A quick foreword:

Welcome to my first post! Once in a while I read something and then obsessively read some more, and start to don my strategic thinking hat. This blog is my space to capture these thoughts and invite discussions from like-minded professionals! Feel free to message me on LinkedIn and start a conversation if you want to chat more!

Introduction

Jamstack is a web development architecture concept that could have major implications in the $100bn web hosting market.

Today, I'm going to try and break down what Jamstack is, explore the impact it could have for millions of web developers and businesses, and see what two very exciting companies are doing to capture this market.

This is a packed post with everything from product strategy to marketing strategy, so strap in!

This post covers the following:

• An introduction to the Jamstack concept
• The growth of Jamstack, and what the future could hold here
• Netlify and Vercel's approaches to capturing the market
• How these platforms could evolve to ensure success

What is a Jamstack?

I'm going to try and explain this without getting neck-deep into overwhelming tech terms. (P.S. I said I'd try, not that I'd succeed. However, if you aren't interested in the foundational concepts, and want to skip to the business end of things, feel free to skip to the next section.)

Jamstack is an approach to build websites that are in practice fast, cheap, easy to build and secure compared to their traditionally architected counterparts.

The term was coined by Matt Billman, CEO of Netlify, by combining 'JAM' + Stack:

• Javascript  -  Any code that the page runs on a web browser. Developed to handle any dynamic requests from the user's web browser.
• APIs  -  Custom built or third party services that can enrich a site with personalised data, and add new functionality like billing, forms and more.
• Markup  -  The static content that users will see and interact with, usually pre-rendered and served via a Content Delivery Network (CDN).

The concept was to take these three elements in order to develop and build sites. To be honest, this wasn't new to anyone 5 years ago, nor is it new today. Since the dawn of microservices and APIs, most websites have a de-coupled architecture and work this way anyway.

The core difference between Jamstack and traditional web architecture is how the pages are created and served to users.

Traditional websites typically generate a fresh copy of a page every time a user visits it. This takes time, resulting in lower performance. It also means there's a server that needs to be running all the time to generate these pages.

Jamstack, however, puts the focus on pre-rendering these web pages when the code is built. In other words, a page is generated once, and shipped out en-masse to a users via CDNs. This means there's no application server creating a brand new version of the page each time a user lands on a website. The page loads fast, and it scales efficiently.

The other aspect of Jamstack sites is that they are intended to be 'serverless' in every aspect, from being built to serving dynamic requests. (There is a server running your code after all, it's just that you don't need to set it up, install and manage it yourself. Here, a managed service like AWS Lambda takes care of it all and you pay only for the time your code runs.)

If you thought this sounded too good to be true, you're right.

Now you're probably thinking, but waiiiiit… there are a number of things that don't make sense here. How can every user be served the same page? What if there are experiences unique to each user which need to be personalized? What if there are pages which contain time-sensitive information or need to load on every refresh?

The truth is, the Jamstack approach is not a silver bullet to the pains of traditional web architecture. However, for mostly static websites it is a blessing. Developers can cook up fantastic web experiences in extremely short periods of time. There is little to no setup on the backend required, the site loads blazingly fast, and is cheap to operate. Plenty of services like Vercel, Netlify, GitHub Pages that will even build, deploy and host your site for free, and only charge for advanced features or high usage.

However, for complex websites which serve mostly dynamic content, Jamstack architecture can become highly convoluted to set up and manage. There are ways to deliver some amount of dynamic content efficiently, but there's a tipping point beyond which the core concept of "pre-rendering pages, bypassing the application server and delivering a webpage straight from a CDN" falls flat.

So is this approach of building websites popular?

Not so much… yet. Just under 1% of all websites out there use the Jamstack approach, but this number has doubled in the past year.

What's more telling though, is that over 15% of the Top 10K websites (by traffic) are at least partially built on some sort of Jamstack approach. It makes sense for the high-traffic sites to adopt Jamstack too, as they need a scalable, fast, website that can handle the load.

There is also a thriving ecosystem of tools and services which fit into every step of the development and hosting workflow for Jamstack sites, from Content Management Systems (to help marketing teams manage and publish content), to Static Site Generators (to help build and generate the site) to web hosting solutions and 'serverless' back-ends (to help add dynamic functionality to the site).

If we take these three data points as a sign of things to come, there is potentially a bright future for Jamstack ahead.

Netlify & Vercel: An Overview

Netlify and Vercel are platforms which provide web hosting and serverless backend solutions to deploy and host based web applications.

