Figma is a popular tool that is used by tons of companies to design user interfaces.  If you’re spinning up a new application, Figma is a product you’d use to lay everything out visually before you start writing code.  Adobe tried to buy them to “usher in a new era of collaborative creativity,” alas that new era was canceled due to regulatory oversight.  Now Figma is going public.

Their S-1 filing reveals how massively dependent on Amazon AWS they are, and it’s a cloud case study.  They’re spending $300,000 per day on AWS services, and recently signed a 5-year deal in which they agreed to spend $545 million.

For a company with annual revenue of $750m, shipping 15% of every dollar coming through the door to Amazon is pretty hefty.

Of course, they have to spend some money on hosting their services.  The question is if AWS is the right choice.

Is that Necessary?

You may recall in 2024, 37signals (the makers of Basecamp and other web-based groupware) decided that spending $3m+ on AWS was wasteful, so they insourced everything to their own gear and hosting and cut the bill by two-thirds.

For firms that are primarily using VMs, AWS (or Azure or GCP) is probably the wrong choice.  They’re premium-priced and bandwidth is extremely expensive.  I’d argue that either renting dedicated servers and building your own fabric, or colocating gear, is a cheaper choice.  Granted, you do have to have technical staff to manage things like high availability, but you’re going to need technical staff regardless.  I’m talking about hosting VMs here, and if you’re running your own images, you need to patch them, secure them, back them up, etc.  Cloud isn’t saving you anything.

Where the analysis becomes…well, cloudy, is all the other hosted services.  DBaaS, S3, CDN, IAM, etc. are very convenient to use.  You can knit together very complicated applications using these services’ APIs.  Spinning up the same services can be done (for most things) but of course that increases the admin burden.  At what point do the lines cross, where it’s more expensive to run and maintain all of these services yourself as opposed to hosting them in the cloud?

And of course, the cloud has other advantages.  If you have a latency-sensitive application, the ability to spin up a service very close to the user base is quite convenient, as is the ability to dynamically vary the resources and sizing for that service.  If you’re Netflix, for example, the cloud is a no-brainer, because the ability to stream near each user and spin up/down the load as needed is invaluable.

But for Figma, what is the cloud buying them?  No doubt, the AWS API is a nice toolbox to work with.  But half a billion dollars’ worth of nice?

What Services is Figma Using?

Their S-1 doesn’t list the Amazon services they use, but their careers page gives us some clues.

For AWS, it mentions DynamoDB, as well as other databases that are presumably DBaaS.  Interestingly, there is also a lot of mention of NetSuite, Salesforce, and Snowflake.  The former two are just a business apps, and presumably systems they’re using to run their business.  But Snowflake is a cloud database service that presents a toolbox for people to use, so that’s another layer of cloud.

According to Figma, their entire customer database is run on a single database:

One of things people are most shocked to learn about Figma’s infrastructure is that it’s backed by a single database instance running on one of the beefiest machine in AWS.

That blog post was from back in 2019, so perhaps things have changed.

What do you think?  Is this a case where someone needs to tell the Figma CTO that the emperor has no clothes, or is this is a rational leveraging of cloud capabilities?  Let us know your opinions in the comments below!

raindog308

Raindog308 is a longtime LowEndTalk community administrator, technical writer, and self-described techno polymath. With deep roots in the *nix world, he has a passion for systems both modern and vintage, ranging from Unix, Perl, Python, and Golang to shell scripting and mainframe-era operating systems like MVS. He’s equally comfortable with relational database systems, having spent years working with Oracle, PostgreSQL, and MySQL.

As an avid user of LowEndBox providers, Raindog runs an empire of LEBs, from tiny boxes for VPNs, to mid-sized instances for application hosting, and heavyweight servers for data storage and complex databases. He brings both technical rigor and real-world experience to every piece he writes.

Beyond the command line, Raindog is a lover of German Shepherds, high-quality knives, target shooting, theology, tabletop RPGs, and hiking in deep, quiet forests.

His goal with every article is to help users, from beginners to seasoned sysadmins, get more value, performance, and enjoyment out of their infrastructure.

You can find him daily in the forums at LowEndTalk under the handle @raindog308.