# Basketly

> Basketly is an operations hub for online shops that advertise on Meta. It keeps every order of a connected shop (Shoper, IdoSell, BaseLinker, WooCommerce, Shopify) with its lines and buyer, sends paid orders server-side as purchases to the Meta Conversions API and, when switched on, to Google Analytics 4, TikTok, Pinterest, ChatGPT Ads, Klaviyo, Google Ads and Microsoft Ads through durable queues (deduplication, SHA-256 hashing of email and phone, test mode), and shows a day-by-day purchase balance of shop orders against what each platform accepted. It optimises product feeds for Google Merchant Center and the Meta catalogue (automatic fixes, rules, flows on a canvas, a diagnosis of every product with a 0-100 score), draws ad images per product from templates, reports per-product results in Meta catalogue ads next to shop views, carts and sales, keeps buyers in one client list with purchase filters and segments that sync to Meta custom audiences, and adds a sales report (revenue, spend of every ad channel, blended MER, order sources), a campaign report for Meta, Google Ads, Microsoft Ads, TikTok, Pinterest and ChatGPT Ads that keeps each platform's own figures apart from the shop's orders, creative analytics, shop automations (paid order, customer stopped buying, shop alert, e-mail the team, SMS, webhooks), a self-check of the setup and multi-organisation workspaces for agencies. A platform accepting an event means it received and counted it; it does not by itself prove optimisation, match quality or attribution.

## Product (English)
- [Home](https://basketly.io/en): overview, how data flows, FAQ
- [Pricing](https://basketly.io/en/pricing): plans, limits, comparison
- [Pricing (markdown)](https://basketly.io/pricing.md): machine-readable plans and limits
- [Orders and buyers](https://basketly.io/en/features/orders-and-buyers): Every shop order with its lines and state, revenue, average basket, new and returning buyers, and 24 months of history.
- [Client list](https://basketly.io/en/features/crm): Every buyer as one person: shop purchases, history, purchase filters, owner, tasks and a data retention period.
- [Segments and audiences](https://basketly.io/en/features/segments-and-audiences): Ready-made and saved customer segments by purchases, and a segment as a Meta custom audience that updates every night.
- [Automations](https://basketly.io/en/features/automations): A paid order, a customer who stopped buying, a shop alert: e-mail the team, create a task, send a text or a webhook, built in a visual editor without code.
- [Product feeds](https://basketly.io/en/features/product-feeds): Your shop’s feed fixed automatically and by your rules, every product diagnosed against Google Merchant Center and Meta catalogue requirements, served at its own address.
- [Image templates](https://basketly.io/en/features/ad-image-templates): Images for carousel and catalogue ads drawn for each product: 122 ready designs, an editor and import from Figma.
- [Product statistics](https://basketly.io/en/features/product-statistics): Impressions, clicks and spend of every product in Meta catalogue ads, next to views, carts and sales in your shop.
- [Sales report](https://basketly.io/en/features/sales-report): One sentence on how the shop is doing, the six key figures and conclusions with a next step — and tabs below, each answering one question. Every number with its truth level: the shop, a platform or the last click.
- [All channels](https://basketly.io/en/features/multichannel-ad-reporting): Meta, Google Ads, Microsoft Ads, TikTok, Pinterest and ChatGPT Ads next to the shop’s sales: total spend, blended MER, each channel on its own campaigns and three truth levels that never mix.
- [Campaign reporting](https://basketly.io/en/features/campaign-reporting): Campaigns of Meta, Google Ads, Microsoft Ads, TikTok, Pinterest and ChatGPT Ads, each platform on its own figures and attribution window, next to the shop’s orders and revenue, with products in catalogue ads.
- [Creative analytics](https://basketly.io/en/features/creative-analytics): Cost per purchase and ROAS from Meta’s report, hook rate and a fatigue signal for every creative.
- [Conversions](https://basketly.io/en/features/conversions-meta-ga4-tiktok): Shop purchases sent server-side to Meta and seven more platforms, and a day-by-day balance: how many orders, how many purchases accepted and where something got lost.
- [Meta Conversions API](https://basketly.io/en/features/meta-conversions-api): Paid shop orders sent to Meta from the server: a durable queue, deduplication, SHA-256 hashing and a result for every delivery.
- [Integrations](https://basketly.io/en/features/integrations): Shop, ads, product catalogues, analytics and marketing in one catalogue, each integration’s state counted from what is actually connected.
- [Self-check](https://basketly.io/en/features/self-check): One check of your whole setup: the shop, purchases sent to Meta, logins, background jobs and automations, with a link to the fix for every problem.
- [For agencies](https://basketly.io/en/features/agencies): A separate workspace for every client shop, roles, an all-clients overview and a job queue for each.
- [Security and privacy](https://basketly.io/en/features/security): Encryption, hashing, roles, retention and self-hosting.
- [AI agents](https://basketly.io/en/features/ai-agents): Claude, ChatGPT and other assistants work with your shop’s data in Basketly: orders, products, feeds, campaigns and conversions.
