TL,DR; (to avoid an irony you’ll soon discover)
We’re adding automatic summarization using GPT-3 for your user feedback, so you don’t have to read through the thousands of feedback you’ve received.
PS: We are live on ProductHunt today, with our first offering of this on our feedback Widget.
Feedback Analysis is one of our core features at Olvy.
It started with a small feature we added to our v1, sentiment analysis on top of user feedback.
It was enough for users to come back asking for more abilities on their feedback, and we could empathise with them because we too had the same problems with feedback.
We realised, to solve this problem you need to solve the organization problem, and also make that information actionable.
To make things actionable, you need organization systems you can set up to make it easy to track things and then a layer of intelligence to highlight and enrich your organised information.
This intelligence layer in Olvy has evolved over time. It started with a little bit of Google’s Cloud Natural Language APIs, then OpenAI entered our stack is currently continuously expanding.
While applications of generative AI and LLMs (large language models) are introducing completely new experiences in writing tools, design tools, and even to writing code. These ideas have gotten to marketing, design, engineering and a bunch of different tools but product workflows still haven’t seen a good level of automation.
That is our focus. We help companies build better products by helping them build a continuous user feedback loop with the help of modern tools.
Today we’re adding another amazing feature to help you understand user needs. It’s powered by GPT-3, which already helps us in feedback processing. It helps us detect user feedback from a ton of noise on social media. And today with the new release of our feedback widget we’re bringing feedback summarization to Olvy.
Every user feedback that lands on Olvy is analyzed using GPT-3 and other APIs and stored for you to read through. If you have a thousand of these feedback, nobody else in your team is going to be reading each and every one of them they’d prefer a TLDR version.
When you’re on your feedback feed, we show you a sidebar with an analysis summary of all feedback you have before you. It gives you a birds eye view of how things look, how much negative feedback you have, which users are most active, what words they’re using the most etc. But these are essentially different data attributes that you can mix and match to find your answers.
With feedback summaries you don’t have to manually go through all feedback one by one. You just click a button and Olvy will use OpenAIs embeddings and GPT-3 to generate a summary of all your user feedback in seconds.
If you want to focus on something specific you’ll just need to add filters to find what you’re looking for and summarise that.
Here is a quick video on what it looks like in action.
So the next time someone asks what is the response like on a feature in one of your meetings, you have an answer ready without calling in expert analytics help.
This is also just the beginning, we’ll soon build more experiences on top these modern systems to make your job easier. The technology problem is solved for us, Olvy is the application layer you’ve been missing.
So let us know what more you’d like us to see.