- Published on
How did I start building saas — tweelog!
- Pramod Kumar
I recently started being active on Twitter. I quickly started posting whatever comes to my mind at that point in time as tweets. It felt so good that Twitter posts are by nature short in length, the content is to the point as the characters are limited and the content on Twitter itself is floating ie, the life span of a Twitter post is very short.
On the other hand, even before being active on Twitter, I used to write blogs on Medium. I like writing about my learnings and experiences. There are two major problems in writing medium blog posts for me
- figuring out the topic that I want to write
- noting down the main points I should cover in the blog post
As I started tweeting whatever comes to my mind and forget it after that, I felt the entire process is so inefficient because I have already shipped the core content to Twitter as tweets. I quickly realized that I should be able to somehow use my tweets to solve the above two problems and publish them on Medium.
I spent some time exploring the idea. There are so many people who are very active on Twitter who tweets very valuable content across the streams. They usually have tweets anywhere above 8k on average. It is a lot of content. Another point is, these tweets are a super short form of what they want to tell. So I figured out this tool if I build it, will be helpful for people who want to create other forms of content, it may be a blog post, LinkedIn post, Twitter thread, newsletter, or even a script for YouTube video.
This is what I visualized this tool would be doing
- Pull the tweets of a particular Twitter user
- Group them into topics
- Show analytics on topics to figure out which is popular
- For a topic, show all the related tweets
- An integrated editor from which I can publish directly to Medium
I wanted to validate this idea with actual real-world users. I quickly spent half a day time to explore Twitter APIs and implemented a script that gives topic-level information in a csv format. I generated this csv for multiple people who are active in my Twitter circle and sent them DM with csv. 80% of people liked it to my surprise, but still, I was not sure if it is worth spending time on it.
After a week of thinking about it, I decided to give it a try. I started building the actual tool. In the process, I built a team of 2 amazing people so that I can focus on marketing and sales.
On the 17th of February, the dev version was ready. We tested it internally, made sure the futures are working fine in general. We kept re-iterating the tool to make sure it is usable. Eventually, we released the very first and raw version to the public on https://tweelog.app. The story just starts now.
I started messaging people again to see if they like the actual tool. People, in general, liked it. In fact, there were people who liked it and reported a few bugs ;) I was happy to take that feedback and fix them quickly. Still, a long way to go! Building a product is not the same as marketing and selling it. I am still learning it. Looking forward!
In fact, I used Tweelog to write this post. Tweelog says I have tweeted 9 times about Tweelog in very recent time. So, I thought why don’t I try it out myself. I used 4 of those tweets to build this post. Hope you liked it!
Check it out here — https://tweelog.app