The E-commerce Glossary

Keep in mind : A common mistake is starting a test without enough traffic to reach a useful conclusion.

Sources: Shopify

Keep in mind : A short, self-contained answer can help readers, but it does not guarantee inclusion in an AI answer or replace SEO fundamentals.

Sources: Google

Keep in mind : Product discovery, referrals to a store, and checkout inside an assistant are different capabilities. Availability depends on the channel, buyer's market, merchant eligibility, and setup. Shopify's ChatGPT channel sends shoppers to the merchant's online store checkout.

Sources: Shopify 1 · Shopify 2

Keep in mind : Availability and supported features vary by market, language, and account. Neither feature guarantees a visit to the websites it cites.

Sources: Google

Keep in mind : Do not add up the sales claimed independently by different platforms. Last click, last non-direct click, and data-driven attribution tell different stories. Attribution alone does not prove that marketing caused additional sales.

Sources: Google

Keep in mind : State how tax, shipping, and refunds are handled. Shopify calculates AOV as gross sales minus discounts, divided by orders, excluding adjustments made after order creation. This is not interchangeable with every definition of net sales per order.

Sources: Shopify

Keep in mind : Universal Analytics used a definition based on a single interaction request, not simply a single page view. Historical UA bounce rates are not directly comparable with GA4. Also check the configured engagement threshold.

Sources: Google 1 · Google 2

Keep in mind : Keep the unit, population, and time window consistent. Do not divide all store orders by sessions with an add-to-cart action. A started checkout without a purchase measures checkout abandonment, which is not exactly the same as cart abandonment.

Sources: Shopify

Sources: Shopify

Keep in mind : Consent Mode does not collect consent itself or make an implementation compliant. Depending on setup, cookieless measurements may be sent when some permissions are denied, and some results may be modeled. A Shopify–GA4 discrepancy is not automatically a bug or solely a consent issue.

Sources: Google

Keep in mind : State whether acquisition spending is included and list the costs deducted. For break-even ROAS, POAS, or an LTV-to-CAC comparison, use contribution before the acquisition expense being assessed to avoid deducting it twice.

Keep in mind : GA4 offers a Purchase journey report and Funnel exploration, which supports up to ten steps.

Keep in mind : Shopify uses sessions that completed a purchase, not the number of orders. In GA4, the overall key event rate counts sessions with any key event; use the rate specific to purchase to track buying. Orders divided by sessions measures order frequency and may differ from the share of sessions that purchase.

Sources: Shopify · Google

Keep in mind : Use the contribution margin as a decimal, such as 0.40 for 40%. A direct cohort measure is cumulative contribution over H divided by the initial cohort size. If comparing LTV with CAC, do not already deduct CAC from contribution. Long-range forecasts may require discounting and explicit retention assumptions.

Sources: Shopify

Keep in mind : On Shopify, do not assume that a theme's data layer is available in checkout: customer events and pixels use supported APIs and sandboxes. A data layer is not necessarily one flat object shared across the entire site.

Sources: Google · Shopify

Keep in mind : Session timeout and the engaged-session threshold are separate settings. Engagement rate is the share of sessions meeting at least one of these criteria.

Sources: Google 1 · Google 2

Keep in mind : It can hurt the experience and contaminate exposure to an experiment. Check its extent, whether it affects variants differently, and how the testing tool loads. Its presence alone does not prove that every result is statistically invalid.

Keep in mind : An achievable threshold can encourage shoppers to add items, but the effect depends on the products, the amount still needed, and the shipping fee avoided. Make conditions clear and test conversion, AOV, and contribution. An excessive threshold may have no useful effect or discourage a purchase.

Keep in mind : GEO overlaps with parts of SEO and AEO. A mention, a linked citation, and a visit are different outcomes. No page structure or markup guarantees a citation.

Sources: Google

Keep in mind : With a $20 product cost and a $50 selling price before tax, gross profit is $30 and gross margin is 60%. Decimal ratios and percentages are alternative ways to express a rate; do not mix them.

