Wondering

Should I join a startup as a junior data scientist?

Junior data scientists and recent graduates can struggle to find their first job. Often, startups are seen as good targets as they can not compete financially to attract the top talent and can see a young and motivated data scientist as a cheaper alternative to contribute to their data needs.

But are startups advised as a first job to break into the data space?

You will (likely) be alone

Start-ups, in particular in the early stages, have very low headcounts. You will likely be the only data scientist and might report to someone who is not an expert. It can be to the CTO who will at least have a vague idea of how things should work and help you with technical stuff or the CMO who will be most interested in results and not on how to set up a full data collection pipeline. In such a case, they and you both will be interested in How to Make Bridges between Data and Business Functions.

“Building a data-driven organization generally starts by having a good data collection pipeline.”

All these elements should be taken into account and you should assess whether this matches your skill level and your ambition. Working in a startup where nobody has an idea of how things should be done (and neither do you) will likely be a bad and unfruitful experience. On the other hand, working with a technical person or someone who has seen how data departments are built in other organisations can be a challenging yet rewarding experience.

What you should ask:

  • Who will you report to? Analyse this person’s background and discuss with them in order to understand how they can help you. Maybe they have an idea on how to get started and have already collaborated with data scientists in the past, which is a good start. Or maybe it is a person who expects you to start tomorrow and deliver results the day after that. In such a case, politely passing might be the wisest decision.
  • How many other data persons are there and will there be? This will give you an idea of who would be able to help you and how do they expect the unit to grow. Are they recruiting someone else at the moment to help you or is growing the team tied to increasing results or a future founding round?

Ultimately, depending on the answer you get to these questions, you will have to decide whether you feel you can achieve what is expected of you under the current conditions. Facing some challenges can be scary but is a good way to grow. But starting in an initiative doomed to fail, can be a demotivating first experience.

They likely need a data engineer

Some companies want to start doing stuff with data without knowing what they want or how it is supposed to be done. This can happen when none of the founders, executives or directors actually have some experience working with or in a data team.

A robust and well designed data collection pipeline is the key to efficient and reliable downstream data analyses.

The 5 Data Profiles
Photo by Christina Morillo from Pexels

“You will likely be the only data scientist and might report to someone who is not an expert.”

And if this is the case, likely, they haven’t considered the problem as a whole. Building a data-driven organization generally starts by having a good data collection pipeline. How to achieve this will depend a lot on the industry you work in but if there is no reliable source of data, your first job will be to build this pipeline before considering doing anything else.

The problem is then twofold: 1) data engineering skills are very specific skills and 2) the company might not even be aware what they need is not a data scientist but rather a data engineer.

What to ask:

  • How is their data currently collected and stored? If they can’t answer this or are very fuzzy with the details, they likely have nothing working yet and it will be your role building it. It can be a cool challenge of course and can be feasible depending on the volume and type of data, but it should be clear to you which kind of tasks awaits you.
  • What will be your duties? If they focus only on results (e.g. building a dashboard or performing a given analysis) but don’t have a data warehouse already in place, you know your role will include some education and managing expectations as these results won’t be generated without some proper data to start with.

Again, your decision will rely on personal criteria and how you feel you can deal with the challenges you will face. But you will make a much more informed decision by gaining as much context as possible.

You will learn a lot

In any event, regardless of how well prepared the startup is to incorporate you, you will face multiple challenges when joining a startup. By definition, these structures are much more likely to often redefine their priorities or even products. And you will have to adapt by working on your prioritization skills and learning new skills as the analytical needs change over time. And you will often have to do it without a more senior person to guide you. This can seem scary but that also means you will learn a lot and will develop in two years a wider array of skills than someone who joins a bigger corporation working on the same problem during the same amount of time.

Some people will think “That sounds like an exciting opportunity” while others will see this as a nightmare when reading this.

Depending on which of these two categories you fall into, you will have a clearer idea of whether taking a startup job with little to no supervision is for you. It is perfectly fine to prefer to be surrounded by other team members to better learn during your first years and knowing which environment you need to thrive in, and knowing your weaknesses and strengths can only help you in the long term.

Salaries might be low, with potential upsides

Early-stage startups generally can’t afford to pay top market rate salaries and this is also why they often consider hiring junior profiles for a lower rate. This is unfortunately often the case although well-founded startups at later stages (or with a functioning and growing product) can be more generous in terms of salaries and perks.

But working for a successful start-up can have some great upside if part of your compensation includes shares or stock options. Owning some parts of the company (or having some options you can exercise later), can provide a future big gain if the company becomes very successful. It of course depends on how many shares you get and how early or late you join the company.

For example, if you join a very early stage startup and get 1% of the company shares (even as stock options) you could find yourself with a good amount of cash if the company is later sold or enter the stock market at a very high valuation. These gains are of course highly hypothetical as many early-stage startups die early or are sold at a low valuation.

Joining a later stage startup will offer fewer opportunities for big equity cash out but such an event is more likely to happen if the company is already established.

In any event, stock options should not be seen as part of the base compensation but rather a complement to drive even more engagement with the company as you also have a strong financial incentive for the company to succeed. Do not accept having a part of your salary decreased and being replaced by stocks just because the company does not want to pay you more. Also, make sure you understand how stock options work as there are a few pitfalls you should be aware of.

Ready to take the leap?

Startup jobs can be unique opportunities or pure traps. And even with the best conditions, they are not done for everybody because of the often-changing priorities and very specific work conditions.

If you consider joining a start-up, make sure you understand where you are stepping and whether it matches your skillset, your ambition and growth capacities. If you’re in, you might be up for the ride of your life.

1 Comment

  • Reallly interesting post guys! It is not only about the company choosing the right candidate, it is also about us candidates choosing the right place to work. Perks is also something to take in consideration, some startups know it and work on that very well.

Leave a Reply

Your email address will not be published. Required fields are marked *