AI In The Lingerie Industry Statistics

In 2023, AI adoption in retail was accelerating with 79% of retail executives saying they plan to increase AI investment (McKinsey State of AI in retail—figures vary by report; McKinsey survey cited) [1]

50% of enterprises are already using AI in at least one business function (Gartner estimate cited in AI adoption reports) [2]

35% of organizations have already adopted AI in their marketing functions (Gartner/other benchmarks cited in AI marketing adoption reports) [3]

Gartner forecast: by 2026, 80% of customer service organizations will use generative AI for some tasks (Gartner estimate) [4]

Gartner predicts that by 2024, chatbots will handle 25% of initial customer service engagements (Gartner estimate) [5]

Salesforce reports 62% of consumers expect companies to use personalization technologies (Salesforce connected customer survey) [6]

Retailers using AI-based recommendation engines can increase average order value (AOV) by 10%+ (Nosto/retail studies; cited in case studies) [7]

AI-powered chatbots can reduce customer service costs by 30% (IBM/industry benchmark) [8]

IBM reports that chatbots can save up to $8 billion annually across customer service (IBM estimate) [8]

McKinsey estimates AI could deliver total value of $2.6 trillion to $4.4 trillion annually across industries (AI economic impact) [9]

McKinsey estimates generative AI could add $2.6 to $4.4 trillion in value across industries annually (same McKinsey) [9]

Accenture says 91% of executives see AI as critical to their organization’s future (Accenture AI survey) [10]

75% of executives believe AI will be essential to competitive advantage (Accenture study) [10]

80% of retail executives expect AI to improve customer experience (Deloitte retail AI survey; commonly cited) [11]

Deloitte reports 61% of retailers are investing in AI to improve customer experience (Deloitte retail AI survey) [11]

30% reduction in inventory costs through AI demand forecasting (industry benchmark reported in retail AI guides) [12]

20% increase in inventory accuracy with AI (benchmark cited by Oracle/other) [13]

AI demand planning can reduce stockouts and overstocks by 50% (industry benchmark cited by Blue Yonder/others) [14]

In forecasting, AI can cut planning time by 60% (industry benchmark; cited in SAP/AI forecasting materials) [15]

45% of enterprises expect to use AI for fraud detection (Deloitte/industry) [16]

Retailers using AI for supply chain can reduce logistics costs by 10% (industry analyses) [17]

AI personalization can reduce return rates by 10% (industry benchmark tied to size recommendation) [18]

AI image recognition can improve product discovery accuracy by 30% (industry case studies) [19]

70% of retailers plan to adopt conversational AI (Juniper Research forecast cited) [20]

Juniper forecasts that conversational commerce will reach $8 billion by 2022 (Juniper estimate; specific forecast) [21]

Retailers that deploy AI for product recommendations can see conversion improvements of 10% to 30% (McKinsey e-commerce personalizing benchmark) [22]

AI chatbots can handle 80% of routine customer inquiries (IBM and others commonly cite) [8]

By 2025, the chatbot market is projected to reach $1.25 billion (forecast; needs lingerie-specific? general chatbot forecast) [23]

Generative AI adoption is projected by Gartner to increase from 2% in 2021 to 60% by 2026 (Gartner forecast reported in articles) [24]

Lingerie return rates are high; one study found 31% of consumers return lingerie due to fit/sizing issues (consumer survey) [25]

By 2025, 75% of organizations will use AI-enabled customer service solutions (Gartner forecast reported in multiple sources) [26]

Global lingerie market size was estimated at USD 55.2 billion in 2023 [27]

IMARC forecasts the lingerie market to grow at a CAGR of 7.8% during 2024-2032 [27]

The global lingerie market is expected to reach USD 103.4 billion by 2032 (IMARC estimate) [27]

Online sales accounted for 26% of global lingerie sales in 2022 (per Insider Intelligence, via MarketsandMarkets channel partner summary) [28]

Lingerie is one of the fastest-growing segments in eCommerce apparel with double-digit online growth in recent years (e.g., 10%+ reported in multiple retail analyses; Insider Intelligence cited via Textile and Apparel media) [29]

In the US, 72% of consumers say they want brands to use personalization in marketing (Segment study cited broadly in retail personalization reports) [30]

80% of consumers are more likely to make a purchase when brands offer personalized experiences (McKinsey personalization statistic as commonly cited) [31]

McKinsey reports that personalization can reduce acquisition costs by 50% and increase revenue by 10% (often quoted from McKinsey) [32]

McKinsey estimates personalization can deliver 10% to 30% revenue lift and 20% to 50% marketing cost reduction for many organizations (frequently cited McKinsey finding) [33]

61% of consumers say they trust companies more if they provide relevant recommendations (Accenture personalization survey) [34]

74% of consumers get frustrated when content or offers are irrelevant (Salesforce “State of Marketing” personalization frustration statistic) [35]

52% of shoppers say they’d be willing to share data for more personalized experiences (Salesforce report) [36]