In a nutshell, you let Vercel or Netlify access your codebase, pre-render your website, host it, and make the process of integrating with other APIs and making dynamic back-end calls easier. Developers can focus on writing code, while these platforms handle the bulk of the heavy lifting.

Vercel and Netlify aren't alone. There are a few other platforms serving this problem space to varying degrees, such as AWS Amplify and GitHub. All of these platforms focus on the developer experience, and centre their pitch around the ease of building and shipping a website at scale.

(Note: This blog is built using Next.js, and hosted on Vercel for free. I tried both platforms out, and they are super intuitive to get started on. It took me more time to write this post than it did to set the blog up!)

VCs and big-names in tech have jumped on Vercel and Netlify betting on the potential growth in this space. Netlify is valued at $2bn and recently raised a $105M Series D, and also boasts private investors such as Tom Preston-Werner (GitHub Founder). Vercel is now valued at over $2.5bn after a recent $150M Series D round, with funding from VCs like Accel, Tiger Global and individual investors like Nat Friedman (GitHub CEO).

Let's take a look at how both Netlify and Vercel are taking markedly different approaches to capturing this market.

Vercel's Strategy: 'Ver'tical Integration

Vertical integration in the developer space is something we see happening more and more often these days. Companies building software for developers know the kind of tech or tool stack their users employ, and how it affects their workflow at different stages of the development lifecycle.

It is a powerful lever, to trap a user in your own ecosystem and influence their behaviour along the way through vertical integration. Companies can extract more revenue and improve stickiness, while creating more value for users.

Enter Next.js.

Next.js is one of the fastest growing Static Site Generators that is used to build Jamstack based websites. Close to 10% of the Top 10K sites by traffic are built with Next.js, almost double from just a year ago.

Why is Next.js so popular?

Next.js is massively popular because it is built on top of React JS. React is the most popular Javascript Library used today, with over 30% market share. Next.js enables React based sites to follow the Jamstack philosophy, by adding a host of handy features.

Naturally, riding off a behemoth like React, and solving real user problems brings Next.js a real edge.

In my opinion, what really makes Next.js attractive is that it really takes Jamstack closer to the 'silver bullet' I mentioned earlier. It brings the dynamic site loading benefits of React, and combines it with the static site generation and performance focus of Jamstack sites.

How can Vercel benefit from Next.js?

For starters, it is a foot in the door. Every Next.js user is already in some ways, a Vercel user, even if they aren't using their hosting services yet.

This gives them access to a thriving community of developers (over 30K on their Discord channel) who will give them invaluable user feedback, and help them keep a pulse on exactly what Vercel needs to build to make the Next.js (and Jamstack) development and hosting experience better.

Which leads to the natural conclusion, that Vercel and Next.js will naturally be designed to work well together, and potentially better than any other platform. Think of this like all of Android's greatest features coming to Pixel phones first, before they trickle down to other phones. A little bit of hunting on developer forums indicates that alternate hosting platforms don't fully support Next.js out-of-the-box, even if they do claim to.

A great example of this that Vercel has built is a super handy feature that allows developers edit their Next.js code, and see it update in real time, all from their web browser. It also combines collaboration features, and allows users to commit and deploy code, in an easy to use interface.

Other examples of features Vercel have built into Next.js exclusive to their platform include Performance Analytics (which sources sites performance data from actual visitors).

Will all this work?

A core part of Vercel's success is tied to the success of Next.js.

It is both a product, and a lead generation engine in its own right. It should come as no surprise that Vercel customers such as GitHub and Auth0 also use Next.js to develop their sites.

As long as Next.js continues to thrive, Vercel should also see their share of the Jamstack market grow.

However, Next.js sites can be deployed anywhere, even on Netlify. This means Vercel will need to make sure that their platform is truly home for Next.js, and build strong synergies that others will be unable to replicate (at least quickly).

Netlify's Strategy: A little bit of everything.

Netlify is an excellent platform in its own right too. They bring a lot of the same features that Vercel already has, and more: Instant deployments, support for a number development frameworks out of the box, identity management, and a couple of other handy features to make managing sites easier.

Brand Marketing 101

What is super interesting about Netlify, is how they've built theJamstack brand, and associated it with themselves. What you have to remember, is that Jamstack is a concept, an approach to building web applications. It isn't a piece of software, or a platform in its own right.

But, if a non-techie visited the Netlify or Jamstack website, they would probably do a double take to be sure. Sure, Netlify's CEO came up with the term, but the fact that Netlify has given Jamstack a logo, and even gone and trademarked it, is quite something. Imagine trying to create a trademark around using algebra to solve math problems.