- [Contact](https://basketly.io/en/contact): contact form (the team answers by e-mail) and help articles grouped by topic

## Produkt (polski)
- [Strona główna](https://basketly.io)
- [Cennik](https://basketly.io/cennik)
- [Rabat dla klientów BooksyAudit](https://basketly.io/cennik#booksyaudit): 30% taniej przez pierwsze 12 miesięcy na płatne plany
- [Zamówienia i kupujący](https://basketly.io/funkcje/zamowienia-i-kupujacy): Każde zamówienie ze sklepu z pozycjami i stanem, przychód, średni koszyk i kupujący nowi oraz powracający, 24 miesiące historii.
- [Kartoteka klientów](https://basketly.io/funkcje/kartoteka-klientow): Każdy kupujący jako jedna osoba: zakupy w sklepie, historia, filtry zakupowe, opiekun, zadania i okres przechowywania danych.
- [Segmenty i odbiorcy](https://basketly.io/funkcje/segmenty-i-grupy-odbiorcow): Gotowe i zapisane segmenty klientów według zakupów oraz segment jako niestandardowa grupa odbiorców w Mecie, aktualizowana co noc.
- [Automatyzacje](https://basketly.io/funkcje/automatyzacje): Opłacone zamówienie, klient, który przestał kupować, alarm sklepu: e-mail do zespołu, zadanie, SMS albo webhook, ułożone w edytorze bez kodu.
- [Feedy produktowe](https://basketly.io/funkcje/feedy-produktowe): Feed ze sklepu poprawiony automatycznie i regułami, z diagnozą każdego produktu według wymagań Google Merchant Center i katalogu Meta, pod własnym adresem.
- [Szablony grafik](https://basketly.io/funkcje/szablony-grafik): Grafiki do reklam karuzelowych i katalogowych rysowane osobno dla każdego produktu: 122 gotowe projekty, edytor i import z Figmy.
- [Statystyki produktów](https://basketly.io/funkcje/statystyki-produktow): Wyświetlenia, kliknięcia i wydatki każdego produktu w reklamach katalogowych Mety obok oglądań, koszyków i sprzedaży w sklepie.
- [Raport sprzedaży](https://basketly.io/funkcje/raport-sprzedazy): Jedno zdanie o tym, jak idzie sklep, sześć najważniejszych liczb i wnioski z następnym krokiem — a pod spodem zakładki, z których każda odpowiada na jedno pytanie. Każda liczba z poziomem prawdy: sklep, platforma albo ostatnie kliknięcie.
- [Wszystkie kanały](https://basketly.io/funkcje/wszystkie-kanaly-reklamowe): Meta, Google Ads, Microsoft Ads, TikTok, Pinterest i ChatGPT Ads obok sprzedaży sklepu: wydatki łącznie, MER łączny, każdy kanał na swoich kampaniach i trzy poziomy prawdy, które nigdy się nie mieszają.
- [Raport kampanii](https://basketly.io/funkcje/kampanie): Kampanie Mety, Google Ads, Microsoft Ads, TikToka, Pinteresta i ChatGPT Ads, każda platforma na swoich liczbach i w swoim oknie atrybucji, obok zamówień i przychodu sklepu, z produktami w reklamach katalogowych.
- [Analiza kreacji](https://basketly.io/funkcje/analiza-kreacji): Koszt zakupu i ROAS z raportu Mety, hook rate i sygnał zmęczenia dla każdej kreacji.
- [Konwersje](https://basketly.io/funkcje/konwersje-meta-ga4-tiktok): Zakupy ze sklepu wysyłane z serwera do Mety i siedmiu innych platform oraz bilans dzień po dniu: ile zamówień, ile zakupów przyjętych i gdzie coś się zgubiło.
- [Meta Conversions API](https://basketly.io/funkcje/meta-conversions-api): Opłacone zamówienia ze sklepu wysyłane do Mety z serwera: trwała kolejka, deduplikacja, hashowanie SHA-256 i wynik każdej wysyłki.
- [Integracje](https://basketly.io/funkcje/integracje): Sklep, reklamy, katalogi produktów, analityka i marketing w jednym katalogu, ze stanem każdej integracji liczonym z tego, co naprawdę jest podłączone.
- [Autodiagnostyka](https://basketly.io/funkcje/autodiagnostyka): Jedno sprawdzenie całej konfiguracji: sklep, wysyłka zakupów do Mety, logowania, zadania w tle i automatyzacje, z linkiem do poprawki przy każdym problemie.
- [Dla agencji](https://basketly.io/funkcje/dla-agencji): Osobna przestrzeń dla każdego sklepu klienta, role, widok wszystkich klientów i kolejka zadań każdego z nich.
- [Bezpieczeństwo i RODO](https://basketly.io/funkcje/bezpieczenstwo): Szyfrowanie, hashowanie, role, okres przechowywania i instalacja na własnym serwerze.
- [Agenci AI](https://basketly.io/funkcje/agenci-ai): Claude, ChatGPT i inni asystenci pracują na danych sklepu w Basketly: zamówienia, produkty, feedy, kampanie i konwersje.