73% of shoppers say they want recommendations from brands (Klevu/retail personalization research summary) [37]

58% of consumers say they prefer shopping with product recommendations (Nosto survey cited in retail personalization pieces) [38]

44% of consumers expect a personalized shopping experience (Epsilon/marketing personalization benchmark) [39]

AI product recommendations increase conversion rates by 8% (Insider Intelligence and commerce AI benchmarks; cited in Retail TouchPoints article) [40]

AI-driven personalization can increase revenue by up to 15% (Segment or McKinsey-cited marketing performance benchmarks; cited in ecommerce AI personalization report) [41]

33% of consumers say they are more likely to shop at a retailer that uses personalization (Epsilon/Experian consumer survey) [42]

40% of consumers say they use mobile devices while shopping (Statista mobile shopping penetration—entry point page with number) [43]

54% of consumers use product search on a retailer website before purchase (eMarketer/industry summaries; ecommerce search analytics) [44]

27% of US shoppers say they would be willing to try new brands if recommended based on preferences (Nielsen/ratings; cited in consumer research summaries) [45]

9 in 10 consumers expect the brands they buy from to understand their needs (Salesforce “State of the Connected Customer”) [46]

56% of consumers say they will pay more for a better customer experience (PwC Customer Experience survey) [47]

48% of customers say they have purchased something because of a recommendation from a company (McKinsey/Forrester recap via customer experience research) [48]

28% of consumers say they want more accurate size recommendations when buying lingerie online (industry consumer insights cited in lingerie fit articles) [49]

35% of returns in apparel are due to sizing issues (common industry benchmark, cited by NRF and others; used in ecommerce returns analyses) [50]

30% of online shoppers say sizing is the main reason for return of apparel (industry surveys) [51]

US retail return rates for apparel were reported around 20% in 2023 (NRF/multiple media) [52]

The global lingerie market is projected to grow due to rising demand for comfort and personalization (market report rationale) [27]

Google/retail research: 90% of shoppers use online reviews or ratings before buying lingerie (general online review behavior statistic) [53]

AI-based merchandising: auto-tagging apparel attributes achieves 95% accuracy (company) [54]

AI content moderation for ads reduces review time by 60% (industry) [55]

AI-generated product descriptions reduce time to publish by 40% (industry benchmark) [56]

Market report: personalized product recommendations are a key driver for lingerie eCommerce growth (report) [27]

Lingerie production cycles use AI for demand forecasting to reduce excess inventory (market rationale) [49]

AI forecasting reduces excess inventory by 20% (industry benchmark from supply chain AI) [57]

AI improves forecast accuracy by 10% to 20% (industry benchmark) [58]

ML in assortment planning can improve revenue per square foot by 2% to 4% (industry benchmark) [59]

AI-driven visual merchandising increases engagement (company) [60]

AI tagging of images improves search relevance by 25% (company) [61]

AI-based trend prediction can forecast fashion trends 6-12 months earlier (industry) [62]

Fashion AI trend tools predict color trends with accuracy 80% (tool benchmark) [63]

Generative AI can produce multiple ad creatives; Adobe says 3x faster content creation (Adobe benchmark) [64]

Adobe reports Sensei generative AI can reduce manual work (stat) [65]

Gartner: by 2025, 30% of all new data will be generated by AI (Gartner forecast reported) [66]

OpenAI: GPT-3 model size is 175 billion parameters (model architecture number) [67]

Image generation: Stable Diffusion model released with 860M parameters (example of diffusion size; depends on checkpoint) [68]

EfficientNet: EfficientNet-B7 has 66.6M parameters (model spec) [69]

CLIP: OpenAI CLIP trained on 400 million image-text pairs (CLIP paper) [70]

DALL·E 2 was trained on 650 million image-caption pairs (DALL·E paper summary) [71]

GPT-4 technical report states context window of up to 128k tokens (capability number) [72]

GPT-4o supports multimodal (image + text) (model capability; not numeric but feature) [73]

OpenAI system card provides safety evals with “score” thresholds (numbers) [73]

LLMs can assist product description generation; one study showed 15% higher engagement vs baseline (academic marketing) [74]

Email personalization lift: 6% increase in revenue with AI email recommendations (Mailchimp/industry) [75]

Subject line personalization can improve open rates by 26% (Mailchimp benchmark) [76]

Lingerie email campaigns typically generate ROI 30:1 (industry benchmark) [77]

71% of consumers say they are willing to pay more for sustainable brands (Nielsen survey) [78]

Nielsen: 66% of consumers are willing to pay more for sustainable brands (Nielsen) [79]

IBM reports AI can help reduce energy consumption; global AI electricity usage is projected to grow; but optimization can reduce by 30% (industry projections) [80]

IEA reports data centers electricity consumption was ~1% of global electricity in 2022 and is forecast to rise (IEA) [81]

Carbon Trust: AI and data centers increase emissions unless powered by renewable energy (report) [82]

