue Why it matters Channel scope is unclear The table uses website sessions but reproduces the headline revenue figures, despite 22% of sales coming through other channels. Confirm that orders and revenue belong to those website sessions. Revenue recognition is undefined Placed-order value, delivered sales, net sales after returns, and collected cash are different measures—especially with deteriorating COD performance. Several interventions overlap Pricing, redesign, availability, logistics, and return-policy changes cannot be separated using monthly totals alone. May is already affected May includes the redesign and May 18 price increase. It is not an untreated baseline. Seasonality and measurement changes remain unchecked Compare equivalent promotional and religious-calendar periods, and verify that the redesign did not change session or checkout-event tracking. May and August both contain 31 days, so their comparison is not distorted by month length. The August year-over-year comparison reduces simple calendar-month seasonality concerns, but does not control for assortment, promotions, customer mix, or moving holidays. Working assumption for the calculations: The performance table represents one consistently measured website population. If its orders include manually entered or marketplace orders, the conversion decomposition must be rebuilt. 3. Sales performance breakdown Metric May August Change Sessions 101,000 94,000 −6.9% Orders 2,505 1,758 −29.8% Conversion 2.48% 1.87% −0.61 percentage points; approximately −24.6% relative Average order value EGP 1,490 EGP 1,590 +6.7% Reconstructed revenue EGP 3.732M EGP 2.795M −25.1% August’s rounded revenue is also 21.8% below August 2025. Using: $$ Revenue = Sessions \times \frac{Orders}{Sessions} \times AOV $$ A sequential bridge, changing traffic first, conversion second, and AOV last, gives: Component Revenue movement May starting revenue EGP 3,732,450 Lower traffic, holding May conversion and AOV constant −EGP 258,685 Lower conversion, at August traffic and May AOV −EGP 854,345 Higher AOV, applied to August orders +EGP 175,800 August revenue EGP 2,795,220 Interpretation: Conversion is the largest observable component of the decline. These are accounting effects, not causal allocations. Their amounts depend on bridge order; they do not establish how much was caused by stockouts, pricing, or the website. A further implication: higher AOV does not prove the price increase succeeded. It measures spending among people who still purchased, excluding those deterred from buying. 4. Segment diagnosis Classification: MIXED, with pronounced concentration in abayas and returning-customer orders. Segment Observation Interpretation Abayas Revenue −31% YoY Strongest problem; coincides with largest price increases and size shortages Modest dresses −18% Suggests the problem extends beyond abayas Basic sets −5% Relative resilience; possible affordability effect, not demonstrated substitution Occasion wear Approximately flat Weakens a uniform demand-collapse explanation Mobile 92% of sessions Large exposure to the redesign; device-specific conversion remains missing Returning customers Order share fell from 30% to 23%, May–August Requires absolute-volume and cohort analysis If customer shares apply to the table’s orders: Order type May, approximate August, approximate Change Returning-customer orders 752 404 −46.2% New-customer orders 1,754 1,354 −22.8% Returning-customer orders therefore fell much faster. However, returning-order share is not a cohort-retention rate. Changes in cohort size, purchase timing, identification, and acquisition history could contribute. Category revenue weights are unavailable, so the exact monetary contribution of abayas cannot be calculated. Geography and channel concentration also remain unmeasured. 5. Internal factor analysis Availability — strong evidence of an operational constraint. Top-20 variant availability fell from 87% in May to 76% in August: an 11-percentage-point deterioration. Advertising products with unavailable common sizes creates a direct mismatch between acquired interest and purchasable inventory. However, average availability is not demand-weighted lost sales. Customers may substitute sizes, designs, or purchase later. Website and checkout — supported conversion hypothesis. The redesign coincided with slower mobile loading and weaker checkout completion. The relatively smaller add-to-cart deterioration directs attention toward the later funnel. The reported LCP deterioration is concerning, but average internal LCP cannot be compared directly with Google’s field-performance classification, which uses the 75th percentile. Google defines good LCP as ≤2.5 seconds and poor as >4 seconds. Google’s LCP documentation An extra checkout step alone does not establish causality. Inspect validation errors, payment failures, shipping-cost presentation, browser compatibility, and abandonment at each step. Pricing and offer — supported hypothesis, magnitude unknown. The example abaya increased from EGP 1,450 to EGP 1,620: 11.7%. Reduced discounts may have raised the effective paid price further. The category most exposed to pricing also suffered the largest decline, but stockouts confound the comparison. Price elasticity cannot be inferred from these aggregates. Retention — demonstrated order weakness, underlying cause unknown. Lower returning-customer orders and email revenue support investigation. A