RIntelligence
Menu
Index/case studies/limechat wow skin science

Proof grade B

WOW Skin Science

GPT bot handles off-hours, multilingual and error-prone WhatsApp support queries.

LimeChat dossier →
BProof grade
9Metrics retained
1Evidence records
LimeChatVendor

Reported outcomes

3 monthsGPT testing durationBefore production · LimeChatBaseline: This is implementation duration. · Formula: Dates and acceptance criteria are absent.
approximately 25000 messagesOff-hours messages handledWithin one month · LimeChatBaseline: Prior volume is not disclosed. · Formula: Deduplication and conversation boundaries are absent.
2 % of queriesQueries not understoodAfter GPT deployment · LimeChatBaseline: 16 · Formula: 16 / 2 = 8x lower; (2 - 16) / 16 = -87.5%
5-7 % of queriesUnseen queries handled autonomouslyAfter GPT deployment · LimeChatBaseline: 0 · Formula: Novel-query classification and accuracy review are absent.
5 % of messagesNon-English messages accurately handledAfter GPT deployment · LimeChatBaseline: Prior accurate handling is not disclosed. · Formula: Ground truth and review are absent.
84 % automationFully automated query rate — bodyAfter GPT deployment · LimeChatBaseline: 72 · Formula: 84 - 72 = 12 percentage points; relative increase = 16.7%
9.8 % increaseAutomation increase — cardAfter GPT deployment · LimeChatBaseline: 72 · Formula: Does not reconcile to 72% to 84%.
23 % better conversionConversion improvement — cardNo period is disclosed. · LimeChatBaseline: Prior conversion is not disclosed. · Formula: Event, denominator and body support are absent.
2 x better experienceExperience improvement — headlineNo period is disclosed. · LimeChatBaseline: Prior score is not disclosed. · Formula: Experience instrument is undefined.

Context and boundaries

Workflowcustomer_support_and_service–marketing_engagement_and_campaignsGPT bot handles off-hours, multilingual and error-prone WhatsApp support queries.
Implementation??
CaveatsCard and body automation claims conflict.–23% conversion and 2x experience lack body definitions.vendor_owned

Source trail

  1. WOW Skin Science delivers 2x Better Experience with LimeChat's GPT-Powered AI Bot

    Vendor case: GPT handles off-hours, free-text and multilingual queries after a three-month test. In month one it processed about 25,000 off-hours messages; unable-to-understand rate fell from 16% to 2% (8x lower), 5–7% of overall queries were newly autonomous, and 5% non-English plus 5% error-heavy messages were handled. Automation rose from 72% to 84% in the body, while a card says 9.8%; a 23% conversion claim and '2x better experience' headline lack definitions.

    company · accessed 2026-07-13
← Back to customer atlas