Hacker News new | ask | show | jobs
Show HN: DataDrivenJS, track and read analytics data with JS
3 points by lu_ 2817 days ago
Hello!

I've been running an analytics company for a couple of years now, and I realised that the data, generated by websites users, is locked in GA or Adobe and there's no easy way for web developers to access it.

So I came up with an idea of DataDrivenJS - a serverless data service that works like web analytics, but its users are developers, not analysts, and the end-users are the very same people who populate the service with data - the website visitors.

You could use DataDrivenJS to quickly add product recommendations, personalize UI, or simply add social proofs. All that without adding any stress to your core stack.

The list of features is quite short. With a light-weight JS library you can: - track visit data - read some of the tracked data

The requests to read data require no authorisation or secret keys, but the very first request must be authorised using a web console in order to create a publicly available data feed. The data feeds contain cached, aggregated results so the read requests are very fast. Obviously, the use of a caching layer, necessary to handle read requests at scale, excludes some of the use cases.

I'd love to find out if the service - or my description of it - makes any sense to you. You can see some working examples on the website:

https://datadrivenjs.com

The service needs a couple more weeks before it can be safely used on production, but if anyone would like to play with it, I'll be happy to provide access right away.

many thanks! Lu

1 comments

Interesting concept. I'm not sure how I feel about the website analytics being global. I think have a simple API for individual users and personalizing their experience could be useful.
Thanks! You could use the service this way already. The last example here https://datadrivenjs.com/docs.html shows how to pull visitors last visit date. Of course, you could just as well pull the list of viewed products, last read article or the navigation items used most often by the user. Do you have any specific use cases in mind?