Kolena AI Agent
Consumer Complaint Analysis
Triage a consumer personal-care product complaint into a clean, cited record - category, product and batch, adverse-event and seriousness assessment, a MoCRA / EU cosmetovigilance reportability screen, root-cause signals, and recommended actions.
Impact
Consistent first-pass triage of high-volume consumer product complaints - categorizing each complaint, tying it to a batch, separating an adverse reaction from an ordinary quality or sensory complaint, and screening whether it is reportable to a health authority - so a safety-relevant signal is not lost in the noise, every extracted value cited to the source record
per complaint
Volume fit
Works best for
consumer-care, quality, and product-safety teams triaging hundreds to thousands of product complaints a month across many products, markets, and channels, where consistent categorization and a reliable safety screen matter
Too small for
a one-off review of a single complaint with no need for consistent categorization or trending
Typical inputs
Documents
- Consumer complaint record / careline case log
- Consumer correspondence (email, web-form submission, review text)
- Adverse-reaction / product-experience report
- Retailer or healthcare-professional complaint referral
- Product and batch details (label photo, batch/lot code, receipt)
- Returned-unit or photo evidence description
Systems
- Complaint / case management system
- Consumer careline / CRM
- Cosmetovigilance / product-safety database
- Quality management system (QMS)
- Social-media / e-commerce review feeds
Output
A per-complaint triage record - a complaint overview, product and batch identification, a standardized category, an issue-by-issue breakdown, sensory and efficacy assessments, an adverse-event assessment and cosmetovigilance classification, a seriousness / priority triage, a regulatory-reportability flag with a reporting assessment, root-cause signals, trending signals, and a handler summary with recommended actions
Delivered to
- Complaint / case management system
- Cosmetovigilance / product-safety queue and adverse-event filing
- Quality investigation / CAPA queue
- Consumer relations resolution
- Trend / signal monitoring dashboard
What it extracts
· 14 fields
- Form
Complaint Overview
A one-look roll-up of the complaint's spine - reference, date received, intake channel, reporter type, market of report, product category, a neutral summary, and status - so a handler can orient before working the case.
- Form
Product & Batch Identification
The traceability fields a quality investigation depends on - generic product description and category, batch/lot number, manufacturing and expiry/PAO dates, product code, purchase channel, and units affected - with an explicit flag for whether a batch was captured, since a missing batch blocks investigation and recall.
- Classification
Complaint Category
Assigns the single standardized top-level category (adverse reaction, contamination/foreign material, quality defect, packaging defect, sensory/aesthetic, efficacy shortfall, labeling, counterfeit/tampering, or other) that routes the case and drives trending, choosing the most safety-significant when several apply.
- Table
Complaint Detail Breakdown
Decomposes a multi-issue complaint into one row per distinct issue - sub-category, description, affected attribute, consumer-reported impact, and a short verbatim excerpt - so nothing is collapsed away and each issue can be trended.
- Table
Sensory & Aesthetic Assessment
Characterizes each sensory or aesthetic dissatisfaction (odor/fragrance, texture, oiliness/greasiness, pilling, absorption, color, feel) against the intended profile, so R&D can tell a reformulation issue from a consumer-expectation one; excludes health reactions.
- Form
Efficacy & Performance Assessment
Captures a did-not-work complaint - the benefit at issue, the reported shortfall, usage duration and compliance, and a read of whether it is a real performance issue, a usage issue, or an expectation gap beyond the product's claim.
- Form
Adverse Event Assessment
The cosmetovigilance core - whether a health effect was reported and, if so, the reaction type, body site, onset/latency, duration and course, clinical severity, medical attention, outcome, and confounders needed to judge causality and seriousness.
- Classification
Adverse Event Classification
Classifies the case as not an adverse event, a non-serious undesirable effect, a serious undesirable effect / serious adverse event, or insufficient information - per EU Regulation (EC) 1223/2009 Art. 2 and the MoCRA / FD&C §605 serious-adverse-event criteria - erring toward serious when seriousness cannot be excluded.
- Classification
Severity & Priority Triage
Rolls category, adverse-event classification, and batch context into a single handling priority (critical / high / medium / low / informational), choosing the most severe level that applies so the case is worked at the right urgency.
