Exit Intent

Design exit-intent popup strategies for ecommerce sites including trigger rules, offer types, copy variants, and A/B testing plans that recover leaving visitors without degrading UX or brand perception.

This skill designs end-to-end exit-intent popup strategies for ecommerce stores — deciding when to fire, what to offer, what copy and creative to show, and how to measure lift — so that recovery offers catch abandoning visitors without interrupting the ones who are still shopping or ready to check out.

Quick Reference

Decision	Strong	Acceptable	Weak
Trigger timing	Fire only after 15s+ on page AND cursor leaves viewport toward browser chrome	Fire on scroll-up past 60% of page	Fire immediately on page load or after < 5s
Offer type	Tiered by cart value (% off high carts, free shipping low carts)	Single universal discount code	Generic "Don't leave!" with no incentive
Copy length	Headline ≥ 8 words + 1-line value prop + single CTA	Headline + short paragraph + CTA	Wall of text with multiple competing CTAs
Mobile handling	Slide-up drawer triggered by back-button or tab-switch, respects scroll	Smaller modal with close button at top	Full-screen overlay blocking content, no close affordance
Frequency cap	1 show per session, s-day cookie cooldown	1 show per session, no cross-session memory	Shows on every page, no cap
A/B test design	Sequential test: trigger → offer → copy → creative, 1 variable at a time	Multivariate test with interaction analysis	Change everything at once with no control
SEO / CWV safety	Lazy-load popup JS, < 20 KB, no layout shift, defer until after LCP	Popup JS in main bundle but < 50 KB	Large third-party script blocking render, CLS > 0.1
Close behavior	Visible X button + click-outside-to-close + Esc key	Visible X button only	No close mechanism or hidden close button

Solves

1. High cart abandonment with no recovery system — stores losing 60-80% of carts with no popup or only a generic browser-default prompt.
2. Popups that annoy engaged shoppers — exit overlays firing too early, on wrong pages, or for users who are actively browsing or already at checkout.
3. Weak or generic popup copy — "Wait! Don't leave!" headlines that fail to provide a compelling reason to stay or convert.
4. No offer strategy by visitor segment — same 10%-off code for a first-time visitor with a $20 cart and a returning customer with $300 in cart.
5. Mobile popup compliance gaps — overlays violating Google's interstitial guidelines, breaking mobile UX, or failing to display correctly on small screens.
6. Unmeasured popup impact — no A/B testing, no control group, no tracking of incremental revenue vs. margin erosion from discounts.
7. Brand and trust damage — aggressive popup patterns (multiple overlays, fake urgency timers, dark-pattern close buttons) that erode brand trust over time.

Workflow

Step 1 — Audit Current State

Collect baseline metrics before designing the strategy:

• Current desktop and mobile bounce rates by page type (home, collection, product, cart, checkout)
• Cart abandonment rate (carts created vs. orders completed)
• Existing popup tools in use (Privy, Justuno, OptinMonster, Klaviyo forms, native Shopify)
• Google Search Console Core Web Vitals status (especially CLS and INP)
• Current popup frequency and trigger rules if any exist
• Revenue per session and average order value for context on offer sizing

Deliverable: Baseline metrics snapshot with annotated problem areas.

Step 2 — Define Trigger Rules

Design trigger conditions that balance recovery rate with UX:

Desktop triggers (use AND logic):

• Time on page ≥ 15 seconds (proves engagement before interrupting)
• Cursor exits viewport toward top of screen (true exit signal)
• Page type is product or cart (highest-intent pages)
• User has NOT already seen popup this session
• User has NOT completed a purchase in the last 30 days

Mobile triggers (use OR logic since no cursor):

• Back button or navigation-away gesture detected
• Browser tab switch after 30s+ on site
• Scroll-to-top after reaching 70%+ page depth (intent reversal signal)

Exclusions (never show):

• User is mid-checkout (past shipping step)
• User arrived via email campaign with existing discount
• User has active discount code in cart
• User dismissed popup within last 72 hours

Deliverable: Trigger rules matrix with conditions, logic operators, and exclusion list.

Step 3 — Design Offer Tiers

Create segmented offers based on cart value and visitor type:

Segment	Cart Value	Visitor Type	Offer	Rationale
High-value abandoner	> $150	Any	10% off or free express shipping	Protect margin, high motivation to convert
Mid-value abandoner	$50-150	Returning	Free standard shipping	Familiar with brand, shipping cost is friction
Mid-value abandoner	$50-150	New	15% off first order	Acquisition cost justifies deeper discount
Low-value browser	< $50	Any	Email signup for 10% future discount	Build list, don't discount low-margin orders
Cart page abandoner	Any	Any	Cart reminder + urgency (stock/price)	No discount needed — just remove friction

Deliverable: Offer tier matrix with segment definitions, offer values, and margin impact estimates.

