Conversion attribution model

How elNudge attributes conversions to nudge interactions and what the lift metric means.

Understanding which conversions elNudge influenced — and which would have happened anyway — is central to measuring its impact on your store.

What counts as a conversion?

A conversion is recorded when a visitor completes a purchase during their session. Technically, this happens when a PURCHASE event is fired — either automatically via the Shopify integration or manually via window.__eln('track', 'PURCHASE', ...) in a custom setup.

One session can have at most one conversion.

What is a nudge-influenced conversion?

A conversion is classified as nudge-influenced when both of the following are true:

1. The visitor interacted with a nudge during the session — this means they clicked a nudge, dismissed it with intent (e.g. a "No thanks" button), or opened the chat.
2. The visitor completed a purchase within 30 minutes of that nudge interaction, either within the same session or immediately after returning.

Simply seeing a nudge (a "shown" event) does not qualify. The visitor must have interacted with it.

The 30-minute attribution window

The attribution window is 30 minutes from the last nudge interaction. This means:

• If a visitor chats with the agent at 2:00pm and purchases at 2:25pm — nudge-influenced.
• If a visitor clicks a nudge at 2:00pm and purchases at 2:35pm — not nudge-influenced (outside the window).
• If a visitor starts a new session at 2:05pm and purchases at 2:20pm — nudge-influenced (window carries across the session gap, provided it is within 30 minutes of the interaction).

Baseline conversion rate

The baseline conversion rate is the conversion rate of sessions where the visitor had no nudge interaction at all — they may have had nudges shown, but they did not click, dismiss, or open chat. This is the best available approximation of what conversion rate your store achieves without nudge influence.

Reading the lift metric

The lift figure shown in the dashboard compares:

• Nudge-influenced CVR — conversions per session, for sessions with at least one nudge interaction
• Baseline CVR — conversions per session, for sessions with no nudge interaction

Example:

• Baseline CVR: 1.8%
• Nudge-influenced CVR: 4.2%
• Lift: +133%

This means sessions where a visitor interacted with a nudge converted at 2.3x the rate of sessions with no nudge interaction.

Limitations

Correlation vs causation. High-intent visitors are more likely to interact with a nudge — they are already engaged and paying attention. They are also more likely to convert regardless. This means some portion of nudge-influenced conversions would have happened even without the nudge.

The lift metric is a useful signal, not a controlled experiment. If you want to isolate the causal effect of nudges, use the A/B testing feature (Growth+ plans), which randomly suppresses nudges for a control group so you can compare conversion rates under identical traffic conditions.

Single-touch attribution. elNudge uses a simple single-touch model: last nudge interaction within the 30-minute window gets credit. It does not account for other marketing touchpoints in the same session.

Purchase event dependency. If your store does not fire a PURCHASE event — or fires it incorrectly — conversions will not be tracked. Verify event firing in Dashboard → Settings → Events before drawing conclusions from conversion data.