stable email list does not establish stable deliverability, engagement, campaign frequency, or reachable buyers. Fulfillment — additional deterioration after purchase. COD failures/cancellations increased by 4.3 percentage points, while average delivery time increased by 0.7 days, or 33%. These may reduce realized revenue and future purchases. They do not necessarily explain checkout conversion, and must not be deducted again if already excluded from reported revenue. Return policy — commercial and compliance concern. Egypt’s Consumer Protection Agency describes a general 14-day return/exchange right, subject to exceptions, and a separate 30-day right for defective goods. A blanket seven-day policy therefore appears inconsistent with the general rule for eligible purchases. Confirm applicable exceptions and correct the published policy and handling practices promptly. Egyptian Consumer Protection Agency The June change cannot explain deterioration that began before June, but could amplify it. Its conversion impact remains unmeasured. 6. External market intelligence Research finding: broad economic pressure is supported; a corresponding contraction in Egyptian online abaya demand is not established. Question Finding Evidential limit Are wider business conditions difficult? Egypt’s non-oil PMI remained below 50 in August Broad business survey, not apparel sales Did conditions worsen continuously through August? Reported PMI improved from 46.8 in July to 49.6 in August Slower deterioration does not mean recovery in sales levels Did competing abaya brands gain share? Not verified No reliable competitor sales or customer-switching evidence Did discount intensity increase before May? Not verified Current offers cannot establish historical changes Did customers move to marketplaces or social ordering? Not established Monthly channel results are missing Did abaya demand decline independently of this brand? Not established No sufficiently specific category series was verified The August PMI improvement is useful contradictory evidence against a simple story of continuously worsening macro conditions causing this brand’s continuously worsening results. It does not rule out household pressure. Egypt PMI series, reporting S&P Global data Research did not establish a defensible matched competitor price basket or historical promotional series. Consequently, competitor aggressiveness remains a possible cause, not a researched conclusion. A useful competitive audit must match fabric, construction, sizing availability, delivered price, delivery promise, and return conditions. Cheap listings alone would not demonstrate equivalent competition. 7. Macroeconomic impact Factor: consumer-price pressure Observed change: Reuters reported urban annual inflation increasing from 14.3% in June to 14.9% in July, citing CAPMAS. This is an observed release, unlike analyst forecasts. Reuters, 10 August 2026 Relevant customer: Urban middle- and upper-middle-income shoppers. Transmission: If household incomes fail to keep pace with essential expenses, discretionary clothing budgets become constrained. Expected effect: Lower purchase frequency, greater price resistance, postponed orders, or movement toward lower-priced options. Compatible internal evidence: Returning orders weakened; basic sets were relatively resilient; the brand increased prices. Contradictory/limiting evidence: Customer incomes and budgets are unknown; occasion wear remained flat; operational problems provide alternative explanations. Conclusion: POSSIBLE material amplifier, not a quantified cause. Inflation alone does not prove falling real income for this customer segment. Factor: broader demand weakness Observed change: Non-oil PMI remained contractionary but improved substantially in August. Transmission: Weak business activity could affect household earnings, confidence, and spending. Compatible evidence: Lower orders and repeat purchasing. Contradictory evidence: Brand performance worsened while the broader survey improved; website and inventory deterioration are business-specific. Conclusion: POSSIBLE background pressure; WEAK as the dominant explanation. Exchange rates, interest rates, and input costs are not assigned causal weight here: verified period-specific changes and business exposure sufficient to complete their transmission chains were not established. 8. Internal versus external diagnosis PRIMARILY INTERNAL — provisional, medium confidence. The classification rests on three observations: Conversion deteriorated much more than traffic. Specific business changes and operational constraints occurred during the deterioration. The weakest category overlaps with the greatest pricing and availability problems. This is not proof that the market was stable. External category evidence is insufficient, and macro pressure could amplify internal weaknesses. Reclassify as MIXED if matched customer or category evidence demonstrates significant external demand contraction. Reclassify toward PRIMARILY EXTERNAL only if conversion weakness persists in clean, in-stock, technically functioning segments and comparable businesses show similar deterioration. 