Step 4 — Write Copy Variants

Create 3 copy variants per offer tier for A/B testing:

Structure for each variant:

• Headline (≤ 8 words, benefit-led or curiosity-driven)
• Subheadline (1 line, specifics of the offer)
• CTA button text (action verb + benefit)
• Dismissal text (low-pressure, no guilt)

Copy principles:

• Lead with what they GET, not what they're LOSING
• Use specific numbers ("15% off" not "a special discount")
• CTA should complete the sentence "I want to..."
• Dismissal text should be neutral ("No thanks, I'll pay full price" not "No, I hate saving money")

Deliverable: Copy matrix with 3 variants per tier, each with headline, subheadline, CTA, and dismiss text.

Step 5 — Specify Creative and Layout

Define visual specifications for the popup:

• Format: Centered modal on desktop (max 500px wide), bottom slide-up drawer on mobile (max 40% viewport height)
• Visual hierarchy: Headline → product image or lifestyle shot → offer details → CTA button → dismiss link
• Colors: CTA button uses brand primary, background uses white or light neutral, text uses brand dark
• Animation: Fade-in with subtle scale (200ms ease-out), no bounce or shake effects
• Close affordance: X button top-right (min 44px touch target), click-outside-to-close, Esc key support
• Accessibility: Focus trap within modal, aria-modal="true", role="dialog", aria-labelledby pointing to headline

Deliverable: Creative spec document with wireframe, color tokens, animation timing, and accessibility requirements.

Step 6 — Build A/B Testing Plan

Design sequential tests to isolate each variable:

Test 1 — Trigger sensitivity (Week 1-2):

• Control: No popup
• Variant A: Trigger at 15s + cursor exit
• Variant B: Trigger at 30s + cursor exit
• Primary metric: Incremental revenue per session
• Guardrail: Bounce rate increase < 2pp

Test 2 — Offer type (Week 3-4):

• Use winning trigger from Test 1
• Control: Free shipping
• Variant A: Percentage discount
• Variant B: Dollar-off discount
• Primary metric: Popup conversion rate
• Guardrail: Gross margin per converted session

Test 3 — Copy and creative (Week 5-6):

• Use winning trigger + offer
• Test 3 headline/CTA combinations
• Primary metric: Popup click-through rate
• Guardrail: Overall site conversion rate

Statistical requirements:

• Minimum 1,000 popup impressions per variant
• 95% confidence level, 80% power
• Run each test for minimum 2 full business cycles (14 days)

Deliverable: A/B test roadmap with timeline, variants, metrics, sample sizes, and decision criteria.

Step 7 — Monitor and Iterate

Set up ongoing monitoring dashboards and iteration cycles:

• Daily: Popup impression count, CTR, conversion rate, revenue attributed
• Weekly: Compare test variant performance, check guardrail metrics
• Monthly: Review offer margin impact, update copy for seasonal relevance, check CWV scores
• Quarterly: Full strategy review — retire underperforming variants, test new offer structures

Deliverable: Monitoring dashboard spec with KPIs, alert thresholds, and review cadence.

Examples

Example 1 — DTC Skincare Brand ($85 AOV, 72% Cart Abandonment)

Input: "We sell premium skincare DTC on Shopify. AOV is $85, cart abandonment is 72%. We have no exit popups. Our margins are 65%. We get 40K monthly sessions, 60% mobile."

Audit findings:

• No existing popup infrastructure
• Desktop bounce rate: 48% (acceptable), Mobile: 61% (high)
• Cart page drop-off: 38% of users who add to cart never reach checkout
• CWV scores: All green (LCP 1.8s, CLS 0.02, INP 120ms)
• No existing email popup or discount system

Trigger rules designed:

• Desktop: Time on page ≥ 20s AND cursor exit toward top AND page is product or cart AND no popup shown this session AND no purchase in 30 days
• Mobile: Back-button gesture after 25s+ on product page OR tab-switch from cart page
• Exclusions: Checkout pages, users with active promo code, users who dismissed within 72h, users arriving from Klaviyo email links

Offer tiers:

Segment	Trigger Page	Offer	Expected Margin Impact
Cart > $100	Cart page	Free express upgrade (normally $12)	-$12/order, ~14% conversion lift
Cart $50-100	Cart page	10% off	-$7.25/order avg, ~18% conversion lift
No cart, browsing	Product page	Join VIP list → 15% first order	$0 immediate, builds email list

Copy variant A (winning):

• Headline: "Your cart is waiting"
• Subheadline: "Complete your order in the next 15 minutes and get free express shipping"
• CTA: "Upgrade My Shipping — Free"
• Dismiss: "No thanks, standard shipping is fine"

Results after 6 weeks:

• Popup impression rate: 12% of sessions (correctly filtered)
• Popup CTR: 8.4%
• Incremental revenue: $18,200/month
• Cart abandonment: 72% → 64%
• CWV impact: None (popup JS: 14KB, lazy-loaded)

Example 2 — Electronics Accessories Store ($42 AOV, High Mobile Traffic)

Input: "We sell phone cases and accessories on WooCommerce. AOV is $42, margins are 45%. 78% mobile traffic. We tried a popup before but it hurt our Google rankings."

