How We QA and Test Every Shopify Store Before Launch

A Shopify store that crashes on launch day, drops orders, or renders incorrectly on mobile is not just embarrassing — it is expensive. Every hour a store is broken costs revenue, erodes customer trust, and creates a recovery effort that consumes far more time and energy than proper testing would have required.

We have a rule: no store launches without passing our full QA process. No exceptions. Not for tight deadlines, not for eager clients, not for "we will fix it after launch." After two decades of ecommerce, we have learned that the cost of fixing issues after launch is always higher — in money, time, and reputation — than catching them beforehand.

This article documents our complete QA and testing process. We publish it for two reasons. First, transparency. If you are considering working with us, you should know the rigour we apply. Second, standards. If you are working with another agency, this gives you a benchmark for what thorough testing looks like.

Why QA testing is non-negotiable

The commercial case for thorough QA is straightforward. Consider the cost of these common post-launch issues:

• Checkout failure: If 10% of customers encounter a checkout error on launch day and abandon, and you would have processed 200 orders at a £60 AOV, that is £1,200 in lost revenue — in a single day.
• Mobile rendering issue: A broken mobile layout on a store where 70% of traffic is mobile effectively makes the store unusable for the majority of visitors.
• Missing tracking: If your analytics are not configured correctly, you launch blind — unable to measure performance or optimise campaigns.
• SEO errors: Missing redirects from a platform migration can cause immediate ranking drops that take months to recover.

Every one of these scenarios is preventable through proper QA. And every one of them, we have seen happen to stores that launched without it.

Functional testing: The purchase flow

The purchase flow is the spine of any ecommerce store. If a customer cannot browse, add to cart, and complete a purchase without friction, nothing else matters. Our functional testing covers every step of this journey.

Product browsing and discovery

• All collection pages load correctly with proper filtering and sorting
• Product search returns accurate results and handles edge cases (misspellings, partial matches, zero results)
• Collection filters work correctly across all filter types (size, colour, price, custom attributes)
• Pagination loads correctly and does not duplicate or skip products
• Product cards display correct information (price, variant availability, sale pricing)

Product pages

• All variants display correctly with accurate pricing
• Out-of-stock variants are handled appropriately (disabled, hidden, or shown with messaging)
• Image galleries function correctly — all images load, zoom works, variant-specific images switch correctly
• Size guides, material information, and other supplementary content displays correctly
• Related products and recommendations populate accurately
• Structured data (product schema) validates correctly

Cart and checkout

This is where we spend the most time testing because the commercial impact of failures here is highest:

• Products add to cart with correct variants, quantities, and pricing
• Cart updates (quantity changes, removals) calculate correctly
• Discount codes apply correctly — percentage off, fixed amount, free shipping, buy-one-get-one, automatic discounts, and combinations
• Shipping rates calculate correctly for all destinations and weight brackets
• Tax calculations are accurate for all applicable regions
• All payment methods process successfully (credit card, PayPal, Apple Pay, Google Pay, Shop Pay, Klarna)
• Order confirmation page displays correct order details
• Confirmation email sends with accurate information

We test checkout using Shopify's Bogus Gateway for initial testing, then validate with real payment processors using test card numbers. We process and refund actual test transactions to verify the complete payment flow.

Edge case testing

Beyond the happy path, we test scenarios that users will inevitably encounter:

• Adding more quantity than available stock
• Applying expired or invalid discount codes
• Checking out with the minimum order value (if applicable)
• Attempting to ship to restricted locations
• Session timeout during checkout
• Multiple tabs open with different cart states
• Back button behaviour during checkout
• Currency switching mid-session (for multi-currency stores)

Cross-browser and device testing

A store that works perfectly in Chrome on a MacBook but breaks on Safari on an iPhone has not been properly tested. We test across a comprehensive device and browser matrix.

Browser coverage

Browser	Versions tested	Platform
Chrome	Current + previous two versions	Windows, macOS, Android
Safari	Current + previous version	macOS, iOS
Firefox	Current + previous version	Windows, macOS
Edge	Current version	Windows
Samsung Internet	Current version	Android

Device testing

We test on actual physical devices, not just browser emulators. Emulators miss issues with touch interactions, viewport rendering, and native browser behaviours. Our test device set includes:

• iPhone (current and previous generation) — covering different screen sizes and iOS versions
• iPad (standard and Pro sizes)
• Samsung Galaxy (current generation) — the most popular Android device family in the UK
• Google Pixel — for stock Android testing
• Desktop at 1920x1080, 1440x900, and 2560x1440
• Ultrawide monitor (3440x1440) to test maximum width behaviour

For each device, we test the complete purchase flow, navigation interactions, form inputs, and visual layout. Touch targets need to be appropriately sized (minimum 44x44 pixels), swipe gestures need to work naturally, and the keyboard should not obscure form fields during input.