9. Root-cause map Confidence below concerns causal contribution, not merely whether a metric changed. Cause Origin / evidence status Priority / confidence Supporting evidence Counterevidence or limitation Confirmation or rejection test Unavailable popular sizes Internal / SUPPORTED HYPOTHESIS High / Medium Declining availability; complaints; ads reaching constrained products Demand weighting and substitution unknown Compare matched variant performance by in-stock status; assess replenishment response Mobile checkout friction Internal / SUPPORTED HYPOTHESIS High / Medium Redesign timing; slower pages; checkout deterioration No device/step-level controls; overlapping changes Instrument funnel and test corrected flow on comparable traffic Price and discount changes Internal, potentially mixed / SUPPORTED HYPOTHESIS High / Medium Largest increases overlap weakest category Stockouts and redesign confound effect Test an offer on consistently stocked products with stable checkout Seven-day return policy Internal / POSSIBLE CAUSE of sales loss Critical for policy review / Low for sales causality Customer questions; reduced reassurance Began after initial weakening Correct eligible rights; track objections and completion; do not test a potentially unlawful restriction Delivery/COD deterioration Internal / SUPPORTED HYPOTHESIS for realized-sales loss High / Medium Longer delivery; higher COD failures Carrier mix, location and cancellation reasons unknown Matched carrier comparison, controlling for region and order type Reduced repeat purchasing Mixed/unknown / SUPPORTED HYPOTHESIS High / Medium Estimated returning orders −46%; email revenue down Symptom rather than identified root cause Equal-age cohorts, repeat intervals, customer interviews Acquisition cost/quality deterioration Mixed / POSSIBLE CAUSE beyond observed traffic loss Moderate / Low CPC and CAC increased CAC and ROAS also reflect downstream conversion Compare source-specific funnels, audiences and delivered-order economics Household budget pressure External / POSSIBLE CAUSE Moderate / Low Inflation evidence; compatible purchasing pattern No target-customer real-income evidence Lost-customer reasons and matched category data Controllability: Website, offer presentation, policy, and inventory allocation are directly controllable. Supply reliability, carrier performance, retention, and acquisition costs are partially controllable. Inflation and competitors’ actions are not directly controllable. 10. Causal chain Three defensible candidate chains emerge: Popular-size shortages → shoppers cannot buy their preferred variant → lower product conversion → fewer orders. Mobile redesign → slower or more difficult purchasing → lower checkout completion → fewer orders from broadly similar traffic. Higher effective prices + reduced return reassurance → greater perceived purchase cost/risk → possible abandonment or delayed repeat purchases. A fourth possible chain affects realized sales: Longer delivery → more cancellation/refusal → lower delivered revenue and potentially weaker future purchasing. The first two have the strongest business-specific support. Their contribution sizes remain unknown, and they must not be added to retention or advertising effects as if those were independent losses. 11. Potential false leads “Meta is the primary cause.” — RULED DOWN as a complete explanation. Lower website conversion can itself reduce ROAS and raise acquisition cost. Audience deterioration remains possible. “Traffic explains the decline.” — RULED DOWN. Traffic fell 6.9%, while orders fell 29.8%. “The economy explains everything.” — RULED DOWN. Broad pressure does not explain the specific checkout and availability problems. “Competitors took our customers.” — POSSIBLE CAUSE, currently unsubstantiated. Competitor presence is not evidence of switching. “Higher AOV proves stronger pricing power.” — RULED DOWN. It excludes non-buyers and does not measure contribution profit. “Returns barely increased, so experience is fine.” — RULED DOWN. COD failures increased, and a shorter return window changes comparability. “Discounting will fix it.” — NOT TESTABLE from current aggregates. Discounts cannot repair unavailable sizes or checkout errors and may reduce contribution. 12. Critical evidence gaps Request these three data packages first: Reconciled sales and funnel export: Daily January–August orders, sessions and funnel events by channel/device/source; placed versus delivered/net revenue; exact change dates. Variant economics and availability: Daily stock status, product views, prices, discounts, units, landed cost and variable fulfillment costs for leading products. Customer and delivery history: Anonymized customer IDs, order dates, first/repeat status, COD outcomes, carrier, delivery times, returns and cancellation reasons. Useful: Focused interviews with abandoned, lapsed and COD-cancelled customers; matched competitor offer histories; comparable 2025 trading-calendar data. Optional: Broad apparel market forecasts and exhaustive competitor lists. These are unlikely to alter the first operational decisions. 