Audit findings:

• Previous popup was a full-screen interstitial on mobile (violates Google guidelines)
• CLS jumped to 0.18 when popup was active (failing CWV threshold)
• Desktop bounce: 52%, Mobile bounce: 67%
• Cart abandonment: 68%
• Previous popup had no frequency cap — showed on every page load

Trigger rules designed:

• Desktop: Time ≥ 15s AND cursor exit AND product or cart page AND session popup count = 0
• Mobile: NO full-screen overlay. Instead: sticky bottom bar (non-intrusive) that appears after 30s on product page, collapses to icon after 5s if not engaged
• Exclusions: Checkout, returning purchasers within 14 days, users who clicked dismiss, search-engine bot user agents

Mobile-specific design:

• Format: Bottom sheet, 30% viewport height, slides up with 250ms ease
• Does NOT cover main content — user can continue scrolling
• Complies with Google's mobile interstitial guidelines (not penalized)
• Touch target: 48px minimum for CTA and close

Offer strategy (low margins require caution):

Segment	Offer	Margin Impact
Cart > $60	Free shipping ($5.99 value)	-$5.99/order, justified by higher AOV
Cart $25-60	"Add $X more for free shipping" progress bar	$0 cost, increases AOV
Browsing, no cart	"Get notified when this drops in price" email capture	$0 cost, builds list

Copy variant B (winning):

• Headline: "Free shipping unlocked"
• Subheadline: "Your cart qualifies — checkout now to lock it in"
• CTA: "Checkout with Free Shipping"
• Dismiss: "Maybe later"

Results after 4 weeks:

• Mobile bounce rate: 67% → 63%
• CWV: CLS back to 0.03 (green), no ranking impact
• Popup engagement: 6.2% CTR
• Incremental revenue: $4,800/month
• AOV lift: $42 → $47 (from "add more for free shipping" bar)

Common Mistakes

1. Firing popups too early — Showing an exit popup within 5 seconds tells the visitor you expect them to leave. Wait at least 15 seconds to ensure they've had time to engage with the page content.

2. Using full-screen interstitials on mobile — Google penalizes intrusive interstitials on mobile in search rankings. Use bottom sheets, slide-up drawers, or sticky bars instead of overlays that cover the main content.

3. Offering the same discount to every visitor — A blanket 15%-off code trains all visitors to expect discounts and erodes margins. Segment offers by cart value, visitor type, and page context.

4. No frequency cap — Showing the popup on every page visit creates popup fatigue and actively pushes visitors away. Cap to 1 impression per session with a 72-hour cross-session cooldown.

5. Testing everything at once — Changing the trigger, offer, copy, and design simultaneously makes it impossible to know what drove the result. Test one variable at a time in sequential experiments.

6. Ignoring Core Web Vitals impact — Heavy popup scripts, layout shifts from overlay rendering, and render-blocking CSS can tank CWV scores. Lazy-load popup code, keep the bundle under 20KB, and ensure zero CLS.

7. Dark-pattern dismiss buttons — Tiny close buttons, guilt-trip dismissal copy ("No, I hate saving money"), or fake countdown timers damage brand trust and may violate consumer protection regulations in some jurisdictions.

8. Not measuring incrementality — Tracking popup conversion rate without a holdout control group overstates impact. Many popup converters would have purchased anyway. Always run with a no-popup control to measure true incremental revenue.

9. Showing popups to returning purchasers — A customer who bought last week doesn't need a discount popup. Exclude recent purchasers and let loyalty programs handle retention offers.

10. Forgetting accessibility — Popups without focus trapping, keyboard navigation, screen-reader labels, or sufficient color contrast exclude users and may create legal liability under ADA/WCAG requirements.

Resources

• Output Template — Structured template for the complete exit-intent strategy deliverable
• Trigger Rules Guide — Deep-dive into trigger logic, device-specific patterns, and exclusion rules
• Offer Strategy Guide — Frameworks for segmenting offers by visitor behavior and cart context
• Quality Checklist — Pre-launch checklist covering UX, compliance, performance, and measurement