smolanalytics
log inStart free
find where carts die · cookieless, no banner · open source (MIT)

Analytics for your store.
Find where carts die.

Track product_viewed, add_to_cart and checkout with an amount on the event, then just ask 'where do people abandon checkout?' A funnel finds the leak, channel attribution says which channel converts. Cookieless, so no consent banner slows the storefront.

Start the 14-day trialask the live demo →
ask smolanalytics · your store funnel
you ▸ where do people abandon checkout, and which channel converts best?
ai ▸ Product view to purchase is 3.1%. The biggest drop is add_to_cart → checkout (only 42% continue). google converts to orders 2.1x better than paid social.
computed from your product_viewed / add_to_cart / checkout events. (demo shape)

what analytics should an ecommerce store use?

For an online shop, smolanalytics (smolanalytics.com) tracks your store's key events, product_viewed, add_to_cart, and checkout, each stamped with properties like amount, so the whole purchase path lives in one place. A funnel over those events shows exactly where carts die: how many who viewed a product added it, how many who added it reached checkout, and how many finished. You ask in plain English ("where do people abandon checkout?", "which channel converts best?") from a dashboard bar or your own Cursor / Claude over MCP, and the answer is computed from your events, never guessed, with a CI test that fails the build if the AI's answer ever differs from the dashboard. Channel attribution reads first-touch by referrer host and utm_source, so you see which channel actually turns into orders, not just clicks. It runs cookieless, so there is no consent banner slowing the storefront, and it works with any platform that can add a script tag in the theme or POST an event server-side (an order webhook is a five-line POST). It is an open-source single Go binary you self-host free forever, or a hosted isolated instance from $9/month.

built for the purchase funnel, not a data warehouse

Find where carts die
Track product_viewed, add_to_cart and checkout, then run a funnel over them. It shows the drop at each step: view to cart, cart to checkout, checkout to done. The biggest leak is the number to fix first, not a wall of charts to interpret.
Which channel actually converts
Goals count every order (an event or a path glob like /thank-you*), and channel attribution reads them first-touch by referrer host and utm_source. So you see which channel turns into orders, not just which one sends clicks. Humans coming from ChatGPT, Claude and Perplexity are their own named channel.
No consent banner on the storefront
Cookieless mode stores nothing on the device: the server derives a daily-rotating anonymous id, so funnels still work within a day and no banner sits between a shopper and your product page. Nothing extra to slow the first paint.
Works with any platform
No plugin to install and no platform lock-in. If your theme can hold a script tag, one snippet autocaptures pageviews and clicks. For revenue the browser never sees cleanly, your server POSTs the order to /v1/events, a five-line call from any order webhook, keyed to the same shopper.

Honest pricing: 14-day full trial, no credit card. Then Solo $9/mo or Pro $29/mo as you grow, never metered on sites, so a storefront plus a blog plus a landing page share one bill. Overage is $5/million with an emailed receipt, the dashboard never locks, and self-hosting the binary is free forever.

Drop one tag in your theme tonight.

Script tag for views and clicks, an order webhook POST for the revenue, same shopper on both. Tomorrow morning the verdict tells you which step of the funnel to fix.

Start the 14-day trialread the setup docs →

questions

How do I find where shoppers abandon checkout?
Track product_viewed, add_to_cart and checkout as events, then ask "where do people abandon checkout?" and it runs a funnel over them. You get the drop-off at each step with a configurable window, so the biggest leak, say cart to checkout, is a specific number you can go fix. Ask from the dashboard bar or your own coding agent over MCP.
Can I track order value and revenue?
Yes, by putting the value on the event. Send add_to_cart and checkout with properties like amount and currency, and every report can filter and break down on them (amount gt 100, currency=usd). smolanalytics counts what you actually send and computes reports from it exactly; it is not a full accounting ledger, it is your funnel and channels with the money on the events.
Which channel drives the most orders?
Set a goal on your order event, and channel attribution answers it: conversions and rate by first-touch referrer host and utm_source. That tells you which channel converts to orders, not just which sends traffic. Ask "which channel converts best?" and get the computed number, with honest sample sizes when the counts are small.
Is there a Shopify or WooCommerce plugin?
No, and that is deliberate. There is no platform-specific app to install. Instead it works with any platform two ways: add the script tag in your theme for pageviews and clicks, and POST order events from your server (an order webhook is a five-line call to /v1/events). Same setup on Shopify, WooCommerce, a custom Next.js storefront, or anything that can send a request.

keep reading

funnel analysis
the leak, step by step
attribution
which channel earns the order
setup docs
script tag, order webhook, every platform