Recommend the right thing. Every time.
AI-powered product and content recommendations that learn what each visitor wants — and serve it across web, email, mobile, and ads in milliseconds.
1,500+ brands
Alpha X Mirrorless Camera
Pro-grade 4K video, 24MP sensor. The compact choice for creators.
The Future of AI: Neural Networks Hit New Milestone
Researchers have unveiled a new processing architecture that mimics the human brain's synaptic efficiency. This shift marks the end of traditional computing as we know it...
Tech Giants Merge
Cybersecurity Alert
Save 20% Today
Just Published
300+ Reads Today
Trending
Recommend products. Surface content. Convert both.
Whether you sell products or publish articles, Personyze adapts to every visitor in real time — lifting AOV for retailers and time-on-site for publishers.
Product Recommendations
Show every shopper the products they’re most likely to buy — on PDPs, category pages, cart, checkout, and post-purchase.
- Bestsellers, new arrivals, related, similar items
- Frequently bought together, complete-the-look
- AI-ranked by purchase intent in real time
Content Recommendations
Keep readers reading. Surface the next article, video, or guide that matches what they’re into — based on topic, behavior, and intent.
- Related articles, popular reads, editor’s picks
- Topic clustering & semantic similarity
- Cross-format recs (article → video → podcast)
Recommendation Badges
Layer dynamic badges over recommendations to drive urgency and trust — trending, low stock, social proof, price drops, new arrivals.
- “Trending” · “3 left” · “Just added”
- Bestseller · Editor’s pick · Top rated
- Real-time inventory & behavioral signals
Same Recs Everywhere
One recommendation engine, every channel. The same personalized picks follow your visitor from web to email to push notifications — coherent, never duplicated.
- Web, email, push, mobile app from one engine
- Open-time rendering for emails — never stale
- Frequency caps prevent same-product fatigue
A recommendation for every scenario.
Pick the right algorithm for the right moment — whether you're powering site widgets, open-time email recs, mobile, or your own API. Personyze handles the data and the math to put the right product in front of every visitor.
Every algorithm, page by page.
27 algorithms ship with Personyze — 24 for products, 20 for content. Pick the page you are building to see the ones made for it; every widget runs one, with a fallback and your own filters. How each family works is in our guide to recommendation algorithms.
Personalized & popular
Good defaults when the visitor’s intent is not clear yet.
Personalized Recommendations
Picks from the visitor’s profile and what similar visitors engaged with.
Most Popular
The most viewed items in the period you choose.
Trending Now
Items whose views are spiking above their weekly average.
Best Sellers
The highest purchase count.
Most-Added to Wishlist
The items most often saved to favorites.
Inspired by Past Purchases
What people who bought the same things went on to view and buy.
Related & cross-sell
Anchored to the product in view, the cart, or what the visitor just did.
Others Who Viewed Also Viewed
Co-viewed by other visitors.
What Others Viewed Then Bought
Bought by visitors who viewed the same items.
Cross-Sells
Co-purchased with the item in context, learned automatically.
Similar ItemsAI
Learned by AI from your catalog text, so a brand-new item has neighbours on day one.
Inspired by Category
More from the category in context.
Up-Sells
Higher-priced alternatives to the product or cart items.
Managed Cross-Sells
Pairs from a list you upload and maintain.
Cross-Sells for Last Item Added to Cart
Co-purchased with the latest item added to the cart.
The visitor’s own history
Best for returning visitors and post-purchase pages.
Recently Viewed
Products this visitor viewed recently.
Matches Their Search
Items matching the visitor’s last search.
Items in Cart
What is in the visitor’s cart right now.
Your Wishlist
The items this visitor saved.
Buy it Again
What this visitor purchased before.
Those Who Bought This Also Bought
Co-purchased with the visitor’s most recent purchase.
Catalog events
Triggered by changes in your catalog.
Price Dropped
Cheaper than at the visitor’s last visit.
Back in Stock
Returned to stock after selling out.
New in Stock
Added since the visitor’s last visit.
Your own rules
Any Products
Your own filter rules; Personyze ranks within them.
Personalized & popular
Good defaults when the reader’s interest is not clear yet.
Content Recommended for You
Articles this reader is likely to read, from similar readers and their profile.
Recommended — Published Since Last Visit
Published after the reader’s last session, picked for them.
Most Popular
The most read articles in the period you choose.
Trending Now
Reads spiking above their weekly average.
Most Liked Content
The articles most often liked or saved.
Most Commented
The most discussed articles in the period.
Co-reads & related
Anchored to the article being read, or the one read last.
Visitors Who Read This Also Read
Co-read by other readers.
Also Read — From Current Category
Co-read within the same topic or category.
What Readers Went On to Convert On
Content that readers of the same articles later converted on.
