When we first started Nepsu, I did not know much about how to successfully run a startup. So let’s say we did some big mistakes at first (that you should avoid).

But before starting, I would say this guide is not for you if you do not like to work hard and learn things on the road.

My partner Julien and I were the only two members of the team with an engineering background, and none of us were specialized in electronic design (He was in aerospace and I was mechanical).

This, combined with the fear of our incapacity to design an electrical circuit within the industry’s standards, got us to conclude that we needed to outsource this complex task, which led us to spend roughly 50000$ to get the PCB design outsourced.

Guess what? We were dead wrong, and here is why.

As a startup, you obviously need to define what your product or service is. This is usually done in the early days, and it will inevitably evolve along the way. Once you know approximately what you’re working on, you can split the workload among your teammates.

Whether you are programming a software, building a physical product, or providing a service, you will need some skills that you do not have yet.

In our case, everything related to electrical design was fairly new and at first, we could not see the benefit of learning these skills to do it in-house. Our reasoning was something like this:

Learning everything will take way too much time compared to outsourcing it. Thus, we have more chances of missing the window of opportunity related to our idea if we take the long way.

This was quite inaccurate for the following reasons:


Outsourcing means getting the job done outside the company. We can assume that this is less efficient in most cases on the following aspects: communications, overall costs, application of the design intent, long term evolution of the product and many more.

Even if you do find a freelancer to design a PCB for less than a hundred bucks on Elance or ODesk, you will have to build physical prototypes, program the controller, debug and assemble the PCB in the product and so on. These steps are a lot easier to go through when you designed the circuit yourself.

If, on the other hand, you want to outsource the whole task, consider that an engineer will cost at least 50$ per hour or 2000$ per week (and we are not talking about the expert engineer here).

When your product is slightly complex, you should consider a few months of development: specifications definition and approbation, circuit design, prototyping, and testing (this is only one iteration! We needed four for the Triangle speaker). This solution will easily cost you tens of thousands of dollars you probably do not have, unless you are heavily financed by angel investors.


Other than your teammates, nobody else will be willing to work up to 100 hours per week to get things done, and to be honest, this is what is needed. I’m a big fan of Elon Musk and he has a nice quote on this topic.


Even more important, as I mentioned earlier, every startup needs to adjust its value proposition along the way. This is in fact the basic principle behind The Lean Startup movement: get constant feedback from real customers and iterate accordingly. When you take this in the equation, you will find it is faster to learn what you need to be able to do these tests and iterations by yourself.

In our case, this was mainly true for hardware design, but we can apply it to software programming as well. Let’s take the creation of Nepsu’s website as an example.

At first, I did know some notions about HTML and CSS and was willing to learn JavaScript, but I’m colorblind and didn’t have any skills in website design (to give you an idea, I was planning on integrating some background music on the landing page… in 2014!). It was not auguring well and my partners did not have much interest in designing the website at that time.

As a matter of fact, we went to a small agency to have them create a nice responsive landing page for a very decent price.

The thing is, a few months later, we could not afford outsourcing the website maintenance anymore so we started to do it by ourselves. We were patching as much as possible but after a while, we decided to program a completely new site by ourselves. As a result, we now have full control on what can be displayed.

If we want to test some new features such as embedding an Instagram gallery or changing keywords for SEO ranking, we can do it without having to contact someone outside the company, which may take time to answer and may prioritize other clients with more urgent demands (more $$$).

In our case, we had access to a government grant to hire a fantastic graphic designer, but if we didn’t have that chance, we would have googled “best website design 2016” to find inspiration and then ask one or two of our talented friends for some help.


You can get help for free. Yes, FREE! It is really important to understand that you are not a hero nor a way too smart person who does not need any help from anyone. But as you may know, you cannot afford to pay for help (even if you think you can, you should not).

Why? Simply because you are a startup and every penny can make a difference for your future. So instead of asking a consultant for services, search for friends (and expert in a skill you need) to come and assist you.

Our product, the Triangle, would not exist today if I didn’t ask help from some friends and their friends!

You are never too smart to ask for help and unfortunately you’re probably just not rich enough to pay for it.

But ask, just ask! It can save you a tremendous amount of time! And as you may know, time is your #1 resource!

So when I say do not outsource, I do not mean to not seek advice, help or knowledge from others. It means you do not have the budget to pay people to execute the task at hand really fast for you. Instead, you should ask your connections to share their knowledge and then execute fast!

There is still so much to say, but to keep it short for now, if I could give you only one advice it would go as follow: If you are starting a business, for each workload you are about to outsource, stop, sit down with your team, and think about a way to do it in-house (flexibility, optimisation, speed, ROI).

What skills and tools are you missing?

Do you have someone in your network who can help you get started?

Will you possibly have to adjust and iterate before getting the perfect result?

And if you have no other way than to outsource it, will it worth the investment?

If you do that seriously, you will save a ton of money that you may not even have yet, and you will learn a lot more as a bonus!

Leveraging your contacts and friends

To help you get started, here is an easy way to find contacts (friends, connections) who could help you:

1- Go to LinkedIn (assuming you have a LinkedIn profile, if not, create one!)

2- Search for keywords like “electronics”, “PCB design” and “electrical engineer” (for example)

3- See which of your 1st degree and 2nd degree contacts pops out and ask them for help

4- Always be clear about what you are asking for and what you’re offering in return.

5- If you don’t know a person very well and do not have their email, use Email Hunter to find his/her email, it’s a little app that embed itself in LinkedIn and find the email of your contacts (if you don’t have LinkedIn gold).

Finally, other sources of advice are all blog posts from startups’ founders sharing their past experiences such as this one. In this stream of thoughts, I am currently working on another post which will take form of a crash-course on hardware design, sharing some basic notions with entrepreneurs who are at the same point I was when starting our project.

To conclude, I would love to hear your opinions and/or story about outsourcing.

Article credit : medium.com
There are currently no comments.