About Video To Blog
Hi, my name is Andrew Petrovics. I am the creator of Video To Blog.
I created Video To Blog as a way for YouTube creators to easily create high quality, SEO optimized blogs from their videos with least amount of effort.
Video To Blog is trusted by thousands of users. We are completely independent, self-funded and bootstrapped. Our legal entity is incorporated in US. As of now, I am the sole member of our team, but hopefully that will change as we grow.
Our Focus
High Quality Content
In our view, the quality of a blog can be measured by how much value the blog brings to the reader. While this largely depends on the content itself, it also depends on how that content is presented (which is where our focus is concerned since the core content itself comes from the video).
In order to bring the maximum value to the reader, our view is that the content should be well organzied (clear and sensible headers), easily digestable (no long drawn out paragraphs), easily scannable (lots of headers / subheaders), and (when appropriate) contain relevant visuals (i.e. screenshots/images).
This is exactly the kind of content we strive to produce with our AI (while still giving users the flexibility adjust the generated content as they see fit). We are constantly testing, iterating, and improving our AI prompts to produce the best quality content possible.
SEO Optimized
Equally important is our focus on ensuring all generated blogs are optimized for SEO in the best possible way. In order for a blog to bring value to a reader, the reader must find it. This is why SEO is improtant.
While optimizing written content for SEO can involve a lot (keyword optimization, metadata optimization, image alt tags, internal/external linking, and much more), our goal is to automate and abstract as much of this as possible so it becomes something our users do not have to worry about.
Minimal Effort
Our other main focus is the ability to accomplish the above goals (produce high quality, SEO optimized content) with the least amount of effort required from our users.
Our north star is an experience where users can simply input a YouTube URL, proofread the generated blog, and click a button to publish to their website and that is it.
We are continuously inching towards this vision with new features like auto screenshots, auto metadata generation, generated AI featured images, auto linking, numerous website integrations, and more.
How It Works
In short, the way Video To Blog works is by transcribing a video and sending that transcript to AI with instructions to write a blog based on the transcript.
Is it really that simple?
The actual implementation is much more complex, but at its core, yes, it is that simple.
What AI models do you use?
At the time of this writing (5/21/2024), we are using OpenAI (GPT-3.5 and GPT-4) for the blog generations, Claude (Haiku) for outline generations and auto screenshots, and DALL-E-3 for image generation.
We are constantly testing and evaluating various AIs so it is very likely we will introduce other AIs for various tasks soon (especially since Google's Gemini is starting to gain traction).