by Dream Intrepid Ltd

Why Canadian E-Commerce Sites Lose Sales to Slow Images (And the Fix Takes 20 Minutes)

By Optimage

Canadian online shoppers abandon carts at a higher rate than the global average when pages take more than 3 seconds to load. This guide shows exactly how image weight is the leading cause and what Canadian store owners on Shopify, WooCommerce, and Squarespace can do about it right now.

Table of Contents

Here is a number that should make any Canadian store owner uncomfortable: according to a Google and Deloitte study on mobile retail performance, a 0.1-second improvement in mobile site speed increases conversion rates by 8.4% for retail sites. Flip that around. For every tenth of a second your store is slower than it should be, you are losing nearly one in ten potential sales. Now think about how many tenths of a second your product images are adding to your load time right now.

Canada is one of the top ten countries globally for online retail spending per capita. Statistics Canada reported that Canadian e-commerce sales exceeded $50 billion CAD in 2023, a figure that has continued climbing. Canadians shop online more than the global average, and they do a significant portion of that shopping on mobile devices. That combination creates an urgent problem for any Canadian store owner who has not paid close attention to image file sizes: your customers are loading large, unoptimized product photos on mobile connections that are often slower than you think, and many of them are leaving before the page finishes rendering.

The good news is that image optimization is one of the few performance improvements you can complete in an afternoon without a developer. This guide walks through everything: why images are specifically the problem, how Canadian mobile infrastructure makes it worse, what to do on each major platform, and a realistic before-and-after example from a fictional but data-accurate Toronto Shopify store.

Canada's E-Commerce Boom Has a Performance Problem {#canadas-ecommerce-boom-has-a-performance-problem}

Let's start with the market context. Canada's e-commerce sector is not just large, it is disproportionately large relative to population. eMarketer's data on Canadian digital commerce consistently places Canada among the highest-spending nations on a per-capita basis, driven by high smartphone penetration (around 91% of Canadians own a smartphone), strong internet coverage in urban centers, and a culture of cross-border online shopping that has pushed domestic retailers to compete on experience and speed.

But "compete on experience" is easier said than done when most Canadian independent store owners are running Shopify or WooCommerce stores they set up themselves, uploading product photos directly from their iPhone or DSLR without any optimization step in between. The result is product pages loaded with 4MB JPEGs, uncompressed PNGs from design tools, and hero images that are completely punishing on a mobile device in Mississauga, let alone in Kamloops or Moncton.

According to the HTTP Archive's Web Almanac, the median page weight for e-commerce sites in 2025 was approximately 3.4MB on desktop, with images accounting for roughly 50-60% of that total. On mobile, where Canadians increasingly shop, the problem is compounded by smaller screens that still load the same large files unless explicit responsive image logic is in place.

The abandonment numbers are stark. Research from Statista and various conversion rate optimization studies consistently shows that 53% of mobile users abandon a site that takes longer than 3 seconds to load. For e-commerce specifically, Walmart famously documented that a 1-second improvement in load time correlated with a 2% increase in conversion. These are not hypothetical numbers. They are the direct, measurable cost of shipping large images to your customers.

The specific shape of Canada's internet infrastructure makes this worse than it looks in the headline statistics, and that is worth understanding in detail before we talk about solutions.

Canadian independent retailers are competing not just against each other but against Amazon.ca, which has invested billions in performance optimization. When a customer's baseline comparison is Amazon loading in under 1 second, your 5-second product page is not just slow, it is a competitive liability. Every dollar spent on marketing to bring a customer to a slow page is partially wasted.

The encouraging reality is that the gap between Amazon's performance engineering and what a Shopify store owner can achieve has narrowed considerably. You do not need a team of engineers. You need to stop uploading large raw photos and start uploading optimized ones. The tools to do this are inexpensive, fast, and require no technical knowledge.

Why Images Are the Biggest Culprit {#why-images-are-the-biggest-culprit}

When a Google PageSpeed Insights audit flags your store for performance issues, image-related warnings almost always appear at the top of the list: "Serve images in next-gen formats," "Efficiently encode images," "Properly size images," "Defer offscreen images." These are not cosmetic suggestions. They represent real bytes being transferred over the network before your customer can see or interact with your product pages.