Similar ContentAI
Learned by AI from titles, topics, authors and descriptions — fresh content included.
More From This Topic
Articles that share the article’s topics.
Managed Related Articles
Related articles from a list you maintain.
The reader’s own history
Best for returning readers.
Recently Viewed
Articles this reader opened recently.
Matches Their Search
Articles matching the reader’s last search.
Content You Liked
Articles this reader liked or saved.
You Commented
Articles this reader commented on.
Reached Goal
Articles this reader converted on before.
Inspired by Reached Goals
What people who reached the same goals went on to read.
Catalog events
Triggered by new content.
Recently Published
Added since the reader’s last visit.
Your own rules
Any Articles
Your own filter rules; Personyze ranks within them.
Around every algorithm, the controls that matter.
The algorithm decides what is relevant. These decide what actually reaches the row.
Sort by what pays
Views, sales, cart adds, wishlist saves, price, rating — or profit, from your feed.
A fallback for every row
A second algorithm tops the row up, or steps in when the first finds nothing.
Your own filter rules
Narrow by category, price or any feed field; Personyze ranks inside what you allow.
In stock only
Sold-out items stay out of the row, without you touching the widget.
Up-sells in a price band
Pricier than the item in view, within the difference you set.
Categories and topics too
Recommend product categories, content categories or interests — not only items.
A feed, or a crawl
Import your product or content feed — or let Personyze crawl your site’s pages to build the catalog.
New visitors covered
Bestsellers, trending or new arrivals first — personal the moment an interest is known.
Go further: Recommendation examples →Recommendation templates →Measure recommendation ROI →
Choose the algorithm, then keep the row full
Two walkthroughs in the real panel — one widget built properly, with a filter, a fallback and a title for each algorithm, and Matches Their Search putting what a visitor searched for on the home page. Then do it with your own catalogue.
From product feed to live recommendation in five steps.
Set up once, then let the engine learn and improve. No data engineers, no ML team, no rebuilds — just measurable lift on day one.
Connect your catalog
Sync product or content data via Shopify, Magento, WooCommerce, BigCommerce, or any REST API. We handle inventory updates, pricing, attributes, and stock in real time.
Choose context & segment
Pick the page (PDP, category, cart, article, homepage) and the visitor segment (new, returning, high-value, mobile, etc). Or run for everyone.
Pick the algorithm or let AI decide
Collaborative filtering, content-based, frequently-bought-together, trending, recently-viewed, or full AI auto-blending. Each scenario has a recommended default.
Preview, simulate, then go live
Test recommendations in our Sandbox simulator with real visitor data. Share preview links with stakeholders. Roll out by traffic percentage when you’re ready.
Recommendations served in milliseconds
Every visitor gets the right products, articles, or content for them — computed in under 50ms and tracked back to revenue. The engine learns and improves automatically.
Test smarter. Win automatically.
Test generic recommendations against AI-personalized picks head-to-head. Auto-optimization promotes the winner the moment it hits significance.
Make every widget feel alive.
Layer dynamic badges and personal tags on top of any recommendation. Build trust with social proof, urgency with scarcity, and warmth with personal touches — all powered by your existing data.
Auto-tag items in the top X% by sales for any time window. No manual flagging.
"#1 in Cameras"
Surface items in the top X% by views or add-to-carts — before they peak in sales.
"Trending in your category"
Tag items added in the last X days. Threshold tunable per category or store-wide.
"Just dropped"
Surface real-time interest. Pull viewer counts, recent buyers, and ratings live.
"1,247 bought this week"
Pull live stock from your inventory feed. Show "only X left" without hardcoding.
"Selling fast in your size"
Auto-flag price-drops on items each visitor previously viewed. Add countdown timers.
"Ends in 02:14:33"
Drop dynamic tokens into any title. Pulled from CRM, geo-IP, or behavioral data.
"Popular in Visitor_City"
Track every click. Prove every dollar.
See exactly which recommendations drive revenue, by widget, page, algorithm, and segment. Attribution that ties recs to actual closed orders.
Everything you need to run recommendations.
From step-by-step onboarding to ROI tracking, every piece of the recs workflow lives in one platform — backed by a team that’s been doing this since 2008.
Book a strategy call →-
Step-by-step onboarding
Catalog connection, widget placement, algorithm selection, and first-go-live — all guided by a dedicated success manager. Most teams ship recs within 5 days.
-
Customizable widgets
20+ widget types — carousels, grids, stripes, popups, banners — all themeable to match your brand. No design or dev needed.
-
API & integrations
Native connectors for Shopify, Magento, WooCommerce, BigCommerce, Salesforce, HubSpot — or roll your own with our API and webhooks.
-
Revenue & ROI tracking
Every rec click tied to a real order. See revenue by widget, page, algorithm, and segment. Custom KPIs for whatever your team measures on.
