Glossary · 43 terms

The vocabulary, defined plainly.

Each term has one definition written to stand on its own, a longer explanation, a source for any number, and a separate note on how Liftable relates to it, so the definition stays neutral.

Conversion

Conversion rate: Conversion rate is the share of sessions, or of visitors, that complete a defined goal, usually a purchase.
Conversion funnel: A conversion funnel is the ordered sequence of steps a shopper passes through on the way to a purchase: landing, viewing a product, adding to cart, reaching checkout and buying.
Add-to-cart rate: Add-to-cart rate is the share of sessions in which at least one product is added to the cart.
Product detail page (PDP): A product detail page (PDP) is the page that presents a single product: imagery, price, variants, delivery information and the add-to-cart action.
Cart abandonment: Cart abandonment is a shopper adding items to a cart and leaving without buying.
Checkout abandonment: Checkout abandonment is a shopper starting the checkout process and leaving before completing payment.
Revenue per session: Revenue per session is total revenue divided by sessions over the same period.
Average order value (AOV): Average order value (AOV) is total revenue divided by the number of orders in a period.

Friction

Friction: Friction is anything in a shopping experience that makes the next step harder than the shopper expected: a control that does not respond, information that arrives too late, a form that asks too much, a page that takes too long.
Dead click: A dead click is a click on a page element that produces no response: no navigation, no change of state, no visible feedback.
Rage click: A rage click is a rapid burst of clicks on the same element in a short window, which shows that a shopper expected a response and got none, or got it too slowly.
Session replay: Session replay reconstructs one visitor’s session as a watchable recording of pages, pointer movement, clicks and scrolling.
Heatmap: A heatmap aggregates interactions across many sessions into a visual overlay on a single page, showing click density, scroll depth or attention.
JavaScript bloat: JavaScript bloat is the accumulated weight and execution cost of the scripts a storefront loads, usually from installed apps.
Largest Contentful Paint (LCP): Largest Contentful Paint (LCP) measures how long the largest visible element on a page, usually a hero image or product photo, takes to render.
Mobile conversion gap: The mobile conversion gap is the difference between the conversion rate on phones and on desktop for the same store.

Testing

Split test: A split test randomly assigns visitors to two or more variants of an experience and compares an outcome between them.
Statistical significance: Statistical significance is the judgement that an observed difference would be unlikely to arise by chance if there were no real effect.
Sequential testing: Sequential testing is a family of statistical methods that stay valid while a test is being monitored, so results can be checked at any point without inflating the false-positive rate.
Peeking: Peeking is checking a fixed-horizon test before its planned sample size is reached and acting on what you see.
Minimum detectable effect (MDE): The minimum detectable effect (MDE) is the smallest change a test can reliably distinguish from noise, given its traffic, baseline rate and duration.
Power analysis: Power analysis is the calculation, made before a test starts, of how likely the test is to detect an effect of a given size if that effect is real.
Always-valid p-value: An always-valid p-value is a p-value that stays valid however often it is checked and whenever the test is stopped: if there is no real effect, the chance it ever drops below the significance level is held at that level.
Mixture sequential probability ratio test (mSPRT): The mixture sequential probability ratio test (mSPRT) is a sequential test that, as each batch of data arrives, compares how likely the data are if there is no effect with how likely they are averaged over a range of plausible effects.
Holdback: A holdback is a small share of traffic deliberately kept on the original experience after a winning variant has gone live.
Novelty effect: The novelty effect is a temporary change in behaviour caused by an experience being new rather than being better.
Sample ratio mismatch (SRM): Sample ratio mismatch (SRM) is a statistically significant difference between the traffic split a test intended and the split it actually received, such as an even split that arrives visibly uneven across a large sample.
Guardrail metric: A guardrail metric is a measure a test must not harm, monitored alongside the metric the test is trying to move, so that a variant that lifts its target while damaging something else is stopped.
Pre-registration: Pre-registration is writing down a test’s plan before it starts: the primary metric, the minimum detectable effect, the significance level, the guardrails and the stopping rule.
Theme app extension: A theme app extension is the way a Shopify app adds content or code to a store’s theme without editing the theme’s files: the app supplies app blocks that you place in the theme editor and app embed blocks that you switch on.
Shopify Function: A Shopify Function is a small program an app deploys to Shopify to customise part of Shopify’s backend logic, such as discounts, delivery options, payment options, or cart and checkout validation.

Traffic

Bot traffic: Bot traffic is automated, non-human requests to a store: crawlers, scrapers, uptime monitors, link-preview fetchers and fraud tooling.
Direct traffic: Direct traffic is the analytics bucket for sessions with no identifiable referrer or campaign parameter.
AI referral traffic: AI referral traffic is visits to a store that come from an AI assistant such as ChatGPT, Perplexity, Gemini or Copilot after it recommended or cited a product or page.
Answer engine optimisation (AEO): Answer engine optimisation (AEO) is the practice of writing and structuring pages so that answer engines, meaning AI assistants and search features that reply with a direct answer, can find a page, lift a self-contained passage from it and cite the source.
Generative engine optimisation (GEO): Generative engine optimisation (GEO) is the practice of structuring content and data so that AI assistants cite and recommend a source in their generated answers.
llms.txt: llms.txt is a proposed convention for a Markdown file at the root of a website, at /llms.txt, that gives large language models a short summary of the site and links to the pages most worth reading.
Structured data (JSON-LD): Structured data is machine-readable markup that describes what a page means.
UTM parameters: UTM parameters are query-string tags added to a link (source, medium, campaign, content and term) that tell analytics where a visit came from.

Money

Revenue leak: A revenue leak is a specific, named point in a store where shoppers who meant to buy are lost, such as a size-guide control on a product page that does nothing when tapped.
Verified revenue: Verified revenue is revenue attributed to a change that has been confirmed after the change went live, by comparing the live experience with a holdback still receiving the original.
Incremental revenue: Incremental revenue is the revenue a change caused, over and above what the store would have earned without it.
Confounding: Confounding is when a third factor drives both the behaviour being measured and the outcome, making an association look causal when it is not.

Scan your store.

See where your store is losing revenue and what to fix first. It is free, takes about a minute, and needs no install.

Scan your store