- [Kontakt](https://basketly.io/kontakt): formularz kontaktowy (odpowiadamy e-mailem) i artykuły z pomocą według tematów

## Videos (20-second clips from the panel, demo data, no sound)
- [For agencies](https://basketly.io/en/video/for-agencies) (en): Each client is its own organisation, roles go down to “only own”, and the all-clients overview puts every client’s results side by side.
- [Dla agencji](https://basketly.io/wideo/dla-agencji) (pl): Każdy klient to osobna organizacja, role sięgają do „tylko własne”, a zestawienie „Wszystkie” pokazuje wyniki klientów obok siebie.

## Developers
- [API documentation](https://basketly.io/en/docs/api) (en): The Basketly REST API: organisation keys, orders, products and feeds, clients and segments, server-to-server conversions and the purchase balance, sales and campaign reports, background jobs and signed webhooks. Examples in curl, JavaScript, Python and PHP.
- [Dokumentacja API](https://basketly.io/docs/api) (pl): REST API Basketly: klucze organizacji, zamówienia, produkty i feedy, klienci i segmenty, konwersje serwer–serwer i bilans zakupów, raporty sprzedaży i kampanii, zadania w tle i podpisane webhooki. Przykłady w curl, JavaScript, Pythonie i PHP.
- [OpenAPI 3.1 spec](https://app.basketly.io/api/v1/openapi.json): machine-readable description of every endpoint and webhook
- [API reference in Markdown](https://basketly.io/docs/api.md): every endpoint, field, scope, error and webhook in one plain-text file (Polish: https://basketly.io/docs/api.pl.md)
- [MCP server in Markdown](https://basketly.io/docs/mcp.md): connecting AI agents (API key or OAuth), every tool with its scope, rules (Polish: https://basketly.io/docs/mcp.pl.md)
- REST API base URL: https://app.basketly.io/api/v1; authentication: `Authorization: Bearer bsk_live_…` (organisation key from the panel); 120 requests per minute per key.
- MCP server for AI agents (Claude, ChatGPT, Cursor, n8n): https://app.basketly.io/mcp with the same organisation key or through OAuth 2.1 "Connect" (claude.ai, ChatGPT custom connectors; the agent then acts as the person who connected it, with their role); tools for orders, segments, integrations, getting started, the purchase balance, products and their statistics, feeds and flows, image templates, the sales, channel, campaign and creative reports, products × channels, automations, clients, tasks and the self-check; agents can fix products, create and change feeds, add notes, statuses, tasks and costs outside ad platforms, but never send conversions.
- Plans: the REST API (keys, webhooks) and the MCP server are included from the Growth plan up (Growth, Agency, Custom). On Free and Start every account owner gets a 5-day trial from creating the account; after it the API answers HTTP 402 with error code `plan_required`.

## Legal
- [Privacy policy](https://basketly.io/en/privacy) (en): What data Basketly processes, why, where it is stored, for how long and what your rights are, including Google and Meta data.
- [Polityka prywatności](https://basketly.io/polityka-prywatnosci) (pl): Jakie dane przetwarza Basketly, po co, gdzie je przechowujemy, jak długo i jakie masz prawa. Także dane z Google i Meta.
- [Terms of service](https://basketly.io/en/terms) (en): The rules for using Basketly: account, plans, what each side does, liability and ending the agreement.
- [Regulamin](https://basketly.io/regulamin) (pl): Zasady korzystania z Basketly: konto, plany, obowiązki stron, odpowiedzialność i zakończenie umowy.
- [Data deletion](https://basketly.io/en/data-deletion) (en): How to delete your data from Basketly: people, Meta and Google connections, SMS accounts, an organisation or the whole account. Including Facebook data.
- [Usuwanie danych](https://basketly.io/usuwanie-danych) (pl): Jak usunąć swoje dane z Basketly: osoby, połączenia z Meta i Google, konta SMS, organizację i całe konto. Także dane z Facebooka.

## Key facts
- Shops: Shoper, IdoSell, BaseLinker, WooCommerce and Shopify, each connected with its own API credentials (no developer). Orders are read every 15 minutes, and by webhook where the shop offers one; once per connection a background job reads the last 24 months of orders (nothing from it is sent to Meta). Products are read on connect, nightly and after product webhooks. Other sources: Google Sheets (synced about every minute), a trusted server endpoint and the REST API.
- Getting started: a checklist for a new organisation read from its own records (shop connected, Meta receiver, a running purchase flow, a product feed, a Meta ad account read, optionally other platforms), each step with a button to the exact place where it is done; it opens by itself while required steps are left.
- Orders page: every order kept whether paid or not (bought, unpaid, cancelled), with lines (name, variant, code, EAN, quantity, price), shipping and the buyer; tiles for revenue per currency, bought orders, average basket and new versus returning buyers. The buyer becomes a person in Clients, matched by phone, then e-mail. Each order shows its source — the channel (Meta, Google Ads, Microsoft Ads, TikTok, Pinterest, ChatGPT Ads, e-mail, price comparison sites, other, direct visit, unknown), campaign, ad set and ad, UTM and click time of the last ad click in the buyer's visits (tied to the order by the order's tracking_id or linkOrder() on the thank-you page; the product's own reading, not a platform's attribution) —, with filters by platform and campaign; Campaigns shows these orders as a separate count next to Meta's purchases, never added up.