The mechanics work like this. A customer on their Bell LTE connection in Toronto clicks on your Google Shopping ad for a handmade candle. Your product page begins loading. Before the customer can add the item to cart, their browser needs to download your HTML, your CSS, your JavaScript, and your images. If your hero product image is a 5.2MB JPEG exported directly from Lightroom, the browser is spending the first 2-4 seconds on that image alone, depending on the customer's connection speed and network conditions at that moment.

Images are the single largest contributor to page weight on most e-commerce sites. Unlike JavaScript (which browsers can partially parse and execute in parallel), large images are a straightforward blocking download. The browser knows it needs to display the image to render the page layout, so it waits.

The format problem compounds the size problem. Most Canadian store owners are uploading JPEGs from their cameras or PNGs from Canva. In 2026, both of these formats are significantly less efficient than WebP or AVIF, which can achieve the same perceived visual quality at 25-50% smaller file sizes. A 2MB JPEG hero image can often be converted to a WebP at 600KB with no perceptible quality loss at normal viewing distances. That is 1.4MB saved on every single page load, for every single visitor, on every single product. See our detailed comparison of AVIF vs WebP vs JPEG for benchmark data on exactly how much you can save.

The Largest Contentful Paint metric, which Google uses as a Core Web Vitals signal affecting your search rankings, is almost always the product hero image on an e-commerce page. Fail that metric and you are paying a compounding tax: slower load times that hurt conversions AND lower search rankings that reduce the traffic arriving at those pages in the first place. For a deeper look at why LCP failures happen and how to fix them, see our dedicated guide.

Consider the scale of the problem across a typical mid-sized Canadian Shopify store. Say you have 100 products with 6 images each: 600 product images. If your average product image is 3MB (a conservative estimate for unoptimized camera files run through minimal editing), your total product image library weighs 1.8GB. After proper compression, that same library might weigh 150-200MB. Every one of those 600 images is being loaded by customers across hundreds of product and collection pages every day. The cumulative bandwidth cost, and the cumulative load time burden on your customers' devices, is enormous.

There is also a less obvious impact: Google's crawl budget. Googlebot crawls your site periodically to index your pages. Slow pages consume more crawl budget, meaning Google crawls fewer of your pages per session. For stores with large catalogs, this can affect how quickly new products appear in search results. Faster pages, driven in part by optimized images, get crawled more efficiently.

The Canadian Mobile Reality: Coverage Gaps and Slow Connections {#the-canadian-mobile-reality-coverage-gaps-and-slow-connections}

Canada has a specific mobile infrastructure challenge that most US-centric performance guides do not account for: geography. Canada is the second-largest country in the world by land area, with a population of roughly 40 million people concentrated in a relatively narrow band within 300 kilometers of the US border. Outside of the major urban corridors (Vancouver, Calgary, Edmonton, Winnipeg, Toronto, Ottawa, Montreal, Halifax), mobile coverage quality drops sharply.

Rogers, Bell, and Telus, the three dominant carriers, have all invested heavily in 5G rollout in dense urban areas. In downtown Toronto or downtown Vancouver, your customers may well be on 5G connections with peak download speeds exceeding 500 Mbps. This is not the average Canadian mobile reality.

The CRTC's annual Communications Monitoring Report consistently shows that rural and remote Canadians have significantly worse broadband access than urban Canadians. Roughly 60% of rural Canadian households have access to broadband meeting the CRTC's basic service objective of 50 Mbps download, compared to over 90% of urban households. For mobile specifically, LTE coverage in rural areas is often intermittent, and 3G fallback is still a real scenario in many parts of the country.

This matters for e-commerce because rural Canadians are not excluded from e-commerce. In many cases, they are among the most enthusiastic online shoppers because physical retail options are limited. A customer in a small town in northern Ontario or rural New Brunswick, shopping for specialty goods that are not available locally, is exactly the customer you want. That customer may be on a 10-15 Mbps LTE connection with variable signal quality. Your 5MB product page is going to take 3-5 seconds to load at best, and they will likely bail before it finishes.

