How CDN Pricing Typically Works: Bandwidth vs. Requests

 Content Delivery Networks (CDNs) are essential for improving website performance, reducing latency, and securing applications. But when evaluating CDNs, understanding their pricing models is crucial because costs can vary significantly depending on usage patterns. The two most common pricing components are bandwidth and requests, and each affects your total CDN bill differently.

---

1. Bandwidth-Based Pricing

Bandwidth refers to the amount of data transferred from the CDN edge servers to end users. Most CDN providers charge based on gigabytes (GB) or terabytes (TB) delivered per month, often with tiered rates.

How It Works:

• Every time a user downloads content (like images, videos, or HTML pages) from the CDN, the data counts toward your bandwidth usage.

• Higher data transfer volumes mean higher costs.

Example:

• Tier 1: First 10 TB per month at $0.08/GB

• Tier 2: Next 40 TB at $0.06/GB

• Tier 3: 50+ TB at $0.04/GB

This tiered pricing incentivizes larger traffic volumes, as the per-GB rate decreases as usage increases.

Pros:

• Predictable for sites with consistent content sizes and traffic

• Easy to calculate cost based on estimated traffic

Cons:

• Can become expensive for sites serving large media files or high-resolution videos

• Traffic spikes can lead to sudden unexpected charges

---

2. Request-Based Pricing

CDN request pricing is based on the number of HTTP/HTTPS requests handled by the edge servers. Each time a user requests a resource (like a web page, API response, or image), it counts as a separate request.

How It Works:

• The CDN logs every user request.

• Pricing is often per million requests, sometimes with separate rates for HTTP vs. HTTPS.

Example:

• First 1 million requests: $0.50

• 1M–10M requests: $0.45 per million

• 10M+ requests: $0.40 per million

Pros:

• Ideal for websites or APIs that serve small amounts of data per request

• Useful when bandwidth usage is low but request volume is high

Cons:

• For large static assets like video or software downloads, bandwidth costs may dominate

• Websites with millions of small requests (like SPAs) can see costs rise quickly

---

3. Combined Pricing Models

Most modern CDN providers use a hybrid pricing model combining both bandwidth and requests:

• Bandwidth charges: For total data delivered

• Request charges: For the number of user requests processed

Some CDNs also differentiate by region, charging more for data delivered to high-cost regions like North America or Europe, and less for regions like Africa or South America.

---

4. Additional Pricing Considerations

Besides bandwidth and requests, other factors can affect CDN costs:

a. SSL/TLS and HTTPS Requests

• Some CDNs charge extra for HTTPS requests because of the CPU overhead of encryption.

• Others include it in the base cost.

b. Invalidation and Purge Requests

• Removing or updating cached content (cache invalidation) may incur extra charges.

• Frequent purges of large content libraries can add to costs.

c. Data Transfer to Origin

• Some CDNs charge for origin fetches, i.e., when the edge server retrieves content from the origin server because it’s not cached.

• This is especially relevant for dynamic content or cache misses.

d. Features and Add-Ons

• Advanced WAF, bot mitigation, edge computing, and analytics may have separate subscription fees.

---

5. Cost Optimization Strategies

To manage CDN costs effectively:

• Maximize caching: Reduce origin fetches by caching content longer at edge servers.

• Compress and optimize content: Minimize file sizes to reduce bandwidth charges.

• Use region-specific caching: Deliver frequently accessed content locally to reduce global bandwidth.

• Monitor usage analytics: Track spikes in requests and bandwidth to prevent surprises.

---

6. Real-World Example

Imagine a streaming website:

• 100,000 users watch 1 GB video each. Total bandwidth = 100 TB.

• The CDN charges $0.05/GB for bandwidth, totaling $5,000.

• Each user makes 20 requests (for video segments, metadata, thumbnails), totaling 2 million requests at $0.50 per million = $1.00.

Even though request charges seem small here, for APIs serving millions of tiny requests, the request-based costs can outweigh bandwidth costs.

---

7. Takeaway

CDN pricing revolves primarily around bandwidth (data volume) and requests (number of accesses). Understanding your traffic patterns, content size, and usage distribution is essential to estimate costs accurately. While bandwidth usually dominates for media-heavy websites, request counts can add up for high-frequency API calls or dynamic web applications.

The most cost-effective CDN strategy often involves optimizing caching, compressing content, and reducing unnecessary requests, while choosing a pricing model aligned with your traffic profile.

By understanding these nuances, businesses can leverage CDNs to improve performance, enhance user experience, and maintain predictable costs.