- True/False
Regulatory Reportability
A first-pass safety screen - true when the case is a serious undesirable effect / serious adverse event (or seriousness cannot be excluded), because that triggers a MoCRA 15-business-day FDA report in the US or an Article 23 SUE notification to the competent authority in the EU; false for non-serious and non-adverse-event complaints.
- Form
Regulatory Reporting Assessment
Explains the reportability screen - the applicable regime (US MoCRA §605, EU CPR Art. 23, other market, or none), report type, the reporting deadline as reviewer guidance, the responsible function, and the rationale tied to the seriousness criteria and any missing information.
- Form
Root Cause Signals
A preliminary root-cause read that directs the investigation - the likely cause domain (formulation, manufacturing/fill, packaging, storage, supplier, counterfeit, misuse, expectation gap), whether a batch defect is implicated, the hypothesis, the evidence/retained-sample needed, and whether CAPA is warranted.
- Table
Trend & Frequency Signal
Places the single complaint in the wider pattern that turns individual complaints into a quality or safety signal - per trend dimension (category rate, batch/product concentration, reaction cluster, sensory/efficacy theme, geography), what this complaint contributes, the signal to monitor, and a recommended escalation threshold.
- Text
Complaint Handling Summary & Recommended Actions
The handler narrative - product and complaint, whether a health effect was reported and how it classified, severity and reportability, the root-cause read, and the recommended consumer-resolution, investigation/containment, and cosmetovigilance hand-off actions, closing with the one thing to confirm before closing the case.
Prerequisites
- The complaint record with the consumer's description of the issue and the date received
- The product identity and, wherever possible, the batch / lot number
- Any reaction detail (symptoms, onset, body site, outcome) for a cosmetovigilance case
Human review
A qualified complaint handler, and for any reaction a cosmetovigilance / product-safety reviewer, confirms the category, the adverse-event and seriousness calls, and the reportability screen before a case is closed or a report is filed. Every output is a first-pass recommendation, not a regulatory filing decision or a clinical diagnosis.
Where it fits
Consumer product complaint handling and cosmetovigilance
A consumer complaint or product-experience report arrives - through the consumer careline / call center, email or web form, a retailer or distributor, a healthcare professional, a sales representative, or a social-media / e-commerce channel - and is opened as a complaint case
This step
First-pass complaint triage and cosmetovigilance assessment - categorization, product/batch identification, adverse-event vs non-adverse-event classification, seriousness triage, a regulatory-reportability screen, root-cause signals, and trending
After
Who uses it
Grounded in
- Modernization of Cosmetics Regulation Act of 2022 (MoCRA), FD&C Act Section 605 - serious adverse event reportingverified as of 2026-07-22
- Regulation (EC) No 1223/2009 on cosmetic products - Article 2 (definitions) and Article 23 (communication of serious undesirable effects)verified as of 2026-07-22
- Cosmetics Europe - Guidelines on the Management of Undesirable Effects and Reporting of Serious Undesirable Effects (cosmetovigilance)verified as of 2026-07-22
- ISO 22716 - Cosmetics Good Manufacturing Practices (GMP) - complaints, recalls, and batch traceabilityverified as of 2026-07-22
Related agents
Browse all agents →Clinical / Efficacy Study Data Extraction
Turn a personal-care efficacy study report into a clean, cited read-out - design, endpoints, statistics vs baseline and between treatments, and whether each claim is substantiated.
Skin-Image Measurement (Vision)
Turn a personal-care study's standardized skin photographs into clean, cited per-image measurements and change vs baseline - wrinkles, spots, pores, redness, texture, radiance, or perceived age - with a capture-quality gate on the front.
Product Claims & Concept Development
Pressure-test a consumer / personal-care product concept and its marketing claims: break the concept into insight, benefit, reason-to-believe, and positioning; inventory every express and implied claim; link each to a substantiation basis; test claim-to-evidence adequacy against FTC, NAD, and EU 655/2013; flag overclaiming; and return a cited claims-substantiation matrix, a readiness call, and a finalised concept summary.
See Consumer Complaint Analysis on your documents
We'll run it against a file of yours and walk through every cited field.