How we built a SaaS in 100 days.
Hey there — we decided to open the Maki hood through a series of articles we’ll be publishing on this Medium account. We’ll detail both the things we are building at Maki and how we’re doing it. And to start, we’ll tell you how we have built a SaaS in just 100 days. Don’t view this as a step-by-step guide to start your next gig, but more as a product philosophy to help you build, as fast as possible.
Start with end-users and keep obsessing about them
This might sound trivial to so many of you, but I want to insist on the importance of talking to users first. So when we started thinking about Maki, we first interviewed 100 HR directors from SMBs, trendy Next40 scale ups and very large Enterprises to get a sense of their problems. Funnily enough, we were initially thinking about the future of work but were surprised to see that most of their problems kept getting back to Recruitment. “We don’t have a clear process to recruit, it’s time-consuming, we recruit clones, we don’t foster diversity, we do very little when it comes to the candidate experience, etc.”. And when reflecting about our own experiences as Hiring Managers we couldn’t but share the problem.
Obsess about your end-users, but more important co-construct your product with them.
We then went to the drawing board, read as many books as possible (here, here & here) as well as scientific literature and when we started getting a sense for a solution, we surveyed another 100 more operational HR specialists to confirm the hypothesis and get a feel for our potential solution. This overwhelmingly confirmed our direction of helping bring more objectivity and more efficiency into the hiring process.
And so we went back, once again, to the drawing board and started designing our very first flows and screens. In parallel, we booked sessions with some of the respondents of our survey to show them our work very early on and gather feedback that way. Not only did we accumulate incredible feedback that helped us build Maki the way it is today, but we also connected with some people that would later become our customers. So I can’t stress that enough, obsess about users, and more important execute together with them.
Don’t shoot for the moon, yet. Make pragmatic choices first.
Once we had a strong enough conviction, we rolled up our sleeves and started building. And to keep our objective of launching a SaaS in 100 days, we had to formulate an operating principle that allowed us to accelerate while still providing our customers with a reliable, secured and scalable experience. In a nutshell, we decided to focus on where we truly added value and outsource pretty much everything else. Let’s take a few examples.
Focus on where you truly add value. Rely on existing bricks for the rest.
When deciding which technologies we would go and use, we did some introspection to understand where our strengths lied and looked ahead to check what would stand the test of time. We naturally opted for a combination of Node.js for backend and React for frontend both coupled with typescript. We felt this approach would scale well as it would allow to attract more talents, and the surrounding mature ecosystem and all its useful tools / libraries would help us go faster.
When scaffolding our project, we looked for ways to bootstrap services while laying down extremely strong foundations that would scale to serve millions of users. We opted to build our backend services on top of the NestJS framework which would later allow us to expand into micro-services should we chose to go that way. We implemented Hasura, a GraphQL APIs engine, to automate the exposure of our Postgres data through a graphQL layer. We leveraged MUI to power our design-system thanks to ready-to-use foundational components. And we obviously implemented some of the best-maintained open-source projects to accelerate on specific tasks.
We eventually decided to completely outsource strategic parts of our platform, leveraging world-class utilities to do so (Auth0 for authentification, Stripe for billing, SendGrid for notifications) because, let’s face it, we wouldn’t do better than these guys. We also chose to host our entire infrastructure within GCP (SQL, Cloud run, Logs, Monitoring) to ease our work between multiple services that would otherwise be spread across different providers.
So I guess, what I’m really getting to here is: focus on your business, rely on existing bricks for the rest. And for the most curious among you, we’ll detail each of these steps in dedicated articles written by our team very soon.
But most importantly, empower people and get out of the way.
Now, truth be told, we wouldn’t have been able to do this without the amazing people that have been part of the journey since day one. And since Powership (a delicious combo of Power & Ownership) is one of our core value, we gave our team the power to make autonomous decisions from day one, no questions asked. Because we believe our company is the sum of what its people think, do and decide, we hired bright builders and got out of the way so they could build.
That’s how Nicolas came from Spendesk and decided to leverage react-router, react-queries and MUI to create scalable ways to build interfaces. How JD joined us from Swile and chose Hasura as our magic GraphQL engine. Or Jérôme decided NestJS was the smartest choice to scaffold our backend services. And we repeated the pattern with Justin, Ben, Anaïs, Yann, Benoit and all the others that have joined us since.
What I’m getting to here is that if you consider your team as a collective brain, you should give them the power and the freedom to make decisions instead of telling them what to do. And yes, that means you’ll get it wrong from time to time, but overall, you’ll move quicker and more importantly, you’ll learn faster.
So my last advice to you, invest in amazing talents and build a strong culture, from day one. This is your most likely chance to actually build a SaaS in just 100 days.
