Shopify Audit

intent

audit your shopify storefront (homepage, collection pages, product detail pages, checkout) for hidden conversion blockers that kill add-to-cart and checkout rates. run this when paid traffic has good clickthrough but low conversion, before major sales events (bfcm, product launches), after a redesign or theme change, or when add-to-cart sits below 5%. outputs a prioritized, executable fix list your team can implement in the theme editor or app ecosystem without a developer.

inputs

required:

• shopify store url (e.g., glowlab.shopify.com or custom domain)
• top product or collection url (the primary audit target)
• current conversion rate or add-to-cart rate (even rough estimates calibrate findings)

strongly recommended:

• device split (% mobile vs. desktop; most stores run 70%+ mobile)
• top traffic source (paid social, search, email, direct)
• current checkout completion rate if available

external connections:

• shopify theme admin access (read-only sufficient; no oauth required if you have store credentials)
• optional: google pagespeed insights api key (env var: PAGESPEED_API_KEY) for automated performance measurement
• optional: heatmap tool (hotjar, microsoft clarity, or native shopify analytics) to corroborate scroll-depth findings

edge cases:

• if store uses custom checkout (shopify plus): audit still applies but checkout section is less actionable (custom build is team's responsibility)
• if store blocks bots (cloudflare challenge, etc.): performance testing may require manual lighthouse runs
• if conversion rate data unavailable: weight findings by funnel position and typical user behavior patterns instead

procedure

step 1: collect inputs and establish baseline

gather store url, top product/collection url, device split, traffic source, and current conversion or add-to-cart rate. if metrics are missing, document which ones and note in the report that prioritization is estimated rather than measured. baseline metrics calibrate what "good" looks like: a store with 80% mobile paid-social traffic gets audited differently than one with desktop search traffic.

output of step 1: baseline summary document with store profile, traffic composition, and current funnel metrics.

step 2: homepage first-impression audit (5-second test)

load the homepage and assess above-the-fold without scrolling. answer: can a stranger identify what the store sells, for whom, and why here in 5 seconds?

check these elements:

• headline: benefit-led, 8 words or fewer, not "welcome to our store"
• hero image: product in use or lifestyle context, not a generic stock photo
• primary cta: single, above-the-fold, action-oriented (not "learn more")
• announcement bar: clarity (e.g., "free shipping over $45" beats vague slogans like "limited time offer")
• navigation depth: 6 or fewer top-level items; drop-downs kept shallow
• hero load weight: measure with pagespeed insights; banner image should be <150kb

output of step 2: list of homepage findings with location, evidence, and estimated friction (high/medium/low).

step 3: collection page browsability audit

if the store has >12 products in a category, check for:

• product card sufficiency: each card shows price, variant selector or cues (color swatches), and review stars visible at default size
• filters and sort: present and functional; no orphaned filters
• image consistency: all product cards use same image dimensions and quality
• pagination vs. infinite scroll: appropriate for catalog size (infinite scroll loses pagination context; pagination loses momentum on large catalogs)
• empty-state handling: if a filter returns zero products, page shows helpful message or refine options, not blank space

output of step 3: collection page findings with card-level and filter-level issues.

step 4: product detail page persuasion audit (core audit section)

audit in this order of visual hierarchy:

title and pricing:

• is title scannable and benefit-focused (not generic: "blue widget" vs. "waterproof blue widget, 40% lighter")?
• is price clearly displayed with compare-at framing if discounted?

image set:

• minimum 5 images present?
• zoomable on desktop and mobile?
• mix of product-only and lifestyle shots?
• at least one image with scale reference (hand, model wearing, size comparison)?

description structure:

• opens with benefit, not brand history?
• scannable (bullets or short paragraphs, not wall of text)?
• addresses 3+ common objections (durability, fit, shipping, returns)?

social proof placement:

• review count and stars visible without scrolling on mobile?
• rating block appears in top half of pdp, not after fold?

shipping and returns:

• shipping timeline and cost visible near add-to-cart button?
• return policy accessible without clicking out to a separate page?

variant picker:

• all variants clearly labeled (not "option 1, option 2")?
• sold-out variants grayed out, not hidden (prevents cart abandonment from invisible stock)?

add-to-cart button and sticky placement:

• mobile: sticky atc button below header, reachable with thumb, not covered by other sticky elements?
• desktop: atc in first viewport, not below mid-fold?
• button text action-oriented ("buy now" or "add to cart", not "submit")?

urgency signals:

• countdown timers present? (flag as trust-risk: reset behavior and fomo manipulation invite regulatory pushback)
• stock scarcity messaging (e.g., "only 3 left") truthful and up-to-date?

cross-sell and recommendations:

• placed below the fold (not blocking atc)?
• related or complementary products shown, not random?

output of step 4: 8-12 pdp findings prioritized by impact, each with location, evidence, and fix.

step 5: cart and checkout friction audit

cart page/slide-out:

• accessibility: is cart a slide-out, page, or drawer? (slide-out is fastest for quick add-ons)
• add-on upsells: present and relevant, or dead weight slowing the flow?

pre-checkout messaging:

• shipping cost visible in cart? (surprise shipping at checkout is the #1 abandonment driver)
• shipping calculator or threshold messaging? (e.g., "free shipping over $45" or "estimated shipping: $6.95 at checkout")
• stock status for cart items visible?

checkout settings:

• guest checkout enabled in shopify payments settings? (required accounts tank conversion 10-30%)
• express payment options enabled? (shop pay toggle on, apple pay, google pay available)
• discount code field prominent? (flagged if actual bfcm codes are invalid or off-site; recommend automatic discounts instead)

checkout form:

• form field count minimal? (each field loses 10-25% of remaining users; pre-fill what shopify knows)
• address validation active? (reduce typos and failed charges)

payment step:

• trust badges visible (ssl, payment provider logos, return guarantees)?
• order summary clear and final total visible before submit?

note: shopify's standard checkout is locked and well-optimized. audit wins come from upstream (shipping expectations, express payments, guest checkout enabled) rather than checkout design itself.

output of step 5: 5-8 cart/checkout findings, with emphasis on surprise-cost and express-payment items.

step 6: performance and mobile-specific audit

performance measurement (pagespeed insights or lighthouse):

• record largest contentful paint (lcp) on pdp: target <2.5s
• check image formats: webp or avif preferred; jpg fallback acceptable
• app script audit: list all installed apps (reviews, upsells, chat, loyalty, etc.); each adds 20-50kb and delays load. mark unused apps for uninstall; lazy-load the rest
• font count: >4 custom fonts often slow render; aim for 2-3

mobile-only pass (separate from desktop audit):

• sticky add-to-cart button: does it stay visible, reachable by thumb (bottom 33% of screen), and not cover essential content?
• text readability: no zoom required to read body copy? (font size >=16px for readability)
• tap-target spacing: all buttons and links 48x48px or larger, with 8px margins between?
• sticky header/navigation: does it cover atc button or image gallery?
• image weight on 4g: realistic load time for buyers on slower connections

output of step 6: performance and mobile findings with specific metrics (lcp time, app list, image formats).

step 7: compile prioritized, executable report

rank all findings from steps 2-6 by estimated revenue impact (traffic to that page × severity of friction) against implementation effort (time for team to execute in theme editor or app settings).

impact tiers:

• high: affects >30% of checkout flow (e.g., guest checkout disabled, surprise shipping costs)
• medium: affects 10-30% (e.g., review stars below fold, missing express payments)
• low: affects <10% (e.g., font size, color contrast)

effort tiers:

• low: <1 hour in theme editor or app toggle (e.g., enable feature in theme settings, rewrite headline)
• medium: 1-4 hours (e.g., edit template, install new app, redesign card layout)
• high: >4 hours or requires developer (e.g., custom code, theme rewrite)

cap the list at 15 findings. a 50-item dump is a backlog, not a roadmap. include only findings with specific, actionable fixes.

per finding, provide:

• location (e.g., "pdp, above-the-fold")
• evidence (screenshot, heatmap data, metric, or user behavior pattern)
• estimated impact (% conversion lift if fixed)
• exact fix (rewritten copy, theme-editor steps, or app setting path)

output of step 7: prioritized report document (see output contract).

step 8: quality check and schedule re-audit

verify the report against assets/audit-quality-checklist.md: all findings name buyer behavior, include actionable fixes, and are ranked. confirm baseline metrics are documented so improvement can be measured post-fix.

close the report with a re-audit recommendation: quarterly or pre-peak season (bfcm, summer, etc.). themes update, apps change, inventory shifts; one-time audits grow stale.

output of step 8: final report with quality sign-off and re-audit calendar entry.

decision points

if device split is unknown or unavailable: weight findings by typical shopify behavior (70%+ mobile assumed). note in report that mobile-first prioritization is structural, not data-driven. request analytics before re-audit.

if conversion rate unavailable: estimate impact by funnel position (homepage findings affect broad traffic; pdp findings affect committed browsers). flag this limitation in the report , recommend baseline measurement to prove improvement post-fix.

if store uses custom checkout (shopify plus): audit checkout findings as recommendations and escalate to dev team. focus the bulk of findings on upstream friction (homepage, pdp, cart, shipping messaging) that your team controls.

if pagespeed api unavailable or blocked: use manual lighthouse runs (chrome devtools > lighthouse tab). record lcp, cls, and fid metrics manually. note that automated testing was not available.

if review app shows zero reviews on pdp: flag as high-impact friction (no social proof = trust gap). recommend seeding reviews via email campaign or review request at post-purchase. if reviews exist but app not installed, install and enable.

if guest checkout is disabled: this is almost always the single highest-impact fix. verify with the store owner that it's intentional (rare). if not, enable and re-test. expected conversion lift: 10-30%.

if countdown timers are present: flag for removal, not optimization. timers that reset are manipulative and risk ftc/legal scrutiny. recommend stock-count messaging instead ("3 left in stock") if truthful.

if surprise shipping cost appears at checkout: recommend announcement bar + cart progress meter showing free-shipping threshold. this is the #1 checkout abandonment driver on shopify stores.

if store traffic is <500 sessions/month: recommend deferring performance optimization. focus on high-impact, low-effort copy and trust-signal fixes instead. revisit performance when traffic justifies the app bloat cost.

output contract

deliver a single markdown report document with these sections:

# Shopify Audit Report: [Store Name]

## executive summary
- store url
- audit date
- device split (% mobile/desktop)
- top traffic source(s)
- current conversion rate and add-to-cart rate (or note if unavailable)
- 1-sentence thesis (e.g., "surprise shipping costs and missing guest checkout are the highest-impact fixes")

## findings (prioritized, max 15)

### finding 1: [location + brief title]
**impact:** high | medium | low
**effort:** low | medium | high
**evidence:** [screenshot, metric, or heatmap data]
**fix:** [exact copy rewrite, theme-editor steps, or app setting path]

[repeat for each finding]

## app audit
- [list installed apps]
- [mark unused for uninstall]
- [note any scripts that slow load]

## performance metrics
- lcp (largest contentful paint): [time]s
- image formats: [webp/jpg/avif status]
- font count: [number]

## mobile-specific notes
[thumbreach, tap-target, sticky element findings]

## implementation roadmap
[order findings by impact/effort; suggest 1-2 week sprint to tackle high-impact, low-effort items]

## re-audit schedule
[recommend next audit date: quarterly or pre-peak season]

data format: markdown (.md) or google doc with same sections. file location: shared with store owner in drive or tool of choice.

outcome signal

the skill worked if:

• report is delivered within 1-2 days of start
• all findings include actionable fixes, not vague recommendations
• at least 50% of findings are high-impact, low-effort items the team can fix this week
• baseline metrics (conversion rate, device split, traffic source) are documented
• store team can implement the top 3-5 fixes without a developer
• after implementing fixes, the team re-runs the audit in 2-3 weeks and measures conversion lift

success is not a beautiful audit report. success is a 5-15% conversion lift within 30 days of implementing the top 5 fixes.