13. Management decision Decision currently justified: run a controlled operational correction and validation programme; defer broad pricing changes and advertising expansion. Why: Observable purchase obstacles are actionable even though their individual revenue effects are uncertain. Evidence: Conversion decline, checkout deterioration, variant shortages, complaints, and higher COD failures. Major assumption: Reporting is broadly comparable and sufficient demand remains for available products. Expected benefit: Recover preventable purchase failures and improve realized contribution from existing traffic. No defensible recovery percentage is available. Primary risk: Simultaneous changes obscure attribution; indiscriminate replenishment ties up cash; poorly tested rollback introduces new problems. What changes the decision: No meaningful improvement after verified operational fixes, combined with controlled evidence of price resistance or wider category weakness. Keep acquisition under a contribution-based spending ceiling, not an unconditional budget freeze. Redirect or pause spend reaching unsuitable inventory immediately. Avoid scaling based solely on platform ROAS. 14. Top three immediate priorities Priority Action and purpose Evidence basis Metric Success / failure signal 1. Repair the purchase path Verify mobile errors and test a corrected checkout; correct return-policy wording and handling where required Checkout deterioration and policy concerns Mobile checkout completion; error rate; realized contribution/session Improvement in comparable traffic without worse order quality; otherwise investigate remaining steps 2. Align inventory and advertising Prioritize proven size gaps; redirect ads to genuinely purchasable assortments Stock deterioration and unavailable-size complaints Demand-weighted availability; conversion on affected variants Availability and completed purchases improve; no purchase response weakens stockout attribution 3. Establish economic and customer truth Reconcile revenue, build equal-age cohorts, and separate COD losses Scope ambiguity; repeat-order decline; higher failures Net delivered contribution; cohort repeat rates; COD success Clear identification of loss locations; unresolved reconciliation blocks scaling For performance, restore comparable pre-redesign speed first; use field mobile p75 LCP ≤2.5 seconds as the longer-term technical target. That target is not a promised sales uplift. 15. Seven-day action and validation plan Timing Work Deliverable Days 1–2 Reconcile channel scope, revenue definitions and event tracking; map intervention dates; review return-policy compliance Trusted baseline and change log Days 2–3 Reproduce mobile checkout problems across major browsers/payment methods; inspect variant availability behind active ads Specific fault list and corrected ad destinations Days 3–4 Release verified fixes with monitoring; start one controlled checkout comparison if feasible Measurable intervention with rollback criteria Days 4–5 Commit selective replenishment; compare carriers within similar routes/order profiles Size-level replenishment and logistics decisions Days 5–6 Build customer cohorts and conduct a small, purposive set of customer interviews Ranked purchase barriers, explicitly not population estimates Day 7 Review immediate operational indicators and test integrity Continue, revise or stop each intervention At August traffic, approximately 21,000 sessions would arrive in seven days across the whole site. Product- and customer-level test samples will be smaller. Seven days is an operational checkpoint, not an automatic statistical verdict. Predefine the minimum commercially worthwhile effect and calculate sample needs. 16. Thirty-day business response Use targeted correction plus validation, with expansion conditional on results. Week 1: Complete the programme above. Week 2: Assess corrected checkout performance and replenished variants. Reconcile delivered revenue as orders mature. Week 3: If operational conditions are stable, test one pricing or offer hypothesis on selected consistently stocked products. Evaluate net contribution per eligible session, not conversion alone. Week 4: Review customer cohorts, carrier outcomes and mature experiment results. Scale only interventions supported by economics and operational capacity. For retention, contact relevant consenting customers about actual replenishments or improved service using a holdout where practical. Do not treat all lapsed customers as needing discounts. Decision branches: Operational correction improves outcomes: Extend it and cautiously increase acquisition within contribution limits. Offer test improves contribution: Apply selectively; do not automatically reduce all prices. Delivery performance remains weak: Reallocate suitable routes based on matched delivered-order results. Results remain weak despite clean execution: Broaden external-demand and customer-loss research before considering repositioning or major assortment changes. Some repeat-purchase and return outcomes will require more than 30 days. Keep those conclusions provisional. 17. Management brief Item Assessment Sales problem Approximately 25% revenue decline, driven principally by fewer converted orders Dominant origin Provisionally internal Primary drivers Availability and checkout friction lead; pricing and reassurance are additional hypotheses Confidence Medium on causality; high on conditional arithmetic Most controllable issue Purchase-path problems and advertising against unavailable sizes Largest external pressure Possible household budget pressure Macroeconomic impact Possible, not quantified Biggest unknown Relative contribution of stockouts, checkout, price and customer mix Immediate decision Correct observed obstacles, reconcile economics, run controlled validation Action to avoid Broad ad increases or blanket discounts before identifying profitable, fulfillable demand Next decision point Day 7 for operational evidence; days 14–30 for sufficiently mature commercial results