DEV Community

Cover image for 3 Engineering Mistakes That Kill Startups

3 Engineering Mistakes That Kill Startups

Michael Lin on June 08, 2023

I’ve been doing engineering consulting in fractional CTO roles for about a year now. And my current work is vastly different from my previous work ...
Collapse
 
stealthmusic profile image
Jan Wedel • Edited

This is exactly what killed the Startup I worked for after university.

We created a huge list of features, designed everything in a perfect waterfall manner. We developed almost two years without ever having a paying customer. It was a great architecture, as you said “perfect”, but it was not what the customer needed.

In fact, customers only needed 5% of the features that we could have delivered 1,5years earlier.

Building something in an agile way, starting with the most valuable parts and getting frequent customer feedback could have saved the company.

I still think it was a good idea…

Collapse
 
scanepa profile image
Stefano Canepa

I suspect we, as developers, should stress the proper agile values more. Following the simple principles of the agile manifesto would avoid all the 3 mistakes highlighted in this article and more.

I worked for companies where agile meant just chaotic anarchic addition of features based on what the single developer liked to hack or projects focused on filling up the agile management tool fields in a way developers thought top managers wanted to read.

Collapse
 
stealthmusic profile image
Jan Wedel • Edited

Yes!

After working for this company for about a year, I figured that our whole product development „process“ does not work.

I started reading about XP back in the days and thought this is exactly what we need.

I tried to convince our CEO to give it a try. He thought Agile means „developers do what they want when they want and you’ll never know when anything is done“.

Sadly, I wasn’t experienced enough to convince him of the opposite. I just got another well-payed job and left. 3 months after the company was bankrupt and the CEOs personal and financial life was devastated.

Collapse
 
_michaellin profile image
Michael Lin

That would definitely help!

Collapse
 
_michaellin profile image
Michael Lin

Very common story - thanks for sharing!

Collapse
 
jaustinuf profile image
jaustinUF • Edited

Excellent points! As a many-years tech-type I’m still amazed by how ‘entrepreneurs’ can develop ideas and get money (sometimes lots of money!) with no real evidence of what might sell.

All these points say get something out there (that shows the basic service) ASAP to see if people will use it/pay for it. If I were a VC I’d require every proposal to be based on the Door Dash story.

Collapse
 
_michaellin profile image
Michael Lin

Totally - I love how Door Dash handled their MVP!

Collapse
 
matpk profile image
Matheus Adorni Dardenne

Not only did I fail to get a single upvote, I also got banned from the Subreddit for spam 😢

Can relate. Reddit sucks.

Collapse
 
_michaellin profile image
Michael Lin

haha yup

Collapse
 
iq9 profile image
Russ Brooks

Yup!

  • I'd rather ship it 75% perfect now, then 100% perfect never.

I keep reminding less-experienced devs, Code is a wiki. It's constantly honed over time by the "crowd" that is your devs.

Collapse
 
_michaellin profile image
Michael Lin

100%!

Collapse
 
dangernoodle profile image
Eszter Szücs-Mátyás

Great! And make sure the POC works well. I have seen so many startups fail because they were faster than their own thoughts.

Collapse
 
_michaellin profile image
Michael Lin

100% Eszter!

Collapse
 
fruntend profile image
fruntend

Сongratulations 🥳! Your article hit the top posts for the week - dev.to/fruntend/top-10-posts-for-f...
Keep it up 👍

Collapse
 
_michaellin profile image
Michael Lin

Yessss .... thank you!

Collapse
 
scanepa profile image
Stefano Canepa

Big corporations make these three mistakes at the project level.

Collapse
 
wadecodez profile image
Wade Zimmerman

Long story short don't aim for perfection. Try to build your start up based on passion and the interest of solving a problem. Whenever you aim to meet demands you will always fall short.

Collapse
 
_michaellin profile image
Michael Lin

Pretty much!

Collapse
 
montyharper profile image
Monty Harper

I'm just one guy, learning how to build an app for a single user, and keeping these three things in mind will be very helpful. Thanks!

Collapse
 
clarance_farley_c14c5d02f profile image
Clarance Farley

...