Caching is often thought of as a complex technical problem. However, in this article we show how it’s much more than that - it’s essential to user experience, especially for global users, and has a huge impact on server costs, directly impacting your companies’ OpEx spend.

What is Caching?

Caching, simply put, means storing frequently accessed data closer to the user. Imagine keeping your office supplies at your desk instead of constantly fetching them from a distant storage room. Similarly, web caching saves copies of web pages, images, or assets, reducing retrieval times and improving performance.

As importantly, it’s important to know that if the webpage itself is cached (along with the resources), it means that your own server infrastructure does not need to render the page. Each time your servers have to render a page it uses often expensive resources to do so - including hitting your CMS or database. If it’s stored on a CDN cache, you can often significantly decrease your server and software licensing bill.

CDN Caching: Performance on a Global Scale

CDN caching specifically refers to storing content on geographically distributed servers (CDN edge servers) around the world. Instead of every user’s request traveling all the way to your origin server (e.g., your main data center), a CDN will serve many requests from a location near the user. The CDN’s globally distributed network *duplicates your website content and stores it across various servers*. This drastically reduces the distance and time needed to deliver data to each user. For example, if a user in Tokyo, Japan, tries to access a website hosted on an origin server in New York, USA, a CDN would serve the website's content from a server located in or near Tokyo, rather than forcing the data to travel all the way from New York. This significantly reduces the round-trip time for data.

This distributed model fundamentally alters the traditional centralized approach to web hosting, where all users, regardless of location, would have to connect to a single origin server. By decentralizing content storage and delivery, CDNs not only enhance speed but also improve the scalability and resilience of web applications. If an origin server experiences an outage, some CDNs can continue to serve cached content from their edge servers, maintaining availability.

Major CDN providers make this process seamless. Cloudflare, for instance, operates 250+ global data centers that cache and serve content closer to users. Akamai, one of the oldest CDNs, has an enormous network of about 365,000 servers in 135+ countries. Fastly and others also maintain hundreds of points of presence worldwide. The presence of these servers around the globe means content can be delivered from a location that is often in the same region (or even the same city) as the end-user, greatly minimizing latency. In summary, caching is like strategically placing warehouses of information across the globe so that customers are always served from the nearest warehouse rather than a single distant factory.

Lower Costs and Better UX: The MediaCo Example

One of the biggest business advantages of CDN caching is dramatically lower server and bandwidth costs. By serving a large portion of traffic from caches, companies can offload work from their origin infrastructure. Fewer requests hitting your origin servers means you can provision fewer servers (or smaller instances) and pay for less outbound bandwidth from your hosting provider.

Consider the fictional company MediaCo, a media publisher serving news and articles. MediaCo receives 2 million page views per day, much of that during peak traffic hours. Each page render takes 1000ms. Each $1,000/month database + application server can render 4 pages simultaneously. To handle peak traffic and spikes, MediaCo runs 16 servers – costing $16,000/month.

After implementing CDN caching for logged-out users, they achieve a 90% cache hit ratio. This allows them to reduce the origin load drastically - only 10% of requests hit the origin, and they can now scale down to just 2 servers, costing $2,000/month.

Result: $14,000/month saved on infrastructure – $168,000/year.

But the financial gains don’t stop there. 40% of their user base is geographically far from the servers – in Europe and Asia. These users previously experienced slow load times and high bounce rates. After enabling caching closer to users, MediaCo saw a 50% increase in engagement from these regions – leading to a 20% overall increase in top-line global engagement.

Faster websites keep users happy. It’s well documented that slow load times drive users away, while speed boosts engagement and conversion. CDN caching’s ability to improve performance is therefore a strategic asset for user experience and revenue growth.

When content is served from a nearby cache server, it loads significantly faster for the end-user, especially if they are far from the origin server. Industry research underlines why this matters: a website taking longer than three seconds to load can lead to over a 50% increase in bounce rate. A one-second delay in response can cause a ~7% drop in conversion rates - conversely, a one-second improvement can increase conversions by that amount.

For MediaCo, this translated into higher ad impressions and longer sessions. Faster load times for the 40% of users who were previously underserved in Asia and Europe led to a 50% increase in engagement among that group, and a 20% lift overall in top-line engagement.

Between server savings and increased engagement-driven revenue, MediaCo effectively generates over $250,000 in net gains per year.

Balancing Freshness and Performance

Whenever we talk about caching, a natural concern arises: How do we ensure users get up-to-date content? Business stakeholders might worry that caching means users see old information. The good news is that modern caching is configurable - you can get the best of both worlds by tuning how long content stays cached and how updates propagate. Two important concepts to understand here are max-age and stale-while-revalidate.

• Max-Age (Cache Lifetime): Think of max-age as an expiration time for cached content. During this window, all users get fast, cached responses.
• Stale-While-Revalidate (Serve Stale While Updating): If a request comes in after that max-age window, the CDN can immediately serve the stale version and fetch a new one in the background. This means fast response times even when cache content has expired, minimizing performance drops.

With max-age and stale-while-revalidate, companies can confidently cache content for performance while still ensuring freshness when needed.

Optimizing Cache Durations: PressHoldings Example

PressHoldings is a fictional portfolio publisher with CDN caching in place. However, they had initially configured a global max-age of just 60 seconds. Their goal was to ensure readers always saw the most recent version of content.

Unfortunately, this aggressive freshness came at a steep performance and cost penalty – only 25% of their traffic was served from the CDN cache, due to the high volume of long-tail evergreen content that wasn’t being cached for long.

By implementing a dynamic cache strategy based on content age and type, they improved cache effectiveness dramatically:

The result: their cache hit rate improved from 25% to 85%.

That jump in cache efficiency significantly reduced origin server requests, lowered bandwidth costs, and ensured a much faster experience for users browsing archive content. Additionally, their editorial team retained control to instantly purge CDN cache for urgent updates.

Cache Invalidation/Purging as a Companion to max-age

What happens if content with a long max-age needs to be updated before its max-age expires? This is where cache invalidation or purging comes in. CDNs provide mechanisms to manually remove specific content from their caches across the globe. This is crucial because it allows for the use of long max-ages to maximize cache hit ratios, while still providing the ability to update content quickly when necessary.

Risks and Considerations

Caching strategies must carefully balance freshness and performance. Misconfigured caches could deliver outdated content or introduce complexity in content invalidation processes. Employing strategic invalidation practices ensures businesses maximize caching benefits without sacrificing content accuracy.

Conclusion: Cache Your Way to a Better Web Presence

CDN caching is far more than an IT tweak – it’s a strategic asset for modern businesses online. By caching content on globally distributed CDNs like Cloudflare, Fastly, Akamai, CloudFront, and others, organizations can simultaneously reduce costs and improve user satisfaction. From our exploration, a few key points for business stakeholders stand out:

• Massive Cost Savings: As MediaCo shows, caching 90% of traffic reduced their server bill from $16,000 to $2,000/month.
• Better User Experience Globally: For 40% of their traffic from distant geographies, CDN caching improved engagement by 50%, leading to a 20% global increase.
• Strategic Caching Beats One-Size-Fits-All: PressHoldings’ hit rate jumped from 25% to 85% with dynamic max-ages.

Caching gives you leverage – you handle more traffic with fewer resources and deliver a better experience to every user. It’s a strategic win no business can afford to overlook.

Catch Metrics is the leading solution to help solve ad stack performance problems.Get in touchto learn more from our experts