Starting immediately, LittleBizzy is suspending MaxCDN implementations for all of our managed hosting clients. This is not a decision that we made lightly, nor does it reflect any ill-will between us and the wonderful folks at MaxCDN.
In short, there are three main reasons for us having reached this decision, mentioned below in more detail. Ultimately, we do aim to keep in touch with MaxCDN and closely monitor their evolution during the 2016 year and beyond, with the hope of possibly re-offering their services in the future (for at least some of our hosting plans).
1. CloudFlare’s exponential growth. CloudFlare has been the subject of much ridicule ever since they first launched, often for reasons that didn’t even make sense and usually at the behest of critics who “didn’t matter” i.e. fanboy bloggers who lacked any real need for enterprise grade SaaS solutions. Over the past few years, they’ve quietly surged ahead, acquiring thousands of corporate and governmental clients around the world, landing another $110 million in funding in late 2015. Already valued at over $1 billion, it is my personal opinion that they’ve been one of the most under-valued and under-rated startups in history from the start, and will ultimately become one of the most important internet companies of the next decade.
But personal opinions aside, the main point here is that CloudFlare’s network map has exploded in growth. Now offering edge locations in places like India, China, and even Kenya, competing CDN services could only dream of even scratching the surface of what CloudFlare currently offers in terms of world-wide network penetration. And while the arguments for using a “real” CDN instead of or alongside CloudFlare still theoretically apply, even in a HTTP/2 world (for example, alleviating your server’s need to process and deliver static files, thus avoiding any potential “uncached” file queries in CloudFlare’s reverse proxies that may hit your origin server), the truth is that with proper website coding and server configuration, the ability to cache “most” file requests via CloudFlare still outperforms other CDN options at this point for the majority of internet users when considering web performance on a global scale. That is, while several case studies (and monitoring tools of our own) have undeniably shown that CloudFlare’s network tends to offer “less consistent” start-to-finish load times for static files than alternatives like MaxCDN (etc), ultimately it comes down to practicality — not theory — when concluding that CloudFlare offers “more” than MaxCDN when considering a global audience, if only because of their sheer number of edge locations.
2. Ongoing CDN bandwidth costs. The truly amazing thing about MaxCDN is that arguably, they single-handedly “created” the CDN industry in terms of small business owners, especially in the United States and Europe. With a slick interface and stellar customer support, they’ve easily maintained their market lead among WordPress site owners and beyond (while “corporate” players like Akamai have quietly enjoyed market domination among enterprise clients for years).
Even the mighty Amazon has struggled to steal market share away from Akamai with their CloudFront CDN service, which despite its impressive network still pales in comparison to CloudFlare’s currently offerings. But whether we are comparing “corporate” CDN players or their more user-friendly competitors targeting small businesses (i.e. MaxCDN or the up-and-coming KeyCDN), the bandwidth costs are becoming more difficult to justify as a managed hosting reseller in a post-CloudFlare world. Despite MaxCDN’s clear lead in this area with very flexible pricing options, and despite the fact that bandwidth prices drop as resellers like us grow, the truth is that these services do not provide enough benefit to the vast majority of our clients to justify the added costs; in fact, there is increasing doubt that much benefit is provided, at all. For example, as LittleBizzy focuses on improving WordPress performance for small business owners, we often get new customers from countries that have historically suffered from poor network performance such as Australia, Israel, Southeast Asia, and beyond. Ironically, it’s within these type of countries that MaxCDN (and other CDN providers) tend to have the least infrastructure, meaning that taking our clients OFF of the CDN and relying on CloudFlare alone actually can improve performance.
Now, this is not to say that we aim to ditch a service like MaxCDN for the sake of our “average” client; nor is it solely a cost/ROI concern by any means. Rather, the truth is that unless your site’s audience is located in the United States or Western Europe, and unless your site is very image/resource heavy, and unless you origin server is negatively affected by occasionally processing files behind CloudFlare, then a CDN is increasingly not necessary for small businesses, especially as the other features and overall performance of the CloudFlare network continues to evolve and improve.
3. Confusion among small business owners. The vast majority of our clients don’t understand or care to study the science behind web performance. To this end, constantly monitoring CDN bandwidth and asking clients to upgrade their CDN package (etc) quickly becomes a drag on our relationships, not to mention a great waste of time on both ends. As one of our more tech-savvy clients once said, “Most customers just want it to work and it’s a pretty binary evaluation, where for most speed quickly gets to be ‘fast enough’…”, which is a fantastic summary, if we are honest.
I am not the type of person to accept a sort of “well, good enough!” mentality, if I know that better performance or better business can be achieved. Still, there is a surprisingly clear line of “good enough” when it comes to website performance, and when we are able to achieve half-second loading times for our clients using nothing but a solid LEMP stack configuration on DigitalOcean paired with CloudFlare’s DNS and optimization, it is even more reason to consider re-simplifying our features. Ultimately, the goal of LittleBizzy has always been bringing “speed, stability, and security” to small business owners in a simple, monthly package that doesn’t require extra logins, extra bills, or extra back-and-forth.
In conclusion, there is still a very solid market for CDN providers like MaxCDN, especially for heavy websites in Western countries looking to stabilize their sites and bring extremely consistent loading times to their static resources. Before writing this post, MaxCDN’s Sean Walsh indeed confirmed to me that HTTP/2 and more edge locations would be coming soon, and even on-the-fly image compression may be coming in late 2016 as well. Sparing a point-by-point comparison for another time, these types of features will be key to ensure MaxCDN’s (etc) long-term survival.
For now, this isn’t goodbye, MaxCDN, but perhaps a “rock on, and hopefully see you again.” ;)