Sources: Shopify

Keep in mind : Use the appropriate GS1 identifiers and retain identifiers already assigned by the manufacturer. Do not create a new GTIN simply because the same product is sold on another website or in another country.

Sources: Google

Keep in mind : The overall rate includes sessions with at least one key event. To measure purchases, use the rate specific to purchase. Adding another key event does not change a correctly scoped purchase rate.

Sources: Google

Keep in mind : It does not grant crawling permission or guarantee citations. Adoption varies between tools. It does not replace robots.txt, accessible pages, or SEO fundamentals.

Sources: Google

Keep in mind : A product costing $20 and selling for $50 before tax has a 150% markup, not 1.5%. Do not confuse markup with gross margin: Shopify calculates gross margin against sales, not cost.

Sources: Shopify

Keep in mind : State whether the effect is a relative change or a change in percentage points. With the same power and error threshold, detecting a smaller effect generally requires a larger sample. A small observed effect is not automatically undetectable.

Sources: Microsoft Research

Keep in mind : Specify the costs deducted: product, logistics, payments, returns, and so on. Under this convention, a ratio of 1 covers advertising costs but not necessarily fixed overhead. Attribution alone does not establish the additional profit caused by advertising.

Sources: Google

Keep in mind : A multiplier of 2.5 means a $20 product cost becomes a $50 selling price before tax. It is not a 250% markup: the markup is 150%. Compare cost and selling price on the same tax basis.

Keep in mind : Formats, required fields, and update frequency depend on the destination. Keep the feed consistent with the product page and checkout.

Sources: Google

Keep in mind : A former selling price, a suggested retail price, and a competitor's price are different comparisons. Label the basis clearly and check the advertising rules for your target market. A crossed-out number alone does not establish a genuine saving.

Sources: Shopify

Keep in mind : Compare cohorts at the same age. The share of returning customers among people buying in a particular month is a different measure: a long-standing customer may have placed only one order during that month.

Sources: Shopify

Keep in mind : The exact identity is RPS = (orders ÷ sessions) × AOV when all definitions match. If conversion rate counts purchasing sessions, also account for orders per purchasing session. RPS measures revenue, not profit.

Sources: Shopify 1 · Shopify 2

Keep in mind : With a 40% contribution margin before advertising, expressed as 0.40, break-even ROAS for ad spend is 1 ÷ 0.40 = 2.5, using consistent revenue definitions. This does not automatically cover fixed costs or prove incrementality. A ROAS of 2.5 is 250% when displayed as a percentage.

Sources: Google

Keep in mind : This setup does not guarantee that every blocker is bypassed or that missing data can be recovered. It does not remove consent obligations or applicable privacy requirements. Check deduplication when the browser and server send the same purchase.

Sources: Google 1 · Google 2

Keep in mind : Leaving and returning before the timeout can remain one session. One person can generate multiple sessions, but three perceived visits do not necessarily mean three measured sessions. Compare session-based and user-based rates separately; neither rate is universally higher.

Sources: Google 1 · Google 2

Keep in mind : Configure masking and data collection carefully. Recording tools can capture personal information; assess consent and privacy requirements for the markets you serve.

Keep in mind : At an alpha of 5%, a p-value of 0.05 or less leads to rejection of the null hypothesis under the protocol. It does not mean there is a 95% chance the variant is better or that the effect matters commercially. Follow the planned duration and stopping rule; sequential methods need their own valid rules.

Sources: Microsoft Research

Keep in mind : Markup must match the real, visible content of the page. The JSON-LD script is not a visual element, but the information it describes must not contradict what shoppers see.

Sources: Google 1 · Google 2

Keep in mind : A published standard is not the same as an available sales channel. Check supported capabilities and merchant eligibility. Shopify currently routes ChatGPT shoppers to the merchant's online store checkout.

Sources: Shopify · Google · OpenAI

Keep in mind : Use them on external campaign links. In GA4, a UTM encountered during a session does not start a new session or rewrite its source, but it can affect event attribution and confuse campaign reporting. Use event tracking for internal links.

Sources: Google 1 · Google 2