- Conversions: a paid order becomes a Purchase for Meta (value, currency, catalogue product ids, hashed e-mail and phone, IP; fbp/fbc when the consent-gated shop script provides them). With "Send purchases" switched on, the same purchase also goes to GA4 (Measurement Protocol event purchase, order id as transaction_id, events older than 72 hours are refused), TikTok (Events API CompletePayment with the same event id as the pixel and the visit's ttclid and _ttp), Pinterest (Conversions API checkout with the same event id as the Pinterest tag and the visit's click id), ChatGPT Ads (OpenAI Conversions API order_created with the same event id as the OpenAI pixel, amount in minor units, order lines, hashed buyer data, and the visit's oppref and obref when the consent-gated shop script reads them; every paid order is sent, not only those after a click), Klaviyo (a Placed Order event by e-mail, only purchases with an e-mail, subscribing nobody), Google Ads (a click conversion to the chosen conversion action, only purchases from a Google ad click (gclid, gbraid or wbraid) or, without a click, with the hashed e-mail when the account has enhanced conversions for leads on) and Microsoft Ads (an offline conversion to the chosen goal, only purchases from a Microsoft ad click (msclkid)) — eight platforms in all, each in its own queue with retries, once per event id. The purchase balance (7/14/30 days) compares shop orders, purchases accepted by the product and each destination (accepted, errors, waiting) day by day, with the five most common error reasons.
- Integrations page: a catalogue by purpose (where you sell, where you advertise, where the catalogue goes, analytics and consent, marketing, developers) with each integration's state counted from existing connections; key-based connections for GA4, TikTok, Pinterest, ChatGPT Ads (Pixel ID and Conversions API key, optionally an Ads API key; "Save and check" sends one validate_only test event that OpenAI does not store) and Klaviyo (up to 5 each for GA4, TikTok, Pinterest and Klaviyo, keys sealed and never shown again); TikTok, Pinterest, ChatGPT Ads and Klaviyo can also read campaign results after the "Read campaign results" switch is turned on (90 days at once, then every 6 hours; a token without reporting permission is refused with a readable reason).
- Product feeds: input from an address (every 1/3/6/12/24 hours), an uploaded file or Shoper products; Google XML, Ceneo XML, CSV/TSV, up to 300 MB and 200,000 products. Automatic fixes (HTML to text, capitals, sales phrases, length limits, availability, invalid GTINs, sale prices, price format; nothing missing is made up, the id is never changed), Google categories picked automatically for every product (from the shop category and product names, with a confidence; a person can confirm or change them), up to 200 rules per feed, flows on a canvas, feeds made from feeds; output at its own secret address as Google/Meta XML, Ceneo XML or CSV, rebuilt on every request. Up to 20 feeds per organisation.
- Products page: 44 checks from Google Merchant Center and Meta catalogue requirements (error -20, warning -6, tip -2, info not scored), a 0-100 score, the product record with input versus output per feed, what each fix or rule changed, history and manual fixes that survive the next read.
- Product statistics (7/14/30/90 days): impressions, clicks and spend per product in Meta catalogue ads (read at most every 6 hours), views, adds to cart and checkout starts from the consent-gated shop script (daily counters without any person), orders and revenue from the shop; exposure tiles (not shown, under 100 impressions, shown) and a one-click flow feed of products without exposure. Meta gives no purchases per product, so product sales come from the shop and are not attributed.
- Image templates: 122 ready designs in four formats (1080x1080, 1080x1350, 1080x1920, 1200x628), an editor with layers, a library of elements and conditions, import from Figma; a feed rule swaps the product image for the drawn one; results per template from catalogue ads over 30 days. Up to 50 templates per organisation.
- Sales report (7/30/90 days or a chosen range, against the period before; tabs named after the questions they answer — Overview, Ad channels, Products with Trends and In channels, Lead to purchases, Visibility in the Meta catalogue, Stock levels): the Overview opens with a one-sentence verdict of the period (revenue change, spend change, MER and its change, figures toned good/bad), six key figures and the conclusions, every number marked with its truth level (shop, platform by its attribution window, last click); revenue, paid orders, average order and new/returning buyers from the shop; total spend of every ad channel (Meta, Google Ads, Microsoft Ads, TikTok, Pinterest, ChatGPT Ads) plus costs outside ad platforms typed in by hand (comparison sites, e-mail tool, agency), blended MER (shop revenue / all spend, no attribution), cost per order and orders by source (Basketly's last-click reading, one source per order); a Channels table with three truth levels never mixed — each platform's own conversions, value and ROAS within its named attribution window (never added across platforms), Basketly's last-click orders per channel and the channel's MER against the shop — and the sum of platform claims next to the shop's orders; spend in other currencies is counted apart unless the optional NBP-rate conversion is switched on; an In channels view of the Products tab (every product against every channel that shows it); rule-based conclusions including channel rules with a channel filter; purchases accepted for sending; shop revenue day by day and spend per channel day by day on separate charts, MER week by week; top 10 products; orders from tagged feed links per feed and landing product (Basketly’s reading from the shop script, not Meta purchases). Stock levels tab: per product with stock from the shop, sales pace of the last 14 days, days of cover, linear stockout date, reorder point, suggested quantity and a level against the organisation’s lead time (7 days by default), with ad and cart demand week over week; stock as last read plus a labelled estimate. Optional e-mail to the owner and admins every Monday with each channel's spend and the blended MER, and a stock section.