-
Set and forget
Algorithms self-tune based on performance. The engine learns from clicks, conversions, and revenue and continuously improves — no manual retraining.
-
Flexible billing
True pay-as-you-go. Pricing scales with monthly visitors and active campaigns — not by catalog size or rec impressions. 30-day money-back guarantee.
Go deeper.
Playbooks, guides, and case studies on building recommendations that actually move revenue — for both e-commerce and publishers.
Product Recommendations Playbook
The end-to-end e-commerce guide: which widgets to put where, which algorithms to use, and how to measure lift.
Read playbook → PlaybookContent Recommendations Playbook
For publishers: how to keep readers reading with related, popular, and personalized article recs.
Read playbook → GuideOpen-Time Email Recs
How to send recommendations that compute at email open — not at send time. The difference can double CTR.
Read guide → ExamplesLive Recommendation Examples
Real Personyze customers and the recommendation widgets they run — with screenshots, configurations, and lift numbers.
View examples →Direct answers. No fluff.
The questions teams ask most when evaluating a recommendation engine. Got something else? Bring it to the call.
Browse the help center →Which algorithms do you support?
Collaborative filtering, content-based (semantic similarity), trending/bestsellers, recently-viewed, frequently-bought-together, complete-the-look, and full AI auto-blending that picks the best mix per visitor automatically. Each scenario has a sensible default; you can override per widget.
What platforms does Personyze work with?
Personyze is fully platform-agnostic. It works on top of any CMS, e-commerce platform, or web framework — including Shopify, Magento, BigCommerce, WooCommerce, Salesforce Commerce, WordPress, Webflow, custom React or Vue apps, and headless setups. Integration is a single JavaScript snippet, with no database changes or replatforming required.
For data, Personyze plugs into your existing martech stack: Google Analytics 4, Mixpanel, Segment, Amplitude, HubSpot, Salesforce, Marketo, ActiveCampaign, and any tool that exposes a REST API or supports webhooks. You can pull visitor signals via product feeds, CRM syncs, CDP integrations, or custom JavaScript — whichever fits your architecture.
How long until recommendations start performing?
Most teams see meaningful lift in the first 7–14 days. Trending and bestseller widgets work from day one. Personalized algorithms need a small history of clicks & conversions to tune themselves — usually about a week of normal traffic.
How do you handle new products with no history? (Cold-start)
New products are surfaced via content-based similarity (matching attributes, category, price band, description) until they accumulate enough behavioral data to enter the collaborative model. You can also boost them manually.
What’s the response time for serving recommendations?
Under 50ms p95 from a global edge CDN. Recs render before the page paints — no flicker, no lazy-load delay. Architected for high-traffic sites with millions of monthly visitors.
Can I A/B test recommendations?
Yes. Built-in A/B testing with traffic splits, real-time stats, and auto-promotion at statistical significance (95% confidence by default). Test algorithms, widget placements, copy, layouts, or rec counts. Auto-deploy the winner the moment significance hits.
Does it work with headless commerce?
Yes. Use our JS SDK, REST API, or GraphQL endpoint to fetch recs server-side or client-side. Works with Next.js, Remix, custom React/Vue/Svelte, native iOS/Android apps, and any backend stack.
How do you track revenue from recommendations?
Every rec impression and click is attributed to a real order via cart-level tracking. We tie back to actual closed orders, not just clicks — with last-click and assisted-revenue models. Custom KPIs supported.
Do you work for content sites without products?
Yes. Content recommendations are a first-class use case — related articles, popular reads, topic clusters, editor’s picks, “you might also like.” Common for publishers, knowledge bases, learning platforms, and media sites.
Can I customize the look of the widgets?
Fully. Widgets ship with 20+ templates you can theme to match your brand, or you can write custom HTML/CSS. Inherits your site fonts, colors, and spacing automatically. Use our visual editor or your own designer.
What about open-time email recommendations?
Yes. Recs compute at the moment the email is opened — not at send time. So if your visitor browsed something new between send and open, the email reflects it. Compatible with Mailchimp, Klaviyo, HubSpot, Salesforce Marketing Cloud, and any SMTP-based ESP.
GDPR, CCPA, privacy?
Personyze is GDPR-compliant and CCPA-compliant. Cookie-less recommendations are available for jurisdictions with stricter requirements. Anonymous-visitor recs are based on session behavior; identified-visitor recs respect whatever consent your CMP tracks.
How does pricing work?
Pricing scales with monthly visitors and active campaigns — not catalog size or rec impressions. Enterprise plans include dedicated CSM and white-glove onboarding. 30-day money-back guarantee, no annual lock-in.
Recommend the right thing. Every time.
AI-powered recommendations that learn what every visitor wants — and serve it across web, email, mobile, and ads in milliseconds.