TinyPNG vs. modern workflows: where image optimization belongs in 2026

We compare manual tools like TinyPNG and Squoosh against automated solutions like build scripts and CDN services to find the right image optimization workflow for founders and developers today.

The Answer Up Front

For a quick, one-off image compression, manual web tools like TinyPNG or Google's Squoosh are adequate. For any production website or application, they are the wrong choice. Production workflows should automate image optimization in the build step or delegate it to a content delivery network (CDN). The goal is to make optimization an invisible, reliable part of your infrastructure, not a manual chore. Manual compression is a repetitive task that introduces friction and is prone to human error; automation solves this systematically.

Methodology

This is a v0 strategic review based on a qualitative user query about image compression workflows, not a quantitative performance benchmark. Our analysis draws from the discussion prompted by a user on Reddit and our knowledge of modern web development practices. Update cadence: this review will be updated to a v1 with reproducible benchmarks when we establish a test corpus.

• Tools Analyzed: This review covers four distinct approaches to image optimization, using specific tools as examples: Manual Web UIs (TinyPNG, Squoosh), Local GUI tools (ImageOptim), Build-Step Automation (sharp, framework integrations like Next.js Image), and Runtime CDN Services (Cloudflare Images, Imgix).
• Source Signal: Reddit post, "Does anyone still use TinyPNG?", URL: https://www.reddit.com/r/SideProject/comments/1unjkjg/does_anyone_still_use_tinypng/, observed July 5, 2026.
• What's Covered: The strategic pros and cons of each workflow, who should use which, and the total cost of ownership (including developer time).
• What's Not Covered: We have not run a controlled benchmark comparing the compression ratios, visual quality (SSIM/Butteraugli), or performance (ms) of these tools. All discussion of performance is based on the architectural implications of the workflow, not on measured output.

The Workflow Options

Image optimization is not a single tool but a step in a process. Where you place that step has significant consequences for developer experience and site performance.

Manual web tools: TinyPNG and Squoosh

This is the simplest workflow. You navigate to a website, drag and drop your images, wait for the compression, and download the results. TinyPNG is famous for its simplicity and effective "lossy" PNG compression. Squoosh offers far more control, allowing you to experiment with different codecs (like AVIF and WebP), resize, and visually compare the input and output before downloading. This approach is fine for non-developers or for assets that are rarely updated, like a logo or a favicon. Its primary drawback is that it is manual, non-reproducible, and disconnected from the development pipeline.

Build-step automation: sharp and framework integrations

This is the modern standard for sites with a build process (e.g., Next.js, Astro, Hugo). A library like sharp (a high-performance Node.js image processor) is added to the project's dependencies. During the build, the script automatically finds all source images, resizes them to various required dimensions, converts them to modern formats like WebP or AVIF, and saves the optimized versions to the output directory. Frameworks like Next.js abstract this further with components like <Image>. You point to a single source image, and the framework handles the rest. This workflow is automated, version-controlled, and highly consistent.

Runtime CDN optimization: Imgix and Cloudflare

This is the most powerful and flexible approach, especially for sites with user-generated content or a large number of images. Instead of optimizing images at build time, you upload high-resolution source images to a service like Cloudflare R2 or Amazon S3. An image CDN like Imgix, Cloudinary, or Cloudflare Images sits in front of your storage. You request images via a special URL that includes transformation parameters (e.g., ?width=800&format=auto). The CDN fetches the source image, performs the transformations on the fly, and caches the result at the edge. The format=auto parameter is key; it enables content negotiation, where the CDN detects browser support and serves AVIF to a modern Chrome browser and WebP or JPG to older ones.

What's Interesting / What's Not

The most interesting shift is that image optimization is no longer a distinct task for a developer to perform but an attribute of the infrastructure. The problem has been largely solved and commoditized, moving from a designer's Photoshop "Save for Web" dialog to an invisible process handled by a build tool or a CDN. The persistence of manual tools like TinyPNG highlights the gap between professional web development best practices and the needs of a much broader audience who just need to make a picture smaller without installing Node.js.

The critical feature that manual tools cannot offer is content negotiation for next-gen formats. Serving a 25KB AVIF file instead of a 70KB JPG is a significant performance win, but you can't just save everything as AVIF because Safari on older desktops might not support it. CDNs and modern frameworks handle this negotiation automatically. This capability alone makes automated solutions superior for any production use case.

Pricing

Pricing models reflect the different workflows. (Snapshot from July 5, 2026)

• TinyPNG: Free for up to 20 images per month (5 MB max each). Pro is $39 per user per year.
• Squoosh: Free (runs in-browser).
• ImageOptim: Free (macOS, donation-supported).
• sharp (library): Free (open source).
• Cloudflare Images: $5/month per 100,000 stored images, $1/month per 100,000 delivered images.
• Imgix: Starts at $75/month for the Basic plan, which includes a base processing fee and bandwidth.

Verdict

Manual tools like TinyPNG are for one-off tasks, not production systems. They are the digital equivalent of a hand tool when the job requires a factory assembly line.

• If you are building a modern static website or web app with a build step (Next.js, Astro, etc.): Use your framework's built-in image optimization component. It's the most integrated and lowest-friction solution.
• If you are handling user-generated content or have a very large, dynamic image catalog: Use a dedicated image CDN like Imgix, Cloudinary, or Cloudflare Images. The flexibility and performance of runtime optimization are worth the cost.
• If you need to compress an image for an email or a PowerPoint deck: TinyPNG is perfectly fine.

What We'd Test Next

A v1 of this review would require a quantitative benchmark. We would create a test suite of 100 images representing common web use cases (photographs, UI screenshots, vector-like illustrations). We would then process this corpus using each tool/workflow with standardized settings. The primary metrics for comparison would be file size reduction at a consistent visual quality level (measured with a metric like SSIM or Butteraugli), the impact on build times for automated tools, and the Time To First Byte (TTFB) for images served from a CDN versus a standard origin.

The investor read

The manual image compression market, occupied by tools like TinyPNG, is a durable, long-tail business driven by SEO and brand recognition. It is not a venture-scale opportunity. The high-value market has moved up the stack to the infrastructure layer. Image optimization is now a feature, not a product. It's sold by CDNs (Cloudflare, Fastly), hosting platforms (Vercel, Netlify), and large-scale media management APIs (Cloudinary, Imgix) as part of a much stickier, higher-ACV offering. The core compression technology is largely commoditized open source (libvips, mozjpeg). An investable opportunity in this space would need to innovate beyond compression, focusing on AI-driven features like generative cropping, subject-aware resizing, or automated visual merchandising. The money is in the automated workflow, not the manual tool.