- Campaign report: a channel switcher — "All channels" (total spend, blended MER, cost per order, orders by source, sum of platform claims, the Channels table) and one tab per platform: Meta (spend per campaign, ad set and ad with purchases and purchase value as Meta reports them, one purchase type per row, never summed, cost per purchase and ROAS, and for the whole account shop revenue and MER; "Show products" lists the catalogue products of a campaign or ad with their shop sales), Google Ads and Microsoft Ads (campaigns with type, ad groups, ads and per-product results), TikTok, Pinterest and ChatGPT Ads (campaign, ad group and ad once results reading is on). Each platform's attribution window is read where the platform exposes it and printed in days.
- Clients: buyers as one person per phone or e-mail, with shop purchases on the card, purchase filters (orders, amount spent, last or first purchase, lapsed for 30-365 days, bought product), owners, internal statuses, tasks and a retention period of 6-120 months (default 24). Segments: five ready-made (new buyers, bought once, returning, lapsed 90 days, no purchases) plus up to 50 saved; a segment can sync to a Meta custom audience (hashed e-mail and phone, only the difference sent, nightly and after a filter change, owner or admin after confirming the legal basis).
- Automations: triggers for a paid order (every, first or next, optionally from a value), a customer who stopped buying (30/60/90/180/365 days) and a shop alert (purchases refused by Meta or any other platform, paid orders without accepted purchases, a feed that failed), plus client-list triggers and inbound webhooks; actions e-mail the team (owner and admins or up to 5 addresses, never shop customers), SMS (SMSAPI, Twilio, an HTTP gateway or the user's own Android phone through smsportal.app; only 08:00-20:00, at most 500 a day, never to numbers on the "Do not text" list), signed webhooks, tasks and owners.
- Creative analytics: channel tabs for Google Ads (ads plus assets: headlines, descriptions, images, YouTube videos), Microsoft Ads, TikTok (video covers), Pinterest and ChatGPT Ads (chat card title and text, no image) list their ads with the platform's figures; for Meta ads, ads reusing the same video or image are grouped into one creative; the main metric is cost per purchase as Meta reports it; spend, CTR, hook rate, hold rate, purchases and purchase ROAS; signals "Possible fatigue" and "Hook rate drop" (to check, not proof of cause); Clusters can create a missing material and copy combination as a new ad, always paused, the only write to Meta.
- Self-check: accounts, sending to Meta, conversions, other sources (including shops with no new orders for 7 days), background jobs, automations and campaign data; runs on "Check now" or 2 minutes after a configuration change, never on a timer; a Meta read that gets no answer is "not checked", never an error; some things repair themselves once.
- Data: secrets, tokens and payloads are encrypted with AES-256-GCM. Roles: built-in owner, admin, operator and viewer, plus custom roles that give each area no access, view or edit, with "only own" for clients and tasks; every check runs on the server. Unlimited organisations per account, data isolated per organisation. Single Go binary, can be self-hosted (Custom plan).
- Plans count orders: the first purchase of each order per calendar month from a shop, Google Sheets or the API, across all organisations the billing owner owns.
- Partner discount: BooksyAudit customers (https://booksyaudit.pl, Booksy profile audits for Polish beauty salons): a customer with a paid BooksyAudit order activates Basketly from the BooksyAudit panel with one click ("Aktywuj Basketly z rabatem 30%"); any paid plan (Start, Growth, Agency) chosen within 365 days of activation is 30% cheaper for the first 12 months, monthly or annual, applied automatically at checkout. The Free plan stays free.
- Help inside the panel: every page has a step-by-step tutorial that highlights its parts (offered once on the first visit, then in the help menu), the help dialog lists guide articles and 20-second clips for the open page, and "Write to us" sends a message to the team from the panel (the reply comes by e-mail). ⌘K / Ctrl+K searches pages, people, products and campaigns; irreversible actions (disconnect, delete, ownership transfer, shorter retention) ask for confirmation naming the object and the effect.
- Contact: kontakt@basketly.io, or the contact form at https://basketly.io/en/contact

## Guides
- [Meta Conversions API: co to jest, czym różni się od Pixela i czego nie zrobi](https://basketly.io/blog/meta-conversions-api-co-to-jest) (pl, 2026-10-25): Czym jest Meta Conversions API, czym różni się od Pixela, jakie dane przyjmuje, jak działa deduplikacja i czego CAPI nie gwarantuje. Przewodnik bez obietnic.
