Glossary

Definitions for every attribution and app-specific term used in these docs and in the console.

AIR Engine Pro feature: Attribution Intelligence Report. A Pro tab in Refer App that analyses all unclassified Direct sessions against 20 deterministic pattern rules. For each rule that fires, the engine produces a finding card with a confidence score, a verifiable fact base, and a recommended action. Findings are grouped into five categories: Dark Social, Cross-Device, Legacy URLs, B2B Authority, and Automated Traffic. See also: Attribution Health Score, Confidence score, Finding card.
Attribution noun: The process of identifying which marketing channel or source brought a visitor to your site. "First-touch attribution" means the source of their very first visit gets the credit — there is no weighted distribution across multiple touchpoints.
Attribution Coverage Insights metric: The percentage of sessions assigned to a classifiable channel. High Direct share with no UTM parameters is the most common cause of low attribution quality.
Attribution Health Score AIR Engine metric: A score from 0 to 100 produced by the AIR Engine summarising the overall quality of the installation's Direct traffic picture. Derived from classified session ratio, problem severity across findings, and channel diversity. Colour-coded: green (Strong, 80+), amber (Fair, 60-79), red (Critical, below 60). Distinct from the Insights Health Score, which measures overall attribution setup quality.
Beacon technical: The JavaScript file added to your site's footer. It fires on every page load and sends the visitor's referrer, landing URL, page path, page title, and viewport category to your installation.
c_ parameter aliases tracking: Shorter, privacy-resilient URL parameters that serve as fallbacks when standard UTM parameters are stripped by browser privacy tools. The three aliases are c_src (fallback for utm_source), c_med (fallback for utm_medium), and c_cam (fallback for utm_campaign). Standard UTM parameters always take precedence. See also: UTM parameters, Short link.
Channel classification: The category the classifier assigns to a session based on its referrer and UTM parameters. The nine channels are: AI Referrer, Paid Search, Paid Social, Email, Organic Social, Organic Search, Internal, Referral, and Direct. The classifier works through these in priority order — the first match wins.
Classifier technical: The server-side logic that reads a session's referrer domain and UTM parameters and assigns it to a channel. Runs on your server at the moment a beacon hit arrives.
Click count Links metric: The number of times a short link has been followed. Incremented atomically on each redirect. Displayed in the Links tab alongside each short link.
Confidence score AIR Engine metric: A number from 0 to 100 attached to each AIR Engine finding card. Computed from five weighted factors: trigger condition match (30 points), cross-channel overlap on the same page path (25 points), Direct-to-known ratio elevation above site baseline (20 points), temporal correlation rate (15 points), and rising velocity trend (10 points). Findings above 80 are Critical, 60-79 are Notable, below 60 are Advisory.
Conversion noun: An outcome linked to a session — a sale, a signed contract, a lead, or any event you consider valuable. Recorded either automatically via a conversion rule or manually in the Conversions tab. See also: Conversion rule, Real revenue, Estimated revenue.
Conversion rule Pro feature: A URL pattern rule that automatically records a conversion against a session when a visitor reaches a matching page. Rules match on the page path using "contains", "starts with", or "equals" logic. One auto-conversion per rule per session is enforced.
Dark social traffic pattern: Shares and referrals through private or closed channels — messaging apps, email clients, Slack, Discord, LinkedIn DMs — where no referrer header is passed to the destination. Dark social traffic arrives with no source and lands in the Direct channel. It is typically the largest single contributor to unexplained Direct traffic on content-led sites.
Direct channel: Traffic with no referrer — visits from typed URLs, bookmarks, or links in apps and messages that strip the referrer header. High Direct traffic usually means links are not tagged with UTM parameters, or traffic is arriving from messaging apps. Using UTM parameters or short links on campaign links is the fix.
Direct-to-known ratio AIR Engine signal: One of the four predictive signals used by the AIR Engine. For each page path, the ratio of Direct first-touch sessions to all first-touch sessions on that path, compared against the site-wide average. A page with a significantly elevated ratio is a stronger candidate for dark social or branded revisit behaviour.
Engagement signal session attribute: A flag indicating whether the visitor clicked through to at least one more page after their first touch. Set to "engaged" the moment a second beacon hit arrives for the same session token. Visible in the Engaged column in Reports and in the Engagement Rate card on the Dashboard.
Engagement rate Dashboard metric: The percentage of sessions where the visitor clicked through to at least one additional page. A measure of traffic quality. See also: Engagement signal.
Estimated revenue conversion type: A conversion value recorded when the actual amount is not yet confirmed. Always shown separately from real revenue and never added to it.
Finding card AIR Engine: A structured result card produced by the AIR Engine when a pattern rule fires. Each card shows: the category and rule name, the confidence score, a fact base displaying the raw database counts used to compute the score, a plain-English explanation of what the pattern indicates, and a specific recommended action. Findings can be marked as reviewed, which moves them to a collapsed section at the bottom of the report.
First-touch attribution model: The attribution approach used by Refer App. The channel and source that sent the visitor on their very first page view receives full credit for any conversion recorded against that session.
Halo-effect attribution traffic pattern: Traffic generated indirectly when a brand is mentioned or cited by another channel, causing users to search for or navigate directly to the site rather than clicking a tracked link. Common with AI tool citations, podcast mentions, and press coverage. Arrives as Direct or Organic Search with no connection to the original channel that generated the awareness.