The fix is not to build a separate "rural version" of your store. The fix is to optimize your images to a standard that serves everyone well, including customers on slower connections. A properly optimized product page with WebP images, appropriately sized for mobile viewports, and correctly compressed, should load in under 2 seconds even on a throttled LTE connection. That is an achievable target for any store, on any platform.

The mobile share of Canadian e-commerce traffic is significant and growing. Industry data from Shopify consistently shows that more than 70% of traffic to Shopify stores originates from mobile devices, and Canada closely tracks this pattern. You are building primarily for mobile users. Optimize accordingly.

There is also a seasonal dimension specific to Canada. Holiday shopping peaks (Black Friday, Cyber Monday, Boxing Day) create simultaneous high-traffic periods on cellular networks in urban areas. Network congestion during peak periods can effectively reduce available bandwidth to users even in cities that normally have excellent LTE coverage. An optimized store loads adequately even under congested network conditions. An unoptimized store may become functionally unusable precisely when shopping intent is highest.

Boxing Day, uniquely Canadian in its e-commerce significance, is the single highest online shopping day for many Canadian retailers. The irony of having a slow, image-heavy store on Boxing Day is that you are failing your customers at the exact moment they are most motivated to buy from you. Optimize well before December 26.

Platform Breakdown: Shopify, WooCommerce, and Squarespace {#platform-breakdown-shopify-woocommerce-and-squarespace}

Different platforms handle images in fundamentally different ways. What you can control, and what happens automatically, varies significantly between Shopify, WooCommerce, and Squarespace. Understanding your platform's behavior tells you exactly where manual optimization is required.

Shopify Canada

Shopify is headquartered in Ottawa, Ontario, and is the dominant e-commerce platform for independent Canadian retailers. Shopify does offer some automatic image handling: it serves images via its global CDN (Fastly), automatically converts images to WebP for browsers that support it, and allows you to request differently-sized image variants via URL parameters.

However, Shopify's automatic optimization is not a complete solution. The platform stores and serves images at the dimensions and file sizes you upload. If you upload a 6MB JPEG, Shopify stores a 6MB JPEG and serves it (converted to WebP at the point of delivery, but at the same underlying quality level and pixel dimensions). The WebP conversion helps with format efficiency, but it does not compensate for images that were never properly compressed before upload. An 800KB WebP derived from a 6MB uncompressed JPEG is still significantly larger than it needs to be. A properly compressed WebP exported at appropriate settings before upload might be 120-180KB.

The Shopify-specific recommendation is to pre-optimize all images before uploading them. Do not rely on Shopify's automatic handling to fix source file problems. Use a tool like Optimage to compress your product photos in bulk before they ever reach Shopify, then upload the optimized versions. Our full Shopify image optimization guide covers detailed upload settings and theme-specific recommendations.

Shopify image dimension recommendations: product images at 2048x2048px maximum (Shopify's image viewer supports zoom up to this resolution), collection page images at consistent aspect ratios defined by your theme, and background/banner images at 2000-2500px wide for full-width sections. At proper compression settings, these dimensions should result in files between 100-400KB each.

WooCommerce

WooCommerce on WordPress gives you more control than Shopify, which is both a strength and a weakness. WordPress generates multiple image size variants automatically (thumbnail, medium, large, full) for every uploaded image, using settings defined in your theme and plugins. If your theme is well-configured, these variants mean the right-sized image is served in the right context. If your theme is poorly configured, you may have hundreds of regenerated image sizes that do not match your current design.

WooCommerce does not automatically convert images to WebP or AVIF without a plugin. You need either a performance plugin (Imagify, ShortPixel, Smush) or to pre-optimize all images before upload and use a caching plugin with CDN support. The pre-optimization approach, compressing images with a local tool before they go into the WordPress media library, is the most reliable approach and requires no additional plugin dependencies.

WooCommerce stores often have larger image libraries than Shopify stores because WordPress makes it easy to accumulate media over years of operation. A 3-year-old WooCommerce store may have 2,000-10,000 images in the media library, the vast majority of which were uploaded with no optimization. That historical debt is a real performance problem, and it is addressable with bulk processing tools.

The WooCommerce-specific complication is image regeneration. When you change themes or update your theme settings, WordPress may need to regenerate all image size variants. This process uses the original uploaded file as the source. If your originals are high-quality, well-optimized files (rather than raw camera output), regeneration produces clean results. Start with good source files.