- [Automatyzacje sklepu: opłacone zamówienie, klient, który przestał kupować, i alarm, gdy zakupy nie dochodzą](https://basketly.io/blog/automatyzacje-sklepu-zamowienie-uspiony-klient-alarm) (pl, 2026-10-26): Trzy wyzwalacze sklepu w Basketly i akcja „Wyślij e-mail do zespołu”: kiedy startują, czego pilnują, gotowe szablony i ograniczenia, które warto znać przed włączeniem.
- [Bilans zakupów: jak sprawdzić, czy zamówienia ze sklepu naprawdę docierają do Mety i innych platform](https://basketly.io/blog/bilans-zakupow-czy-zakupy-dochodza-do-mety) (pl, 2026-10-26): Jak czytać bilans zakupów dzień po dniu: zamówienia w sklepie, zakupy przyjęte przez Basketly i wysyłki do Mety oraz innych platform. Co znaczy każda różnica.
- [Integracje sklepu: od czego zacząć, żeby zamówienia, reklamy i katalog działały razem](https://basketly.io/blog/integracje-sklepu-od-czego-zaczac) (pl, 2026-10-26): Co połączyć najpierw w sklepie na Shoperze: API Shopera, konto Meta i odbiorca, przepływ zakupów, skrypt po zgodzie, feedy, potem inne platformy. Stany i ograniczenia.
- [Mapa połączeń sklepu: cała konfiguracja na jednym płótnie i 5-minutowy przegląd raz w tygodniu](https://basketly.io/blog/mapa-polaczen-sklepu) (pl, 2026-10-26): Jak czytać mapę połączeń: skąd przychodzą dane, co robi z nimi Basketly i dokąd trafiają, co znaczą kolory i jak w 5 minut tygodniowo wyłapać, co przestało działać.
- [Produkty bez emisji w reklamach katalogowych Meta: jak je znaleźć i dać im szansę](https://basketly.io/blog/produkty-bez-emisji-w-reklamach-katalogowych) (pl, 2026-10-25): Jak sprawdzić, które produkty Meta prawie nie pokazuje w reklamach katalogowych, czytać wyświetlenia, oglądania i sprzedaż oraz zrobić z nich osobny feed.
- [Raport sprzedaży sklepu co poniedziałek: przychód, wydatki wszystkich kanałów, MER i ROAS bez mieszania źródeł](https://basketly.io/blog/raport-sprzedazy-sklepu-co-poniedzialek) (pl, 2026-10-26): Jak czytać Raporty i poniedziałkowy e-mail: przychód sklepu, wydatki wszystkich kanałów, MER łączny, ROAS według platform i najlepsze produkty.
- [ROAS, MER i koszt zakupu w reklamach Meta: co mierzy każda liczba i której ufać](https://basketly.io/blog/roas-mer-i-koszt-zakupu-w-meta) (pl, 2026-10-26): Czym różni się ROAS z Mety od MER liczonego z przychodu sklepu, jak liczy się koszt zakupu i jak czytać te liczby w Kampaniach i Kreacjach bez podwójnego liczenia.
- [Segmenty klientów sklepu i grupy odbiorców Meta: nowi, powracający, uśpieni i co z nimi zrobić](https://basketly.io/blog/segmenty-klientow-i-grupy-odbiorcow-meta) (pl, 2026-10-25): Jak wybrać klientów po zakupach, zapisać segment i wysłać go do Mety jako grupę odbiorców, która co noc sama się aktualizuje. Z warunkami i ograniczeniami.
- [Wytyczne danych produktowych Google, Meta, TikTok i Pinterest: zasady ze źródłami i tytuły od AI](https://basketly.io/blog/wytyczne-danych-produktowych-google-meta-tiktok-pinterest) (pl, 2026-10-25): Zasady danych produktowych Google Merchant Center, katalogu Mety, TikToka i Pinteresta: jak sprawdzić, które łamią Twoje produkty, i jak działa propozycja tytułu.
- [Zakupy ze sklepu do GA4 i TikToka z serwera: co się wysyła, jak włączyć i czego nie oczekiwać](https://basketly.io/blog/zakupy-do-ga4-i-tiktoka-z-serwera) (pl, 2026-10-26): Jak wysyłać opłacone zamówienia ze sklepu do Google Analytics 4 (Measurement Protocol) i TikToka (Events API) z serwera: połączenie kluczem, przełącznik i ograniczenia.
- [Zamówienia ze sklepu w jednym miejscu: jak czytać przychód, średni koszyk i kupujących](https://basketly.io/blog/zamowienia-ze-sklepu-przychod-koszyk-kupujacy) (pl, 2026-10-26): Co pokazuje strona Zamówienia: kupione, nieopłacone i anulowane zamówienia, przychód w każdej walucie, średni koszyk, nowi i powracający kupujący, 24 miesiące historii.
- [Źródło zamówienia: która kampania i reklama przyprowadziła kupującego](https://basketly.io/blog/zrodlo-zamowienia-ktora-reklama-sprzedala) (pl, 2026-10-26): Jak Basketly ustala źródło zamówienia: parametry URL w reklamach, skrypt w sklepie, wywołanie na stronie podziękowania i identyfikator kliknięcia dla każdej platformy.
