5 mistakes made when starting a startup
Good time of day, Habr.
During the previous year, I spent a lot of time watching movies, and since not every movie I wanted to watch then I had to spend about an hour picking up a movie (the theme is not the same, the actor is not liked, this one is very old, etc. )
As a result, in January, I thought why not create a kind of “Spotify only for the movie.”
')
It all started as usual, at first a little google, then we will think about a week about how it can be implemented and whether there will be a point in such a startup.
The scheme was pretty simple. The user enters, clicks the button and gets a “movie of his dreams” which he has not seen. To ensure the work, the service had to know what the user likes, therefore, there was a list of favorite films, and also to exclude films that the user had already seen there was a list of watched movies, quite simple, isn't it?
I will not talk for a long time, this post is not about how I ran the Spectaculr, this post is about how everything went to hell.
So, let's begin.
It is clear that in order to run such a service you need a base of films, or an API. I was not going to buy the base of films, because I thought that it would be difficult to maintain information up-to-date. Therefore, I chose the option with the API.
But I did not even think about the fact that there are not so many such APIs. As a result, I had two options:
In general, there was a feeling that you have to choose, but there is nothing to choose from. I chose TMDb because there was the functionality I needed - Discover.
And here, probably, my biggest mistake is - I didn’t test the work of this Discover API itself, because first I had to write everything else and then start working on the “crown” of the service. And when the time came, I just realized that it was impossible to use, because in response the user received absolutely not what he had to (you could be a lover of melodramas and get some comedies, for example).
Perhaps it will seem to many that this is rather trivial, but nevertheless, when I began my work on the service, I didn’t particularly think about whether this scheme would work for a movie.
Eventually:
Before that, I had never really thought about writing an application for any OSes, but then I thought it would be better to run on a mobile platform and not have a web version initially.
I chose the Android platform (not because I love it, but because there are more users), because I didn’t have knowledge, I couldn’t write a native application, I wasn’t going to hire anyone either due to lack of funds and the desire to do everything by myself.
Well, here I thought, why not make a SPA using all the advantages of web development. In general, everything turned out quite well (I thought so), then simply using an ordinary WebView in Eclipse, I created the application.
After a certain time of using the application, I concluded: it is very difficult to write a really good, non-native application. Please note that I did not use PhoneGap. Perhaps if I did this, the result would be better.
Verdict. If you have the tools or the knowledge - cut the API and write the native application, the user will definitely appreciate it, but if not, spend a little more time testing the application on different screens / devices, etc. because this is the only way to come across problems of a non-native application.
I certainly understand this feeling when you are eager to share your startup with “the whole world” and “change it", but if I could, I would return in February of this year and force myself to work another week (maybe two) and only then run
I often hear that it is better to run the so-called MVP (Minimum Viable Product), see how people react and only then if users are interested in working on a better version. I think this is not right. After all, a user who once tried your raw product will most likely not try it again.
Probably, for a start-up there is nothing more painful than the moment when, in his opinion, nobody wants to try a brilliant product. It wasn’t quite like that with this startup, given my work (small) and knowledge in PR (zero), I reached enough users to identify weak points, understand my mistakes, etc.
But nevertheless, the point is that if you do not have knowledge in this area, it is better to try to find a person who has.
And of course, think about what you will do after launch, to whom you will send a link, where you will tell about your startup, where you will order advertising and so on, otherwise the meaning of your launch, if no one knows what you are.
That's all. Of course, these are not all the mistakes that were made by me, but these are my biggest mistakes.
Well, after all this, it is worth noting that despite the fact that you have a bad product in the end and how disappointed you feel, you should think well before closing your startup, because there is a way out.
By the way, for myself, I found this exit. I spent a certain amount of my free time to develop a new version. I started from scratch, this time I used my movie database, chose NodeJS and MongoDB instead of PHP and MySQL, wrote a new design, rethought the idea, etc.
Thank you, successful startups.
Start
During the previous year, I spent a lot of time watching movies, and since not every movie I wanted to watch then I had to spend about an hour picking up a movie (the theme is not the same, the actor is not liked, this one is very old, etc. )
As a result, in January, I thought why not create a kind of “Spotify only for the movie.”
')
It all started as usual, at first a little google, then we will think about a week about how it can be implemented and whether there will be a point in such a startup.