The Jamstack.org website is particularly interesting. There are sections listing out various different tools to build Jamstack sites, with prominent CTAs urging users to try them out on the Netlify platform.

Jamstack references are peppered all over the Netlify website too (from their main nav bar to dedicated landing pages) leading a reader to believe that Netlify is the home for Jamstack development as a whole.

This sort of brand association is extremely hard to achieve, but if Netlify is able to make themselves synonymous with Jamstack, that's priceless. If under 1% of websites are built using Jamstack technologies today, and this trend kicks off hard, I shudder to think at how many users Netlify could acquire purely on name recall.

I'd love to see how this strategy plays out long term, and some of the web traffic stats to see how much these efforts are contributing to the growth of Netlify's brand.

However, it isn't all gimmicky marketing. Netlify is investing into the Jamstack ecosystem too. (P.S. Vertical Integration is back again)

Netlify recently announced the $10M Jamstack Innovation Fund, offering up to $100K in early investments to founders building products that serve the Jamstack developer. The fund also comes with business, product & engineering advisory services, and the power of the Netlify founder network and marketing machine.

For Netlify, this is a great opportunity to further seize the Jamstack narrative.

Netlify can't solve every single developer problem. However, what they can do, is create synergies with those who are solving crucial parts of the puzzle they aren't. With the Innovation Fund, Netlify could possibly attract a diverse selection of early-stage products, and nudge them in the right direction. Naturally, these products will be distributed via Netlify, and integrate well with the platform. Vercel has a heavyweight with Next.js, but Netlify could assemble an entire roster of featherweights to take it on.

Netlify also recently acquired 'OneGraph', a tool to help developers connect and manage a complex system of multiple APIs. Depending on how successfully Netlify is able to integrate OneGraph into their ecosystem, it could make Netlify a great place to build websites, regardless of the framework they use to develop websites.

In summary…. and what's next for these platforms?

Well done on making it this far. (I'm patting myself on the back, but you should give yourself one too.)

There's a lot more about the Vercel and Netlify products, and their competitors that I've not talked about in this post, as there is little that differentiates players like Netlify and Vercel right now, making choosing one over the other rather trivial at times.

Here's what I think will contribute to the success of platforms like these two in this space:

1. Provide a robust offering and continue to solve the core problem they set out to solve.

• Making the workflow in developing, hosting and managing websites easier, while providing best-in-class performance is essential.
• Developers demand the best experience, even from a free tier offering. And if one platform doesn't cut it, they will move, as long as it is easy to. Right now, the switching costs aren't even that high.

2. Move beyond Jamstack thinking.

• If Jamstack as a concept doesn't kick off as predicted, this leaves platforms with a weak foundation, and a limited Total Addressable Market. It is one thing to position yourself around Jamstack architecture, but another to become fully dependent on it to succeed.
• While Jamstack is being positioned as a really cool way of building websites, I still feel it isn't anything brand new, and it isn't a silver bullet to all the problems developers face today. Being stuck in this thought process could severely limit how these platforms evolve.

3. Bridge the gap between static and dynamic websites. Solve business problems, not just user problems.

• Jamstack's performance benefits hinge on serving pre-rendered, static content to users. But we live in a world where businesses are edging towards ultra-personalised digital experiences. Platforms, and Jamstack ecosystem tools need to bring static site performance without compromising on dynamic functionality. Both platforms are making strides in this already (Vercel with Next.js Edge Functions, and Netlify with its OpenGraph acquisition.)
• I don't have a crystal ball, but my gut tells me that businesses rank performance benefits highly, but personalisation will rank higher. It's all about the incremental gain/loss: imagine shaving off 1s in page load time, only for users to now take 10s more to find a product they are looking for. Sure your website loads faster, but there's a net loss of 9s in the time to viewing relevant content. (Not the best example, but I hope that illustrates the performance vs. personalisation debate.)
• Providing a clear cut reason for a paid-tier is the next big challenge. Many tools like Next.js are free to use and can be deployed anywhere. Realistically, enterprises will not immediately pack up and move to Vercel or Netlify for hosting outright because their existing setup could give them more flexibility. Making the Vercel/Netlify walled garden more enticing is key.

Fin.

And now we wait. I'm excited to see how the Jamstack ecosystem evolves in the coming years!

If you enjoyed reading this, share it! I'm always up for a chat, so feel free to reach out to me on LinkedIn or Twitter!