- [Asystent AI w sklepie: jak podłączyć Claude i ChatGPT do zamówień, produktów i kampanii](https://basketly.io/blog/asystent-ai-w-crm-claude-chatgpt-mcp) (pl, 2026-10-26): Jak podłączyć Claude, ChatGPT, Cursora albo n8n do Basketly przez MCP: przycisk „Połącz” albo klucz, uprawnienia, narzędzia, skill i czego asystent nie zrobi.
- [Hook rate i hold rate w reklamach Meta: jak je liczyć i co naprawdę znaczą](https://basketly.io/blog/hook-rate-i-hold-rate-meta) (pl, 2026-10-26): Hook rate i hold rate to nie są wskaźniki Meta, tylko proste ilorazy z jej danych. Jak je policzyć, gdzie się mylą i dlaczego nie mówią nic o sprzedaży.
- [Event Match Quality: co to jest i jak poprawić jakość dopasowania zdarzeń w Mecie](https://basketly.io/blog/event-match-quality-jak-poprawic) (pl, 2026-09-16): Czym jest Event Match Quality w Menedżerze zdarzeń, które dane klienta naprawdę pomagają w dopasowaniu, jak je przygotować i kiedy wynik EMQ w ogóle się nie pojawia.
- [Raportowanie kampanii Meta dla agencji: wielu klientów, role i jeden widok wyników](https://basketly.io/blog/raportowanie-kampanii-dla-agencji) (pl, 2026-10-26): Jak agencja może prowadzić raporty kampanii Meta dla wielu klientów: dostęp partnera w Mecie, osobna przestrzeń na klienta, role zespołu i widok wszystkich wyników.
- [RODO a Conversions API: hashowanie, podstawa prawna i jak długo trzymać dane klientów](https://basketly.io/blog/rodo-conversions-api-hashowanie) (pl, 2026-06-10): Czy hash SHA-256 wystarczy, żeby wysyłać dane klientów do Mety zgodnie z RODO? Role stron, podstawa prawna, umowa powierzenia i okres przechowywania danych.
- [Shoper i Meta Conversions API: jak wysyłać zamówienia jako serwerowe zdarzenia Purchase](https://basketly.io/blog/shoper-meta-conversions-api) (pl, 2026-10-26): Jak wysłać opłacone zamówienia ze Shopera do Meta jako serwerowe zdarzenia Purchase: połączenie z API sklepu, webhooki, zgoda, deduplikacja z Pixelem i sprawdzenie.
- [Deduplikacja zdarzeń Pixela i Conversions API: event_id, event_name i okno 48 godzin](https://basketly.io/blog/deduplikacja-zdarzen-piksel-capi) (pl, 2026-04-21): Jak Meta łączy to samo zdarzenie z Pixela i z Conversions API: event_id, event_name, okno 48 godzin, metoda fbp, typowe błędy i sprawdzenie w Menedżerze zdarzeń.
- [Konwersje offline z Arkusza Google do Meta: kolumny, hashowanie i deduplikacja](https://basketly.io/blog/konwersje-offline-z-arkusza-google-do-meta) (pl, 2026-10-13): Jak wysyłać sprzedaż telefoniczną i stacjonarną z Arkusza Google do Meta przez Conversions API: kolumny, format daty i telefonu, hashowanie, limit 7 dni.
- [What Is the Meta Conversions API? How It Differs from the Pixel and What It Won't Do](https://basketly.io/en/blog/what-is-meta-conversions-api) (en, 2026-10-25): What the Meta Conversions API is, how it differs from the Pixel, which data it accepts, how deduplication works and what CAPI does not guarantee.
- [Shop customer segments and Meta audiences: new, returning, lapsed buyers and what to do with them](https://basketly.io/en/blog/customer-segments-meta-custom-audiences) (en, 2026-10-25): How to pick customers by their purchases, save a segment and send it to Meta as a custom audience that updates itself every night. With conditions and limits.
- [Order source: which campaign and ad brought the buyer](https://basketly.io/en/blog/order-source-which-ad-sold) (en, 2026-10-26): How Basketly works out an order's source: URL parameters in your ads, the shop script, a call on the thank-you page and each platform's own click id.
- [Product data guidelines of Google, Meta, TikTok and Pinterest: rules with sources and AI titles](https://basketly.io/en/blog/product-data-guidelines-google-meta-tiktok-pinterest) (en, 2026-10-25): Product data rules of Google Merchant Center, the Meta catalogue, TikTok and Pinterest: how to see which ones your products break and how title suggestions work.
- [Products Meta barely shows in catalogue ads: how to find them and give them a chance](https://basketly.io/en/blog/products-not-shown-in-catalog-ads) (en, 2026-10-25): How to see which products Meta barely shows in catalogue ads, read impressions, product views and sales side by side, and turn them into a separate feed.
- [Purchase balance: how to check that shop orders really reach Meta and other platforms](https://basketly.io/en/blog/purchase-balance-are-purchases-reaching-meta) (en, 2026-10-26): How to read the day-by-day purchase balance: shop orders, purchases Basketly accepted and sends to Meta and other platforms. What each gap means and what to check.