Squarespace

Squarespace is popular with design-conscious Canadian retailers, particularly in categories like fashion, jewelry, art, and home goods. The platform handles images reasonably well by default: it serves images via Fastly CDN, automatically generates multiple size variants, and supports WebP delivery for compatible browsers.

Squarespace's maximum image upload size is 20MB, which has led some users to upload unnecessarily large files in the belief that "bigger is better." It is not. Even if Squarespace serves a smaller web-optimized version, the upload itself takes longer, and the original file consumes storage. Pre-optimizing before upload improves upload speed, reduces storage footprint, and ensures your source files are clean regardless of what Squarespace does with them in delivery.

The 20-Minute Fix: A Practical Bulk Compression Workflow {#the-20-minute-fix-a-practical-bulk-compression-workflow}

This workflow is designed for a non-technical Canadian store owner who has a product image library that has never been optimized. It assumes you are on a Mac and comfortable with basic file management. The entire active work time is about 20 minutes, though the compression itself may run for longer depending on your library size.

Step 1: Gather your existing product images (5 minutes)

Create a new folder on your desktop called "product-images-original." Copy (do not move) all of your current product images into this folder. If you are on Shopify, you can download your images from the Shopify admin under each product, or use a bulk export tool. If you are on WooCommerce, they live in your WordPress media library and are accessible via the server or an FTP/SFTP client. If you have them locally, collect them from wherever they live on your drive.

Step 2: Audit what you have (5 minutes)

Sort the folder by file size, largest first. Look at the top 20 files. Are they larger than 500KB? Larger than 1MB? Anything over 500KB for a web product image is worth compressing. Anything over 1MB is definitely worth compressing. Note how many files you have total and the total folder size. This is your "before" baseline.

Step 3: Compress with Optimage (5 minutes setup, then let it run)

Optimage is a Mac application for bulk image compression that handles JPEG, PNG, WebP, and AVIF files. Drag your entire product image folder into Optimage, set your output quality level (85 is a good starting point for product photos; it preserves visible detail while achieving meaningful file size reduction), and specify an output folder called "product-images-optimized." Click compress and let it run. Depending on library size, this takes anywhere from 2 minutes to an hour. You do not need to watch it.

Step 4: Verify quality (3 minutes)

When compression finishes, open a few of the output files and compare them visually to the originals. Check images with fine texture detail (fabric, jewelry, food), images with text overlays, and images with high contrast. At quality 85, you should see no visible difference in a side-by-side comparison. If any images look worse than acceptable, re-run those specific files at quality 90.

Step 5: Upload optimized images to your store (ongoing)

Re-upload the optimized images to your store. On Shopify, this means going into each product and replacing the images. On WooCommerce, use a plugin like "Enable Media Replace" to swap files in place without breaking existing links. On Squarespace, re-upload through the block editor.

The total active effort is about 20 minutes. The ongoing habit is simple: run new product photos through Optimage before uploading, every time. Build it into your product listing workflow as the step that happens before the image goes anywhere online.

Sizing guidelines to use when compressing:

  • Product photos (main image): 1200x1200px, quality 85, target under 200KB
  • Product photos (lifestyle context shots): 1600x1067px, quality 82, target under 250KB
  • Collection/category banner images: 2000x800px, quality 80, target under 300KB
  • Hero/homepage banners: 2400x1200px, quality 78, target under 400KB

These are target ranges, not hard limits. Some complex images need slightly higher quality settings. Some simple images (flat lay shots with plain backgrounds) compress remarkably well at quality 75 with no visible degradation.

Canadian Marketplace Considerations: Etsy, Kijiji, Facebook Marketplace, Wayfair {#canadian-marketplace-considerations-etsy-kijiji-facebook-marketplace-wayfair}

Canadian sellers operating across multiple platforms need to understand that each marketplace has different image requirements and different performance implications.

Etsy Canada uses the same global image specs as Etsy US: minimum 2000px on the shortest side, JPEG or PNG format, with a practical maximum upload size of 20MB per image. Etsy's own CDN handles delivery, but you still want your source images to be clean and properly exposed. Heavily compressed or visually noisy source images look worse after Etsy's own processing pipeline. Aim for 1-2MB per listing image as your upload target: large enough to maintain quality through Etsy's processing, small enough to upload quickly when you are adding 50 listings in a session.

