This January's Mobile Monday brought together a range of kick ass specialists from different startups (Fortumo, Pipedrive, Testlio and Messente) to give some insights to the important question of choosing a platform for your startups and how much weight does this decision carry. It is a burning question as it will determine how you'll do both in the short and long run.
Martin Henk from Pipedrive lead the evening in with a very simple analogy: the Marshmallow Test. Originally used to measure self-control in a series of tests in Stanford study, it featured children who were given one marshmallow and told that if they managed not to eat it for 15 minutes, they'll be given a second marshmallow too. Good things come to those who wait ey? In the Stanford test, those who did wait were more successful at school and career.
However, the Marshmallow Test in startups is not that straightforward: should you get something – even a prototype – completed and out asap or hold out and tweak your product? Should you choose a platform that is easier because your friend is really good at it and could get the product up and running in a week, you would get your first user etc, or choose something that would make more sense in the long run but your friend has to learn it or you have to find someone else to do it, yet you'd have a more stable product? So many options!
The evening's speakers had different experiences from the opposite end of the spectrum. For an example, Messente got their product out quickly and thanks to that, quickly discovered the bottle necks which lead to writing a new and better code and better product. But there were also those who benefitted from more preparation. Taavet Tamm from Fortumo pointed out had they thought about the question of scaling in the past, the present would've been easier.
But why is it so important to start with the right software platform choice? A good example was in the room: the oldest code still in use dates back to 2008. It makes sense to choose a platform that doesn't become completely obsolete in the long run. Yet, it's not wise to be paralized by this question - sometimes the "F*ck it, let's do it" will get you further as the user really doesn't care about what's going on behind the scenes, the user cares if the product is fast and usable. Look around who (which specialists) are around you and get a move on because the worst decision is avoiding one.
The evening ended with panelists sharing insights on what was the biggest surprise in the last 5 years. For most, it was that Java would still be "a thing". Some admitted seriously underestimating the public cloud. One misses physical keyboards on smartphones.
Last but not least, here are the slides from the speakers. Stay tuned for the next MoMo!