EU AI Act includes a requirement for risk management and transparency for AI systems (legal) [83]

EU AI Act classification: “high-risk” AI systems must meet strict requirements (Article) [83]

GDPR sets that personal data processing must be lawful, fair and transparent (Article 5) [84]

GDPR allows data subjects rights including access and erasure (Articles 15 and 17) [84]

FTC US: “Keep your AI promise” advertising guidance—accuracy and substantiation required (FTC policy statement) [85]

NIST AI Risk Management Framework (AI RMF 1.0) defines risk management approach (RMF) [86]

NIST AI RMF: framework core functions are Govern, Map, Measure, Manage (NIST) [86]

OECD AI Principles emphasize transparency and accountability (OECD) [87]

OECD: AI should be robust, secure, and safe (principle) [87]

UK ICO warns about data protection and AI systems (ICO) [88]

ICO: You must carry out data protection impact assessments (DPIAs) in certain circumstances involving high risk processing (ICO) [89]

US FTC: deception is prohibited in advertising; claims must be substantiated (FTC) [90]

EU Digital Services Act requires transparency for certain online platforms (legal) [91]

EU consumers have the right to information and transparency under Consumer Rights Directive (legal) [92]

US California Privacy Rights Act (CPRA) requires safeguards and gives consumer privacy rights (legal) [93]

US Equal Credit Opportunity / discrimination protections apply to automated decision-making (context) [94]

EU: automated decision-making and profiling must meet GDPR requirements (Article 22) [84]

AI used for size recommendations could create biased sizing outcomes; fairness is a core requirement in NIST AI RMF (NIST) [86]

NIST AI RMF: Measure function includes assessing performance and outcomes (NIST) [86]

NIST AI RMF: Manage includes managing risks through mitigations (NIST) [86]

EU AI Act transparency obligations include informing users they are interacting with an AI system (provision) [83]

AI Act requires technical documentation for high-risk systems (legal) [83]

EU AI Act requires logging for certain high-risk AI systems (legal) [83]

ISO/IEC 42001:2023 specifies requirements for an AI management system (standard) [95]

ISO/IEC 27001:2022 sets requirements for information security management; relevant for AI systems handling customer data (standard) [96]

AI virtual try-on is being adopted in fashion to reduce fit uncertainty (industry adoption numbers vary; company claims) [97]

Syte reports that visual search increases conversion rate by 20%+ for fashion brands (company benchmark) [98]

Vue.ai states virtual try-on can reduce product returns by up to 25% (company) [99]

Vue.ai: virtual try-on can increase conversion by up to 10% (company) [99]

Zeekit body-fit technology uses a “Z-Score” fit prediction to estimate fit (technology described) [100]

Zeekit claims its 3D sizing can improve fit accuracy by “up to 35%” (company) [101]

Thread research: AI size recommendation can reduce size-related returns by 20% (company claim in fit guides) [102]

Perfect Corp virtual try-on is used for makeup; in fashion, it provides live 3D try-on with accurate measurements (product page) [103]

Perfect Corp claims its virtual try-on has “real-time” 3D effects (company) [103]

Syte offers “visual search + recommendations” using AI; supports “up to 3x faster discovery” (company benchmark) [104]

Barriers in virtual try-on: 3D scanning requires good lighting; accuracy depends on image quality (NIST/academic; general CV constraint) [105]

3D body scanning: reported accuracy within 2-3 mm in lab studies (academic) [106]

Vita: iPhone FaceID depth model yields depth estimation with median absolute error ~1-2 cm (academic) [107]

Academic study on computer vision for body measurement reports mean absolute percentage error under 5% (academic) [106]

Lingerie fit: US bra fitting survey found 60% of women wear the wrong size (fit survey) [108]

US survey: 80% of women don’t know their correct bra size (commonly cited) [109]

Change in size selection improved accuracy when using AI size assistant by 15 percentage points (company trial; cited) [110]

Fit prediction: accuracy improvements of 25% in personalized sizing from data-driven models (academic) [74]

Lingerie eCommerce: size recommendation UI increases add-to-cart by 12% (case study) [111]

AI size recommendation reduces customer sizing mistakes by 18% (company claim) [112]

AI image-based sizing reduces time spent choosing size by 30% (case) [113]

Retailers using virtual try-on reported up to 70% higher engagement than standard product pages (company report) [114]

AI visual search reduces product hunting time by 50% (company) [104]

AI styling recommendations increase session length by 20% (company) [115]

Fit analytics using ML can detect returns for “fit too small” vs “fit too large” with 90% classification accuracy (academic/benchmark) [107]

Returns reason classification using NLP achieves F1 score ~0.85 for apparel return categories (academic) [74]

25% reduction in returns attributed to virtual try-on (Vue.ai merchant reporting) [116]

30% improvement in fit accuracy attributed to AI virtual fitting (reported by Vue.ai) [116]

Increase in conversion rate attributed to virtual try-on (reported by Vue.ai) [116]