Performance testing and optimisation

Performance testing is not just about running a PageSpeed Insights scan and hoping for a good score. It is a systematic evaluation and optimisation process.

Our performance targets

// Minimum performance standards
const targets = {
  mobilePageSpeed:  85,      // Minimum score
  desktopPageSpeed: 95,      // Minimum score
  LCP:              '2.5s',  // Largest Contentful Paint
  TBT:              '200ms', // Total Blocking Time
  CLS:              0.1,     // Cumulative Layout Shift
  FCP:              '1.8s',  // First Contentful Paint
  TTFB:             '600ms', // Time to First Byte
};

What we test and optimise

• Image optimisation: All images compressed, served in modern formats (WebP/AVIF), properly sized for their containers, and lazy loaded below the fold
• JavaScript audit: Identifying and removing unused JavaScript, deferring non-critical scripts, and evaluating the performance impact of every app installed
• CSS optimisation: Removing unused styles, inlining critical CSS, and ensuring render-blocking stylesheets are minimised
• Font loading: Preloading critical fonts, using font-display: swap, and limiting the number of font weights and styles loaded
• Third-party script impact: Measuring the performance impact of each third-party script (chat widgets, reviews, analytics) and implementing loading strategies that minimise their impact on Core Web Vitals

We test performance on throttled connections (simulating 3G and 4G) because real-world mobile performance is significantly worse than what you see on a fast office connection. A store that takes three seconds to load on your Wi-Fi might take eight seconds on a customer's mobile connection.

Performance work during QA is not about achieving perfect scores — it is about ensuring the store meets the thresholds that matter for both SEO rankings and conversion rates. Studies consistently show that each additional second of load time reduces conversion rate by 7-12%.

SEO validation

Launching a store with SEO issues means starting from a disadvantaged position in search. Our SEO validation covers both technical foundations and content optimisation.

Technical SEO checks

• Meta tags: Every page has a unique, properly formatted title tag and meta description
• Heading hierarchy: Single H1 per page, logical heading structure
• Canonical tags: Correct self-referencing canonicals, proper canonicalisation for variant URLs and paginated collections
• XML sitemap: Auto-generated, includes all indexable pages, excludes non-indexable pages
• Robots.txt: Properly configured to allow crawling of important pages and block irrelevant ones
• Structured data: Product, BreadcrumbList, Organisation, and FAQ schema validated against Google's requirements
• Redirects: All old URLs redirect to their equivalents (critical for migrations)
• Internal linking: Key pages are properly linked, no orphaned pages, breadcrumbs function correctly
• Image alt text: All images have descriptive, keyword-relevant alt text
• 404 handling: Custom 404 page with helpful navigation

For migration projects, redirect validation is particularly critical. We test every redirect from the old URL structure to the new one, verify that redirect chains do not exceed two hops, and confirm that no important pages return 404 errors. The work we do during project discovery identifies all the URLs that need redirecting, and QA verifies they are all in place.

Accessibility testing

Accessibility is not optional. Beyond the ethical imperative of making your store usable by everyone, the Equality Act 2010 requires service providers (including online retailers) to make reasonable adjustments for disabled users. Accessibility failures also impact SEO, as many accessibility best practices overlap with search engine requirements.

Our accessibility testing covers

• Keyboard navigation: Every interactive element is reachable and operable via keyboard alone. Tab order follows a logical sequence. Focus states are clearly visible.
• Screen reader compatibility: Page structure is announced correctly. Form labels are properly associated. Images have meaningful alt text. Dynamic content changes are announced.
• Colour contrast: Text meets WCAG AA minimum contrast ratios (4.5:1 for normal text, 3:1 for large text). This applies to all states — default, hover, focus, and active.
• Form accessibility: All form fields have visible labels, error messages are descriptive and associated with the relevant field, and required fields are clearly indicated.
• Motion and animation: Animations respect the prefers-reduced-motion media query. No content relies solely on animation to convey information.
• Touch targets: Interactive elements are at least 44x44 pixels with adequate spacing between them.

