@Jagannath Ramesh

Introduction 🙋‍♂️

Easy Summaries is a Chrome extension that summarizes web content such as articles, PDFs, and YouTube videos. This is an essay of our story on how we started with the idea, how we built it, how we tried to market it, and why we decided to shut it down.

In the summer of 2023, my brother Jagannath and I were brainstorming different ideas in the AI space when it was being adopted everywhere right after the launch of ChatGPT. We did not know about fine-tuning the LLMs or building RAGs, but we had an idea. We did what we had to do—build a wrapper!

Never realized everyone and everything is a wrapper!

Never realized everyone and everything is a wrapper!

Trivia: My younger brother @Seshadri designed the extension’s user experience and interface.

Build…Iterate…Repeat… 🔂

1. Design & Competitive Analysis

Being the product mastermind, Jagan created the Google forms with some basic questions to identify the pain points in the existing products. We got 30+ responses and were able to locate a bunch of problems:

  1. The summaries were not well-structured
  2. They were wordy and took too much time. Summaries are supposed to be concise. Don’t they?
  3. Too many distractions and they were not engaging(ads, ads, and ads)

Thus, the idea of Quick and Easy Summaries(later renamed Easy Summaries) was born and designed to be simple and minimal.

responses from the survey

responses from the survey

Did we evaluate the idea without doing a competitive analysis?

No, we found numerous summary-generating extensions in the Chrome Store, but they were overly complicated—requiring users to input links, type prompts, or even provide an API key. Imagine expecting users to dig up their API keys just to generate a summary! (a big red flag) Our mission was clear: address these pain points with a simple, one-click summary solution!

2. Engineering

In a week, we created an MVP(minimum viable product) that could capture a site's contents and provide a summary.

  1. The backend written in Node JS will receive the contents of the user — say a blog.
  2. It will condense the content and send it to Open AI API with a prompt — like “Summarize this ${condensed_content}
  3. After all the processes, the backend will pass the content to display on the front end.