The scheme was pretty simple. The user enters, clicks the button and gets a “movie of his dreams” which he has not seen. To ensure the work, the service had to know what the user likes, therefore, there was a list of favorite films, and also to exclude films that the user had already seen there was a list of watched movies, quite simple, isn't it?
I will not talk for a long time, this post is not about how I ran the Spectaculr, this post is about how everything went to hell.
So, let's begin.
Using someone else's API is not always good.
It is clear that in order to run such a service you need a base of films, or an API. I was not going to buy the base of films, because I thought that it would be difficult to maintain information up-to-date. Therefore, I chose the option with the API.
But I did not even think about the fact that there are not so many such APIs. As a result, I had two options:
- RottenTomatoes (poor functionality, good metadata, well-built API)
- TMDb (good functionality, average metadata, unorganized and sometimes tangled API)
In general, there was a feeling that you have to choose, but there is nothing to choose from. I chose TMDb because there was the functionality I needed - Discover.
And here, probably, my biggest mistake is - I didn’t test the work of this Discover API itself, because first I had to write everything else and then start working on the “crown” of the service. And when the time came, I just realized that it was impossible to use, because in response the user received absolutely not what he had to (you could be a lover of melodramas and get some comedies, for example).
If it works in one industry, it does not mean that it will work in the second
Perhaps it will seem to many that this is rather trivial, but nevertheless, when I began my work on the service, I didn’t particularly think about whether this scheme would work for a movie.
Eventually:
- Not every person wants to watch old movies (people who only watch in cinemas and only freshly released movies, as well as people who see the 2008 film as old). After a little analytics, it became clear that 1/3 of users just want to have a convenient service for storing watched / favorite movies, as well as viewing information about different kinas (sorry)
- In Spotify, a person can not only find music for himself, but also listen, and since there was no opportunity to watch a movie in the service, some people lost interest
Native application against web
Before that, I had never really thought about writing an application for any OSes, but then I thought it would be better to run on a mobile platform and not have a web version initially.
I chose the Android platform (not because I love it, but because there are more users), because I didn’t have knowledge, I couldn’t write a native application, I wasn’t going to hire anyone either due to lack of funds and the desire to do everything by myself.
Well, here I thought, why not make a SPA using all the advantages of web development. In general, everything turned out quite well (I thought so), then simply using an ordinary WebView in Eclipse, I created the application.
After a certain time of using the application, I concluded: it is very difficult to write a really good, non-native application. Please note that I did not use PhoneGap. Perhaps if I did this, the result would be better.
Verdict. If you have the tools or the knowledge - cut the API and write the native application, the user will definitely appreciate it, but if not, spend a little more time testing the application on different screens / devices, etc. because this is the only way to come across problems of a non-native application.
Do not rush, spend more on development
I certainly understand this feeling when you are eager to share your startup with “the whole world” and “change it", but if I could, I would return in February of this year and force myself to work another week (maybe two) and only then run
I often hear that it is better to run the so-called MVP (Minimum Viable Product), see how people react and only then if users are interested in working on a better version. I think this is not right. After all, a user who once tried your raw product will most likely not try it again.
Think about where and how to promote
Probably, for a start-up there is nothing more painful than the moment when, in his opinion, nobody wants to try a brilliant product. It wasn’t quite like that with this startup, given my work (small) and knowledge in PR (zero), I reached enough users to identify weak points, understand my mistakes, etc.
But nevertheless, the point is that if you do not have knowledge in this area, it is better to try to find a person who has.
And of course, think about what you will do after launch, to whom you will send a link, where you will tell about your startup, where you will order advertising and so on, otherwise the meaning of your launch, if no one knows what you are.
The end
That's all. Of course, these are not all the mistakes that were made by me, but these are my biggest mistakes.
Well, after all this, it is worth noting that despite the fact that you have a bad product in the end and how disappointed you feel, you should think well before closing your startup, because there is a way out.
By the way, for myself, I found this exit. I spent a certain amount of my free time to develop a new version. I started from scratch, this time I used my movie database, chose NodeJS and MongoDB instead of PHP and MySQL, wrote a new design, rethought the idea, etc.
Thank you, successful startups.
Source: https://habr.com/ru/post/224903/
All Articles