Kijiji (the dominant Canadian classifieds platform) accepts JPEG and PNG images with a maximum of 20 images per listing and a 20MB total upload limit. For casual sellers, this is rarely a constraint. For high-volume sellers listing 50-100 items per week (used furniture, clothing, electronics), the upload friction of large unoptimized photos is real time lost. Compressing Kijiji listing photos to 300-500KB per image maintains perfectly acceptable display quality on the platform while reducing upload time by 60-80%.

Kijiji Autos specifically recommends "high quality" photos and allows up to 50 images per listing. For dealers listing dozens of vehicles, uploading 50 unoptimized photos per vehicle across a 200-vehicle inventory means potentially 10,000 photos in a single catalog session. At 5MB average per photo, that is 50GB of upload. At 400KB average per compressed photo, it is 4GB. This is not a trivial difference.

Facebook Marketplace Canada has the same behavior as Facebook globally: the platform aggressively recompresses any image you upload. This means uploading a 5MB photo versus a 500KB photo results in nearly identical display quality on the platform, because Facebook applies its own compression pipeline regardless of what you send. The practical implication: you do not need to upload pristine full-resolution files to Facebook Marketplace. A 1024px wide JPEG at 80% quality is completely adequate.

Wayfair Canada has more specific requirements for its professional seller catalog. Wayfair's image guidelines specify white or near-white backgrounds, minimum 1500px on the short side, no watermarks, and explicit file naming conventions. For Wayfair specifically, submit clean, properly exposed JPEGs at around 2000x2000px, compressed to approximately 500KB-1MB. Going larger does not improve display quality through their pipeline but does slow down your catalog submission workflow.

The Quebec Market: Bilingual Product Images and French-Language SEO {#the-quebec-market-bilingual-product-images-and-french-language-seo}

Quebec represents approximately 23% of Canada's population and a significant share of Canadian e-commerce volume. If you are selling physical products to Canadian consumers and you are not serving the Quebec market in French, you are leaving money on the table. But the Quebec market creates a specific image optimization consideration that most guides overlook: bilingual product images.

Many brands create product images that include text overlays, labels, promotional callouts ("Free Shipping," "Best Seller," "New Arrival"), or instructional content baked directly into the image file as a flat raster. When you need French and English versions of these images, your image library doubles overnight. Suddenly instead of 500 product images you have 1,000. The optimization burden scales proportionally.

The practical recommendation is to avoid text-in-image wherever possible. Use HTML/CSS text overlays for promotional callouts rather than baking text into images. Reserve image text only for cases where it is truly necessary (product packaging that shows French/English bilingual text, as required by Canada's Consumer Packaging and Labelling Act, for example). This keeps your image library manageable, your translation workflow clean, and your optimization burden reasonable.

When bilingual product images are unavoidable, build a naming convention from the start: product-name-en.jpg and product-name-fr.jpg in parallel folders. Compress both versions in the same Optimage session. Some Shopify themes support localization apps (Langify, Weglot) that can serve different image variants based on the customer's language preference, which is the most elegant solution for large bilingual catalogs.

For French-language SEO specifically, your image alt text and file names matter. An image file named boucles-oreilles-argent.jpg with alt text Boucles d'oreilles en argent sterling will perform better in French-language Google searches than a generic product-123.jpg. This is not strictly a compression issue, but it is an SEO issue that Canadian e-commerce operators serving Quebec should be aware of, and it costs nothing to implement when you are already renaming files as part of your optimization workflow.

The hreflang attribute on your pages, combined with properly localized image alt text, creates a complete French-language SEO signal that helps Google understand which version of your store to show to Quebec-based searchers. This is table stakes for any Canadian store with serious Quebec ambitions.

Google Shopping Canada: How Image Quality Affects Your Ad Performance {#google-shopping-canada-how-image-quality-affects-your-ad-performance}

Google Shopping ads are a critical acquisition channel for many Canadian e-commerce stores. If you are running Shopping campaigns through Google Ads, the quality and relevance of your product images directly affects your campaign performance in ways that go beyond simple aesthetics.

