Short answer: use a hosted link when reliability and fast launch matter most; an inline embed when the surrounding page must remain visible; a full-page embed for a dedicated on-brand quiz route; and a triggered pop-up, drawer, or side tab only when the launch trigger, focus, close control, frequency, mobile behavior, and non-script fallback are tested. “Embeddable” is not evidence that all four modes behave alike.
Key findings, verified 18 October 2026
- involve.me documents responsive standard, full-size, and pop-up embeds, dynamic-height resizing, page- or domain-based hiding for supported triggered modes, and hidden-field context through
data-paramsor matching parameters on the parent page URL. - Interact requires the quiz to be published before its embed reflects the latest changes and requires a host block that accepts custom HTML or embed code; its editor preview may not show the live embed.
- Typeform documents Standard, Full page, Popup, Slider, Popover, and Side tab modes. Its mobile embeds default to full-screen pop-ups, with an inline override available for Standard embeds, and the host page must use HTTPS.
- Outgrow documents full-page, in-page, pop-up, chat, custom, floating, bar, and side-note modes. Its full-page mobile flow displays a welcome screenshot before the content takes over the screen.
What is a quiz embed?
A quiz embed places vendor-hosted interactive content inside a page controlled by another site. A hosted link sends the visitor to a vendor-controlled destination. A full-page embed keeps the host URL while the quiz fills the main viewport. A pop-up, drawer, or side tab launches from the host page after a trigger. These choices alter navigation, sizing, mobile behavior, accessibility, analytics, consent, and recovery even when the quiz itself is unchanged.
Quiz publishing-mode matrix
| Mode | Page ownership | Best fit | Critical test | Failure fallback |
|---|---|---|---|---|
| Hosted link | Vendor owns the destination | Email, social, QR, or fast launch | Domain trust, return path, campaign parameters | Link from a stable explanatory page |
| Inline embed | Host page surrounds the quiz | Buying guide, lesson, or product page | Dynamic height, nested scroll, keyboard order, content blocking | Visible link to the hosted version |
| Full-page embed | Host URL; vendor content fills the main view | Dedicated quiz landing route | Viewport sizing, browser chrome, back behavior, mobile takeover | Direct hosted URL |
| Triggered overlay | Host page plus scripted pop-up, drawer, or tab | Optional product finder or feedback prompt | Trigger, focus, close, repeat frequency, mobile equivalent | Persistent inline link or button |
How do current vendor modes differ?
| Product | Documented modes | Sizing or mobile distinction | Publication or host constraint | Context or state distinction |
|---|---|---|---|---|
| involve.me | Standard, full-size, pop-up, plus triggered variants documented in the help center | Responsive iframe; dynamic height is on by default when height and width remain auto | Use the generated code; changing the funnel URL requires replacing the embed code | Hidden fields can receive data-params; matching parent-page parameters can also be passed |
| Interact | Inline host-specific embed workflows; separate host instructions | Validate in preview and on the published host page because an editor may not render the embed | The quiz must be published and the host must accept custom HTML or embed code | The embedded version reflects the most recently published quiz |
| Typeform | Standard, Full page, Popup, Slider, Popover, Side tab | Mobile defaults to full-screen pop-ups; Standard can be forced inline with the documented attribute | HTTPS host required; use the generated Live embed code | Form edits require Publish edits; saved embed-setting changes can apply without re-embedding |
| Outgrow | Full page, in-page, pop-up, chat, custom, floating rectangle, greet bar, side note | Full-page mobile begins with a welcome screenshot and then takes over; custom iframe can vary dimensions by device | Full-page HTML needs the documented responsive viewport tag | Mode and device behavior must be recorded with the generated code |
Sources: involve.me share and embed documentation, Interact embed guide, Typeform embed documentation, and Outgrow embedding options.
Ten-field embed audit
- Host URL: record the exact preview and production route.
- Publishing mode: hosted, inline, full page, or named overlay.
- Vendor destination: retain the stable hosted quiz URL.
- Required code: save the generated script, iframe wrapper, or SDK version.
- Launch trigger: button, page load, delay, scroll, exit intent, tab, or direct navigation.
- Sizing: width, minimum height, dynamic resize, overflow, and nested scrolling.
- Mobile behavior: inline, placeholder, full-screen pop-up, takeover, or a different trigger.
- Context passing: parent URL parameters, hidden fields, campaign values, and consent scope.
- Completion signal: visible result, redirect, event, contact update, or no exposed event.
- Fallback: a keyboard-reachable hosted link that still works without the embed script.
Run the audit with keyboard-only desktop, a narrow mobile viewport, slow loading, blocked third-party scripts, restricted third-party storage, and the real CMS preview plus production page. This is a protocol to perform, not a claim that those tests were completed for the four products here.
Worked example: an embedded product finder
A retailer places a product-finder quiz inside a buying guide. Inline mode keeps sizing advice and comparison copy visible, but a short fixed container creates nested scrolling on mobile. Full-page mode removes that immediate layout conflict but also removes most of the guide context. A pop-up protects the page layout, yet adds a trigger, focus transition, close control, repeat-frequency decision, and mobile equivalent.
The retailer chooses inline delivery with automatic height, passes only approved campaign context, and keeps a visible “Open the finder on its own page” link below the container. The release is not complete until the result state, fallback, keyboard path, consent text, and analytics boundary work on the published CMS page.
Common embed failures
- A page builder strips the script while its editor shows a convincing placeholder.
- The quiz stays inline on desktop but launches as a full-screen flow on mobile.
- A fixed-height iframe cuts off results or creates two independent scroll areas.
- Exit intent is described as universal even though pointer exit has no identical mobile or keyboard equivalent.
- Campaign or known-contact values are added to the wrong URL or without an approved data contract.
- The embed has no usable fallback when a privacy tool, content blocker, network failure, or CMS policy prevents its script from loading.
Method and limitations
We compared all four vendors under the same architecture fields: documented modes, page ownership, host requirement, sizing, mobile transformation, context passing, update behavior, completion boundary, failure surface, and fallback. Findings were verified against current first-party documentation on 18 October 2026. Documentation establishes supported configuration, not compatibility with every CMS, theme, consent manager, browser, analytics stack, or security policy.
No subscriptions, authenticated embeds, production CMSs, live scripts, real contact data, analytics events, consent managers, device lab, content blockers, performance benchmark, security review, or accessibility-conformance audit were used. Plan gates and generated code were not inspected inside paid accounts. Revalidate the exact code, mobile mode, data parameters, and plan before launch.
For adjacent checks, use the participant-access matrix, mobile compatibility matrix, and accessible live-quiz checklist. Business teams deciding what should happen after completion can use the business lead-generation guide.
Correction path: report a changed mode, mobile default, script requirement, update rule, context-passing mechanism, plan gate, or host limitation through the corrections process with the exact claim and a current first-party source. Verification date: 18 October 2026.