Korean Cosmetics Firm Maps Kaspi.kz For Kazakhstan Market Entry | Actowiz Solutions

By Author: Actowiz Solution
Total Articles: 6
Comment this article

About the Client
Our client is a publicly listed South Korean cosmetics company preparing to enter the Kazakhstan beauty market through Kaspi.kz — the country's dominant e-commerce and marketplace platform. K-beauty carries strong appeal across Central Asia, but "we think it will sell" is not a strategy a listed company can take to its board. The team needed evidence: what actually sells on Kaspi.kz, at what price, from which brands, and which product attributes are trending — to decide what to source from Korea and how to position it.
The Challenge
An unfamiliar, single-platform market. Kaspi.kz dominates Kazakhstani e-commerce, but there was no off-the-shelf market report granular enough for category-level entry planning. The insight had to come from the platform's live data.
Depth across the whole hierarchy. The client needed a complete map from the top category (Beauty & Health) down through subcategories — Facial Care → Serums & Creams, Cleansing, and so on — with the brand and product landscape inside each.
Demand signals, not just listings. A catalogue alone doesn't reveal ...
... what sells. The team needed sales-volume and revenue proxies, review counts, ratings and customer-feedback signals to gauge real demand.
Sourcing intelligence. The strategic question was what to bring from Korea — so the data had to surface ingredient trends (Cica, Vitamin C, PDRN and others) and identify high-growth segments worth sourcing into.
Language & localization. Listings in Russian/Kazakh had to be extracted and structured so a Korea-based team could analyse them.
The Actowiz Solution
1. Full Category-Tree Extraction
We mapped the complete Beauty & Health hierarchy on Kaspi.kz down to the deepest subcategory, so the client could see the market's structure the way the platform organizes it — the foundation everything else hangs on.
2. Brand & Product Landscape per Subcategory
Within each subcategory, we extracted the brand set and product listings — titles, prices, variants, images and attributes — giving a clear view of who competes where, and at what price points K-beauty would be entering against.
3. Demand Proxies from Public Signals
Where direct sales figures aren't published, review count, rating and rating-velocity are the strongest public demand proxies. We captured these per product and rolled them up per brand and subcategory, turning a static catalogue into a demand-ranked landscape.
4. Ingredient & Attribute Trend Analysis
Product titles and descriptions were parsed for ingredient and claim attributes (Cica, Vitamin C, PDRN, retinol, and others), letting the client see which actives were proliferating and which subcategories were growing fastest — the core input to the sourcing decision.
5. Localized, Analysis-Ready Delivery
All fields were delivered structured, with original Russian/Kazakh text preserved alongside English for the Korea-based strategy team — plus a dashboard preview so stakeholders could explore before diving into the raw dataset.
Data Fields Delivered
Taxonomy data captured:
Full category-to-subcategory path
Product position within the category hierarchy
Brand & product details collected:
Brand
Product title (original and English)
Price
Product variants
Images
Product attributes
Demand signals tracked:
Product rating
Review count
Rating velocity
Seller/Merchant count
Attribute trends analyzed:
Parsed ingredients and claims (e.g., Cica, Vitamin C, PDRN)
Product type
Audit information recorded:
Capture date
Source reference
Subcategory context
The Results
Whole tree — the Beauty & Health category mapped end-to-end — the client saw the market's structure and its own entry points at a glance
Demand-ranked — brands and subcategories ordered by review-based demand proxies, replacing assumptions about "what sells in Central Asia" with platform evidence
Sourcing shortlist — high-growth ingredient segments (e.g., trending actives) surfaced directly from listing data — feeding the decision on which Korean lines to bring first
Board-ready — a listed-company strategy team got a defensible, data-backed entry case — not a hunch — before committing sourcing capital
Why It Worked
Structure first. Mapping the full category tree gave every later insight a place to live — entry planning starts with knowing the shape of the market.
Proxies where truth is hidden. No platform hands out sales figures; review signals are the honest, available stand-in for demand — and they ranked the opportunity credibly.
Answer the real decision. The client wasn't buying "data about Kaspi" — they were buying a sourcing shortlist. Framing the extraction around that decision is what made it useful.
Frequently Asked Questions
Can you cover marketplaces beyond Kaspi.kz?
Yes — Central Asian, GCC, SEA, LatAm and global marketplaces. The category-tree + demand-proxy + attribute-trend method transfers to any listing-based platform.
Can you estimate sales when the platform doesn't publish them?
We don't fabricate figures. We deliver robust public demand proxies — review counts, rating velocity, seller counts — which reliably rank relative demand across products and segments for entry decisions.
Is this a one-time dataset or ongoing?
Market-entry projects are usually one-time, but many clients convert to a recurring feed post-launch to monitor their own listings, competitors and price positioning.
Is marketplace data extraction compliant?
We collect only publicly displayed catalogue, price and review information — no accounts, no personal data. Collection follows Actowiz's responsible-scraping framework and applicable regulations.