Health Score Insights metric: A score from 0 to 100 summarising the overall quality of your attribution setup. Calculated from Attribution Coverage, UTM Completeness, AI Taxonomy Activity, and Conversion Tracking (25 points each).
Ingestion endpoint technical: The server-side URL that receives every beacon hit. If you rename the refer-app folder, this URL changes and the beacon script must be updated to match.
Internal traffic channel: Sessions where the referrer is the same domain as the tracked site. With the self-referral exclusion enabled in Settings, Internal sessions are not recorded.
Organic Search channel: Traffic from a search engine with no paid-search UTM signal. Recognises Google, Bing, DuckDuckGo, Yahoo, Baidu, Yandex, and other major search engines.
Page depth Insights metric: The number of pages a converting visitor viewed between their first touch and their first recorded conversion. Used in the "How People Convert" panel in Insights.
Paid Search channel: Clicks from search engine ads, identified by utm_medium=cpc, utm_medium=ppc, or utm_medium=paid.
Pattern rule AIR Engine: One of 20 deterministic conditions evaluated by the AIR Engine against each page path in the selected date range. Each rule targets a specific traffic behaviour pattern and fires when the session data on a path meets the rule's threshold conditions. A fired rule produces a finding card.
Predictive signal AIR Engine: One of four computed metrics the AIR Engine uses when scoring a finding: Direct-to-known ratio, temporal correlation, viewport divergence, and velocity trend. Each signal is calculated from the session data already in the database. No external data is used.
Real revenue conversion type: A conversion value recorded against a confirmed transaction. Always shown separately from estimated revenue.
Referral channel: Traffic from any website that is not a known search engine, social network, email provider, or AI tool, and has no UTM signal overriding the classification.
Referrer technical: The URL of the page a visitor was on before arriving at your site. Some browsers, privacy tools, and link sources strip the referrer — traffic that arrives with no referrer is classified as Direct.
Session noun: A single visit by one browser to your site, defined by the session cookie window. All page views within the session window are grouped together. Attribution is determined by the first page view (first-touch).
Session Journey Reports feature: An expandable inline view of every page a visitor touched during their session, in chronological order with the time between each step. Available for sessions with more than one page view. Click the Touches count in Reports to expand.
Session token technical: A random identifier stored in the visitor's session cookie that groups page views from the same visit together. Contains no personal information.
Short link Pro feature: A branded redirect URL in the format yoursite.com/go/{slug} that redirects to a destination URL with campaign parameters appended server-side. Browser privacy filters cannot strip the parameters because the redirect happens before the destination page loads. See also: Slug, c_ parameter aliases.
Slug Links: The identifier in a short link after /go/. In yoursite.com/go/summer-launch, the slug is summer-launch. Slugs must be unique and can contain letters, numbers, hyphens, and underscores.
Taxonomy match classification field: The human-readable label assigned when a referrer domain matches an entry in the AI Taxonomy. For example, a visit from chat.openai.com produces a taxonomy match of ChatGPT.
Temporal correlation AIR Engine signal: One of the four predictive signals used by the AIR Engine. Measures how many Direct sessions on a page path arrived within 72 hours of a known-channel session on the same path. A high correlation rate is consistent with cross-device revisit behaviour: a visitor saw the page through a tracked channel, then returned directly on a different device or browser shortly after.
Touches Reports column: The total number of page views recorded within a session. Sessions with more than one touch show a clickable count that expands the Session Journey view.
UTM parameters tracking: Query string parameters added to URLs to provide attribution context. The five standard parameters are utm_source, utm_medium, utm_campaign, utm_term, and utm_content. Refer App also supports c_ aliases as fallbacks when UTM parameters are stripped by privacy browsers.
UTM Completeness Insights metric: The percentage of sessions that have at least one UTM parameter. Low UTM Completeness means marketing links are not tagged, causing traffic to fall back to Direct even when it came from a specific campaign.
Velocity trend AIR Engine signal: One of the four predictive signals used by the AIR Engine. Splits the selected date range in half and compares Direct session counts on a path in the recent half against the earlier half. A rising trend increases the urgency of the finding and contributes 10 points to the confidence score.
Viewport divergence AIR Engine signal: One of the four predictive signals used by the AIR Engine. Compares the desktop/mobile split of Direct sessions on a path against the known-channel sessions on the same path. A large gap — for example, Direct sessions are 80% desktop while organic sessions are 60% mobile — is a strong indicator of cross-device revisit behaviour.
Viewport category session attribute: A coarse device classification recorded at first page view: mobile (under 768px), tablet (768-1023px), or desktop (1024px and above). Visible in the Device column in Reports and the Visitors by device section on the Dashboard.
Webhook Pro feature: An HTTP POST request sent automatically by Refer App to a URL you configure, carrying attribution data as a JSON payload. Session webhooks fire when a new session is logged. Conversion webhooks fire when a conversion event is recorded. Webhooks let you push attribution data to external systems without a plugin connector. See also: Integrations.
White Label Pro feature: A Pro feature that replaces the default Refer App branding on the Branded Report with your own agency name, logo, and footer text. Applies only to the client-facing PDF report, not the console.

Terms cover CQI Referrer Attribution (WordPress), Refer App, and the AIR Engine. Back to documentation.