We use a combination of automated testing (axe DevTools, Lighthouse accessibility audit) and manual testing (keyboard-only navigation, screen reader testing with VoiceOver and NVDA). Automated tools catch approximately 30-40% of accessibility issues — the rest require human evaluation.

Analytics and tracking validation

Launching a store without validated analytics is like opening a shop with the lights off — you cannot see what is happening. Our analytics validation ensures every tracking system is configured correctly and collecting accurate data.

What we validate

• GA4 enhanced ecommerce: All ecommerce events fire correctly with proper parameters — view_item, add_to_cart, begin_checkout, purchase, and refund
• Conversion tracking: Google Ads, Meta Pixel, and other advertising platform conversions fire accurately with correct revenue values
• Server-side tracking: Meta Conversions API, Google Enhanced Conversions, and any other server-side implementations are sending data correctly
• Email platform integration: Klaviyo or other email platforms are receiving customer and order data for flow triggers
• UTM parameter handling: UTM parameters are preserved through the checkout flow and appear correctly in GA4
• Internal traffic exclusion: The client's IP addresses and internal traffic are properly filtered
• Consent management: Cookie consent banner functions correctly, analytics only fire after consent is granted, and consent preferences are respected

We validate analytics by processing test transactions and verifying that the correct data appears in each platform. We check GA4's DebugView, Meta's Test Events tool, and Klaviyo's activity feed to confirm end-to-end data flow. This validation process connects directly to the broader analytics setup framework we implement for every store.

Security and compliance checks

Security and compliance issues can have serious legal and financial consequences. Our pre-launch security review covers:

• SSL certificate: Active and correctly configured across all pages
• Password protection: Development password removed, any gated sections properly secured
• Admin access: Temporary development accounts removed, client admin accounts properly configured with appropriate permission levels
• Privacy policy: Published and linked from footer and checkout, covering GDPR requirements including data processing, cookie usage, and customer rights
• Terms and conditions: Published and accessible, with checkbox at checkout if required
• Cookie consent: Banner compliant with PECR regulations, properly blocking non-essential cookies before consent
• Payment security: PCI compliance confirmed (handled by Shopify's infrastructure, but custom implementations need verification)

The pre-launch checklist

Everything above feeds into a structured pre-launch checklist that must be completed and signed off before we remove the password and point the domain. Here is a summary view:

Category	Items	Status required
Functional	Purchase flow, discounts, shipping, payments, emails	All pass
Cross-browser	Chrome, Safari, Firefox, Edge + mobile browsers	No critical issues
Performance	PageSpeed 85+ mobile, Core Web Vitals pass	Targets met
SEO	Meta tags, schema, redirects, sitemap, robots.txt	All validated
Accessibility	WCAG 2.1 AA keyboard, screen reader, contrast	No critical issues
Analytics	GA4, advertising pixels, email integration	All verified
Security	SSL, privacy policy, cookie consent, admin access	All compliant
Content	All pages proofread, images optimised, links working	All reviewed

The checklist is a living document. We update it after every project based on issues we encounter. If something slips through on one project, it gets added to the checklist for every project that follows. This continuous improvement approach means our QA process gets more comprehensive over time.

Post-launch monitoring

QA does not end at launch. For the first 48 hours after going live, we actively monitor:

• Real-time analytics for anomalies in traffic patterns or conversion rates
• Server response times and error rates
• Order processing to confirm transactions are flowing correctly
• Email delivery to verify transactional emails are sending
• Search Console for crawl errors or indexation issues

We are available for immediate response during this monitoring period. If an issue emerges, we address it within minutes, not hours. This rapid response capability is part of our commitment to a smooth launch — because the reality of ecommerce is that some issues only manifest under real traffic conditions, no matter how thorough the pre-launch testing.

---

Thorough QA testing is one of the things that separates professional Shopify development from amateur work. It is not glamorous. Clients rarely see the hundreds of test cases we run. But the absence of problems on launch day — that smooth, confident, revenue-generating launch — is the direct result of this work.

If you are planning a Shopify launch and want the confidence that comes from rigorous testing, let us have a conversation. We will show you exactly how our QA process works and how it protects your investment.