- [ROAS, MER and cost per purchase in Meta ads: what each number measures and which to trust](https://basketly.io/en/blog/roas-mer-cost-per-purchase-meta) (en, 2026-10-26): How Meta's ROAS differs from MER based on shop revenue, how cost per purchase is counted, and how to read them in Campaigns and Creatives without double counting.
- [Shop purchases to GA4 and TikTok from the server: what is sent, how to switch it on and what not to expect](https://basketly.io/en/blog/send-purchases-to-ga4-and-tiktok-server-side) (en, 2026-10-26): How to send paid shop orders to Google Analytics 4 (Measurement Protocol) and TikTok (Events API) from the server: a key connection, one switch and honest limits.
- [Shop automations: a paid order, a customer who stopped buying and an alert when purchases stop arriving](https://basketly.io/en/blog/shop-automations-paid-order-lapsed-buyer-alert) (en, 2026-10-26): The three shop triggers in Basketly and the "E-mail the team" action: when they start, what they watch, ready templates and limits to know before switching on.
- [The shop connection map: your whole setup on one canvas and a 5-minute weekly check](https://basketly.io/en/blog/shop-connection-map) (en, 2026-10-26): How to read the connection map: where data comes from, what Basketly does with it, where it goes, what the colours mean and how to spot breakage in 5 minutes a week.
- [Shop integrations: where to start so orders, ads and the catalogue work together](https://basketly.io/en/blog/shop-integrations-where-to-start) (en, 2026-10-26): What to connect first in a Shoper shop: the Shoper API, Meta login and receiver, the purchase flow, the consent script, feeds, then other platforms. States and limits.
- [All your shop orders in one place: how to read revenue, average basket and buyers](https://basketly.io/en/blog/shop-orders-revenue-basket-buyers) (en, 2026-10-26): What the Orders page shows: bought, unpaid and cancelled orders, revenue per currency, average basket, new and returning buyers, and 24 months of order history.
- [A weekly shop sales report every Monday: revenue, spend of every channel, MER and ROAS without mixing sources](https://basketly.io/en/blog/weekly-shop-sales-report) (en, 2026-10-26): How to read Reports and the Monday e-mail: shop revenue, spend of every channel, blended MER, ROAS as each platform reports it and best-selling products.
- [An AI assistant for your online shop: connecting Claude and ChatGPT to orders, products and campaigns](https://basketly.io/en/blog/ai-assistant-crm-claude-chatgpt-mcp) (en, 2026-10-26): How to connect Claude, ChatGPT, Cursor or n8n to Basketly through MCP: the Connect button or a key, permissions, tools, the skill and what the assistant won't do.
- [Hook Rate and Hold Rate in Meta Ads: How to Calculate Them and What They Really Mean](https://basketly.io/en/blog/meta-hook-rate-hold-rate) (en, 2026-10-26): Hook rate and hold rate aren't Meta metrics, just simple ratios of Meta's numbers. How to calculate them, where they mislead, and why they say nothing about sales.
- [Event Match Quality: What It Is and How to Improve It in Meta Events Manager](https://basketly.io/en/blog/improve-event-match-quality) (en, 2026-09-16): What Event Match Quality in Meta Events Manager measures, which customer parameters actually help matching, how to prepare them, and when EMQ does not appear at all.
- [Meta Ads Reporting for Agencies: Many Clients, Clear Roles and One View of Results](https://basketly.io/en/blog/agency-meta-reporting-multiple-clients) (en, 2026-10-26): How an agency can report on Meta campaigns for many client shops: partner access in Meta, one workspace per client, team roles and an all-clients overview.
- [GDPR and the Meta Conversions API: Hashing, Lawful Basis and How Long to Keep Customer Data](https://basketly.io/en/blog/gdpr-conversions-api-hashing) (en, 2026-06-10): Is SHA-256 hashing enough to send customer data to Meta under GDPR? Who is controller and processor, lawful basis, the DPA and how long to keep customer data.
- [Server-Side Purchase Events for Online Stores: Sending Orders to Meta Conversions API](https://basketly.io/en/blog/ecommerce-server-side-purchase-events) (en, 2026-10-26): How to send paid online store orders to Meta as server-side Purchase events: a direct Shoper API connection, webhooks, consent, Pixel deduplication and checks.
- [Pixel and Conversions API Deduplication: event_id, event_name and the 48-Hour Window](https://basketly.io/en/blog/pixel-capi-event-deduplication) (en, 2026-04-21): How Meta merges the same event from the Pixel and the Conversions API: event_id, event_name, the 48-hour window, the fbp method, common mistakes and how to check.
- [Offline Conversions from Google Sheets to Meta: Columns, Hashing and Deduplication](https://basketly.io/en/blog/offline-conversions-google-sheets-meta) (en, 2026-10-13): How to send phone and in-store sales from a Google Sheet to Meta via the Conversions API: required columns, date and phone formats, hashing and the 7-day limit.

## Authors
- [Aleksander](https://basketly.io/en/author/aleksander) (en): Founder of Basketly.io, BooksyAudit.pl and SMSPortal.app
- [Aleksander](https://basketly.io/autor/aleksander) (pl): Założyciel Basketly.io, BooksyAudit.pl i SMSPortal.app