Google's Quality Score for Shopping ads incorporates click-through rate as a major component. Your product image is the most visually dominant element of a Shopping ad unit. In a row of 6 competing product listings, the image is what determines whether a shopper pauses and clicks or scrolls past. Research from Shopify and various CRO studies consistently shows that product images with clean white or neutral backgrounds, good lighting, and sharp focus significantly outperform cluttered or dark images in Shopping ad click-through rates.

The image quality standards Google enforces for Shopping are documented in the Google Merchant Center guidelines: minimum 100x100 pixels for non-apparel, minimum 250x250 for apparel, recommended 800x800 or larger for optimal display. Google specifically warns against "unclear, blurry images" and images with "promotional text, watermarks, or borders." Products disapproved from Shopping due to image quality issues do not run at all. Products with marginally compliant but low-quality images run but underperform on click-through rate.

The Canadian-specific dimension here is the Google Shopping Canada feed. If you are advertising in Canada and in the US simultaneously (a common scenario for Canadian stores with cross-border shipping), your product feed needs images that meet quality standards for both markets. The standards are effectively the same, so a well-optimized product image library serves both.

One nuance: Google Shopping uses your product image at relatively small display sizes in the Shopping tab (typically 200-300px wide on desktop), but serves full-size images in the product detail panel when a user clicks. This means you need high-quality, well-optimized images that look good at both sizes, not just thumbnail-optimized tiny files. The target sweet spot: 1200x1200px at approximately 200-400KB, clean background, sharp focus, accurate color representation.

For Canadian retailers running bilingual campaigns targeting both English and French searches, the same product image typically serves both language variants in Shopping. Google does not require separate French and English images in the Shopping feed. However, your product titles and descriptions in the feed should be in the language of the ad campaign, which means you may need separate English and French Shopping campaigns pointing to language-specific product pages. The image itself remains constant.

A well-optimized image library also improves your Shopping ad loading speed on the Google Shopping tab, which is a factor in ad quality measurement. Faster-loading, correctly-sized product images display more crisply in Shopping ad units, particularly on high-DPI mobile screens where low-quality images look noticeably soft.

What Canadian Agencies and Freelancers Are Using Right Now {#what-canadian-agencies-and-freelancers-are-using-right-now}

The Canadian digital agency and freelance web design community has settled on a fairly consistent tool stack for image optimization work, based on conversations with practitioners in Toronto, Vancouver, and Montreal.

For Mac-based workflows, Optimage is the consistent recommendation for bulk local processing. It handles JPEG, PNG, GIF, WebP, and AVIF, runs entirely offline (no cloud upload of client files, which matters for confidentiality and PIPEDA compliance), and produces reliably excellent results with minimal configuration. Agencies doing site migrations or full catalog optimizations use it to process hundreds or thousands of images in a single session before uploading to the client's CMS.

For WordPress-specific workflows, ShortPixel and Imagify are widely used as plugin-based solutions that process images on upload and can bulk-process existing media libraries. These work well for clients who want a "set it and forget it" approach, though they add monthly cost (ShortPixel's API pricing is based on image credits, running roughly $9.99 CAD/month for a typical small to medium site).

For Shopify-specific workflows, pre-optimization before upload remains the agency standard. Most Shopify themes are not designed to be modified at the image delivery layer, and Shopify's own CDN handles format conversion reasonably well once files are uploaded at a sensible source quality. Agencies typically deliver a pre-compressed image library to clients with an Optimage preset, training clients to compress before uploading going forward.

For performance auditing, Google PageSpeed Insights (free), GTmetrix (free tier available), and WebPageTest (free, open source) are the standard tools. Canadian agencies often run audits from both US East and Canadian server locations in GTmetrix to capture the real-world latency experience of Canadian customers, particularly since some CDN edge nodes are located across the border.

A note on PIPEDA (Canada's Personal Information Protection and Electronic Documents Act) and cloud-based image optimization tools: when you upload product images to a cloud optimization service, you are sending files to a third-party server. For most product images, this is not a privacy concern. However, if your product photos inadvertently contain identifiable customer information (a photo taken in-store with a customer visible in the background, for example), uploading to offshore cloud services may create PIPEDA compliance questions. Local tools like Optimage, which process entirely on your own hardware, sidestep this issue entirely.

