Scale or Fail: How to Build Processes and Mechanisms the Amazon Way

Photo by Austin Distel on Unsplash

Guest post by Richard Howard, AWS Startup Business Development 

“Scaling” is a term that’s ubiquitous in the startup world. Founders are always talking about how they scaled up their infrastructure to cope with a spike in usage or why they scaled up their hiring after a big round of fundraising. On the flip side, founders also bemoan having to let people go because they couldn’t “scale with the business.”

Despite founding my own startup several years ago, I ironically learned how to scale as a business development manager at Amazon, a little company that employs over 500,000 people. I want to share my journey with you, as not understanding the basics of scaling can be the death of your startup.

I started at Amazon’s Web Services division (AWS) roughly three years ago after six years at various startups, including my own. All of those startups were at the very early stages when I joined—a time when the axioms “Getting sh*t done” and “Done is better than perfect” are most apt.

When I started, the AWS EMEA Startup team really did operate like an early-stage startup. There were around 12 of us, sprinkled throughout EMEA to cover all of the startups on two and a half continents. We were encouraged and goaled on meeting as many startups as possible in order to help them and influence them to use AWS. Our jobs were incredibly broad and somewhat ambiguous and I loved it. It was the perfect combination for a former CEO.

When the team and our ambitions grew, we needed to become more “Amazonian” and create processes and mechanisms that meant that we could work with the ever-increasing number of startups in a more scalable way.

Because of my years of just getting it done, I was stuck in the old way of doing things. It was comfortable, and I loved meeting and helping entrepreneurs. What I’d failed to realize, however, was that to be really successful and help as many customers as possible, I needed to scale the impact I was having.

How does this apply to founders? The endorphin rush that you get from successful one-to-one customer interactions won’t sustain your startup, no matter how great it feels in the moment. You need to constantly be thinking about how to inject scalability into what you’re doing.

Set a breaking point

Startups are all about having hypotheses and testing them out. It makes sense to do things that don’t scale at the beginning. Brian Chesky went to New York to take photos for early Airbnb customers. Whilst it might not make sense just yet to process-ize what you’re doing, you should give yourself a target—whether that’s total number of users, growth numbers, revenue numbers, or something else—that is the point at which you’ll remove yourself from the process and make it truly scalable. And you should be thinking the whole time about what this process will look like when it’s implemented in a scalable way. How are you actually going to do it? Do you need to hire more people? Can it be automated? Can you outsource it?

Apply trial and error

Despite how much you think about what your new process is going to be, it’s unlikely that you’ll hit a home run straight away. Don’t feel too married to the new process if some of it isn’t working. Experiment, iterate, and improve until you’re truly happy with how it’s working. Also, know that even though your latest plan may be working well at the present, it might need changing all over again when you hit that next growth curve.

Expect discomfort

It feels great to speak to customers directly and get immediate and actionable feedback. But doing that is only going to feel great right up until the point that your startup dies. I was lucky that I was at Amazon when I learned this lesson and could take my time. If you’re running your own startup, you don’t have that kind of time.

If you really love the direct interaction with customers, putting a layer of process between you and them will feel unnatural and it’s going to make you feel removed from what you used to think of as the ‘pulse’ of your business. But it is absolutely critical if you’re going to scale up successfully.

Keep your hands dirty

Adding scalable processes doesn’t mean you stay away from getting your hands dirty and can never speak to customers ever again. It just means that you’ve recognized that to truly service as many customers as possible and grow your company, you’re going to need to stop doing everything so manually. Don’t worry though: your startup is never too big for you to still occasionally get involved in the nitty gritty. After all, if Deliveroo Co-Founder and CEO Will Shu can still make deliveries every fortnight, you can do some heavy lifting as well.

Conclusion

Pulling yourself away from customers can be a scary proposition for any founder. It’s where you really have to trust the people you’ve hired to relay the passion for the product that you have. But if you’re going to scale and become the kind of company that you want to be, then it’s imperative to do it. Building processes and mechanisms are just part of your startup growing up. This applies across the board from customer interactions to hiring to infrastructure.

The need to scale up on these things is a sign that something is going right at your startup. Embrace it. If you implement the processes in the right way, it can be an incredible growth engine in itself and allow you to focus less on the day-to-day and more on the future. And remember, just because you’re scaling up doesn’t mean you should’t get your hands dirty every now and then.