Product recommendations on your site and in emails
Upload a product feed — the platform builds bestsellers, “also viewed”, “also bought”, similar items and personal picks for every customer.
Recommendation models
Models retrain daily on views, carts and purchases; recently purchased items are excluded. The catalog updates from your feed on schedule or via the API.

Website widgets without development
A recommendation block is set up in an editor with a live preview on your own products: card design, colors, fonts, a separate phone layout, display rules by page, device and segment, and up to four A/B variants compared by clicks and revenue. Copy the style from a screenshot of your website — AI fills in the fields. One line of code on the page; impressions and clicks are tracked automatically.

Personal blocks on your website
A banner or your own HTML anywhere on the page, with content chosen for the visitor: segment members get their own offer, everyone else gets the default content with A/B variants. The visitor’s name in the text, translations into your site’s languages, a live preview of every arm, and views, clicks and goal conversions within 7 days after the view. The block renders in isolation and never breaks your site’s styles.
- Up to 5 “segment → content” rows and up to 4 A/B variants
- On the Growth and Enterprise plans

In emails and journeys
Personal products are inserted into the email at send time — for abandoned carts, cross-sells and win-backs.

Tuned to your store’s industry
The industry tells the platform what matters in your catalog: clothing size, tire size and wheel bolt pattern, part number. Search understands “205/55 r16 winter” and “alloy wheels 5x112”, “similar” shows only compatible items, picks match the shopper’s size, and groceries and beauty get “time to reorder”. Claude suggests the industry of the project and catalog branches.

Smart catalog search
A ready-made search widget for any website — suggestions with products, categories and brands under the search box and a results page with filters and an “Add to cart” button, no development needed. Search fixes typos (“Did you mean”), understands price and brand in the query (“nike sneakers under 50”), finds products by meaning (“something warm for winter”) and learns from clicks and purchases every day. In WooCommerce, Magento 2 and 1C-Bitrix it is one setting.

AI search
Shoppers type a query in plain words — “waterproof running shoes under 100 euros, but not Nike”, “a gift for a runner” — and AI turns it into catalog filters in one call: categories, brands, attributes, price and exclusions. Only what exists in your catalog, nothing made up. The usual results appear at once, AI results replace them a second later, and above the products there is a short hint based on the product cards and refinement chips: “cheaper”, “for women”. Frequent queries are parsed overnight and answer instantly from the cache. Pay as you go from the balance; in WooCommerce, Magento 2 and 1C-Bitrix AI results appear on the store’s search page.
- When to use AI: for descriptive queries, on an “Ask AI” button or always
- Report: share of searches through AI, parsing speed, cache and the cost per 1,000 searches
- Proof of value: half of the visitors get the usual search, and the report compares clicks and purchases

Ask AI about a product
A block on the product page: shoppers ask “will this size fit?” or “does a 15-inch laptop fit?” and get an instant AI answer from the product description and attributes and your store knowledge base, in the language of the question. It never makes things up: without an answer it offers to ask a person in the chat, and the operator sees which product it is about. The report shows frequent questions, products that need a better description and purchases after a question. Pay as you go; one setting in WooCommerce, Magento 2 and 1C-Bitrix.

Product reviews
After a purchase the customer gets a link to a review page for the products they bought. Reviews are premoderated: AI flags spam, abuse and personal data, and clean 4–5 star reviews can be published right away. The product page gets a widget with the rating, the score distribution and schema.org markup for search engines; stars appear in recommendation widgets, emails and the AI chat; your store gets a reviews API. From the Growth plan.

- Profiles and eventsA single customer profile from website, app and CRM events.
- SegmentsAudiences by events and attributes — in real time.
- PredictionsRFM, churn risk, purchase probability, revenue forecast, best send time and channel.
- JourneysVisual scenarios: triggers, waits, branches and A/B.
- LoyaltyPoints, paying with points, tiers, friend referrals, promo codes and Wallet cards.
- Channels and campaignsEmail, web and mobile push, in-app messages, SMS, messengers and A/B-tested campaigns.
- Live chatWebsite chat, messengers and email in one inbox for your whole support team, with the customer profile alongside.
- Signup formsPopups and website forms: the contact goes straight into the profile with consent.
- AnalyticsReports, funnels, retention, revenue, traffic, channel attribution, ROPO and a control group.
- AI assistantsThe in-app assistant figures things out and builds campaigns; Claude and ChatGPT over MCP.
- White-labelThe platform under your brand and domain.
Try it on your own data
14 days on the Growth plan, no card required: up to 1,000 profiles and 500 emails a day.