Before and After: A Toronto Shopify Store That Recovered $4,200 in Monthly Revenue {#before-and-after-a-toronto-shopify-store-that-recovered-4200-in-monthly-revenue}

This is a fictional but data-accurate example based on typical numbers for a mid-sized Canadian Shopify store.

The store: Hazel & Oak Studio, a fictional Toronto-based home goods brand selling candles, ceramics, and linen goods. The store launched in 2022. By late 2025, it has grown to 180 products, each with 6-8 product images. The owner uploads all product photos directly from her Canon EOS R6 in original size: each file approximately 12-20MB, exported from Lightroom with minimal compression.

The audit: A PageSpeed Insights audit of the best-selling product page (a $45 beeswax candle) returns a Performance score of 41 on mobile. The Largest Contentful Paint is 7.2 seconds. The audit flags "Efficiently encode images: potential savings of 4.8MB" on that single page. The page is loading 6 product images at an average of 1.1MB each.

The store generates approximately $28,000/month in total revenue. Mobile accounts for 68% of traffic. Mobile conversion rate is 1.8% against a desktop conversion rate of 3.4%, an unusually large gap that is a textbook indicator of mobile performance problems.

The image library stats before optimization:

  • Total product images: approximately 1,260 files (180 products x average 7 images)
  • Total library size: 15.1GB
  • Average file size: 12MB
  • Format: 100% JPEG (raw Lightroom exports)

The optimization process:

The owner downloads all product images from Shopify admin (this takes about 45 minutes as a background task). She installs Optimage, drags the entire folder in, and sets quality to 85 with WebP output enabled. The compression run takes 52 minutes on her MacBook Pro M2. She makes lunch.

The image library stats after optimization:

  • Total product images: 1,260 files
  • Total library size: 1.04GB
  • Average file size: 830KB
  • Format: WebP
  • Reduction: 93% smaller total library size

She re-uploads the optimized images to Shopify, replacing originals. This project takes about 3 hours spread over two evenings, mostly waiting for Shopify's upload interface to process files.

The result two weeks after re-uploading:

  • PageSpeed Insights mobile Performance score: 41 rises to 79
  • LCP on best-selling product page: 7.2 seconds drops to 2.1 seconds
  • Mobile conversion rate: 1.8% rises to 2.6% (a 44% relative improvement)
  • Monthly revenue from mobile traffic: approximately $9,600 rises to approximately $13,800
  • Net gain: approximately $4,200/month in additional revenue
  • Time invested: 20 minutes of active work plus 52 minutes of automated compression

The math on image optimization return on investment is consistently one of the most favorable in all of web performance work. There is no other 20-minute change you can make to a Shopify store that has this kind of revenue impact.

The mobile-to-desktop conversion rate gap also closes significantly. Before optimization, mobile converts at 53% the rate of desktop. After, it converts at 76% of desktop, which is much closer to industry benchmarks for well-optimized stores. That remaining gap is likely attributable to the natural difference in shopping behavior between mobile and desktop, not performance issues.

For a broader look at the revenue impact of image optimization across e-commerce categories, our image optimization and e-commerce revenue analysis covers the research in detail.

Your Next Step {#your-next-step}

If you manage a Canadian e-commerce store on any platform and you have never done a systematic image optimization pass, the exercise in this guide is your next 20 minutes of work. The before/after numbers in this example are not outliers. They represent the expected outcome for any store that has been uploading unoptimized images since launch.

The competitive reality of Canadian e-commerce in 2026 is that your customers have fast devices, increasingly mixed connection quality, and zero patience for slow pages. The stores that load in 1-2 seconds convert. The stores that take 4-7 seconds lose the sale to whoever loads faster.

Audit your store with Google PageSpeed Insights right now. If your mobile Performance score is below 70, image optimization is almost certainly the highest-leverage improvement you can make this week.

Ready to optimize your Canadian store's images? Try Optimage free and compress your first batch in under 2 minutes. No subscription required to get started.

Related reading:

Try Optimage Free

Compress, convert, and optimize your images in seconds. No sign-up needed to start.

Start Optimizing Free
Back to The Optimage Journal