In today's business environment, no piece of information is too small. No piece of information--at least, when combined with other pieces of information--is insignificant. It's all about big data, but they don't call it "big" for nothing. Some companies don't have the resources to develop their own analytics infrastructure. Service offerings can fill the gap, but only if the services can meet companies' many--and ever-changing--needs. Enter Keen IO, which brings to the table scalable customization and calibration, along with simplicity in implementation.
Keen IO collects and stores massive quantities of event data (big data)--all the interactions that happen during the course of a day. Users can determine the type of events and data they want to track--anything from signups to impressions to errors--and can use either use a simple REST API or pre-made client SDK. "We believe in the power of data to uncover new truths about what’s important in your application," said Keen representatives in an email exchange with ProgrammableWeb, adding:
Keen IO makes APIs to collect, analyze and visualize data. Keen IO is unique because of how flexible it is. We're analytics by API – we don't have a pre-defined solution or a finite set of use cases. Anything related to event data can be tracked and built on top of our platform. This empowers developers to collect and explore their data. Compared to a lot of solutions out there, with proprietary and closed solution sets, our platform approach allows for an amazing amount of flexibility, control, and customization for our users. On top of this, we make it incredibly easy to get set up and start sending events.
What makes Keen IO extremely robust is its ability to store arbitrary JSON data, based on custom properties and attributes, that makes sense to you. Users do not have to worry about how big data is stored and scaled. The complexities involved in configuring complicated data analysis tools while balancing multiple servers is taken out of the equation--enabling even small companies and startups to focus on critical business matters and building their value proposition.
Keen IO Features
Keen IO's model is to expose APIs that will allow users to perform data collection and data modeling, as well as data analysis.
Keen will help get you started with numerous templates, to start you on your way to building a killer dashboard
Keen also provides bulk loading of existing historical data events, through Keen’s official Ruby client,or Python client, or by directing using the REST API POST feature, allowing companies to leverage their already-mined analytical data.
In fact, Keen is planning to boost its backend infrastructure to support companies' growing--and increasingly dynamic--data needs. According to Keen:
Our customer base is growing quickly, so one of the biggest areas of focus for us in the next 6-12 months will be scaling our backend infrastructure to accommodate the surge in activity – ensuring optimal speed, performance, and reliability is super important to us, and we are working on a lot of things to make sure we sustain the same amazing experience for our customers.
Keen IO has been – and always will be – first and foremost about Analytics for Developers. But we're also working on some new tools that will be beneficial for every member of a company who wants to get value out of their data. Since we exist to empower people to collect and explore their own universe of data, we’re rolling out a set of new tools to level-up the data exploration side, and then taking it one step further: open sourcing it all, so users can share this power with their customers and teammates.
We’ve also updated our JS library (currently in beta), which includes two new tools - dataviz and dataset. These brand-new tools can be used for creating dynamic, highly-customizable data visualizations.
...we are looking at ways to create open source projects that track the most common data points for websites and give their owners insights into usage and retention.
We also want to start collecting and sharing more open data sets - similar to what we did recently with the New York Times Military Surplus Gear Data.
There is a wealth of amazing, interesting data out there that is too often locked away or left untouched. We want to lead the effort to surface this data, provide broad access to it, and empower people to play around with it and learn from it. We’re doing this in the hopes that people will leverage the incredible power and potential of data to discover really interesting things that we haven’t even thought of yet.
Keen IO is available in a free developer-tier edition, capped at 50,000 events per month. This will allow companies to dip their toes in the water and gauge whether the Keen IO service proposition is suitable for their needs. The next pricing tier is the Startup tier, which costs $20 per month. The pricing model scales from there, and companies can add capacity and functionality as their analytics needs change and grow.
Getting Started with Keen
1. Once you have registered for a developer account and created a new project, you are presented with the Overview dashboard, as shown below:
2. Take note of your API Keys, which are obtainable from the dashboard, which you will shortly use.
Working with Events
The core of Keen IO is on tracking Events, triggered actions that you are interested in following, with similar events grouped together in Event Collections. We will create a sample collection and event, called enrolments and enrolment1, respectively. Here we create a json file containing an enrollment event:
"class": "computer studies",
Using cURL, we will post the event, using the Project_ID and Write_Key which you can obtain from your dashboard, and our newly named EVENT Collection. We also pass the JSON file we just created above.
If your event posting went OK, you will get the confirmation JSON response “created: true” returned. Refreshing our dashboard, we can see that the event collection and event were registered:
There are tons of methods we can look at using to analyze APIs, but to illustrate how easy it is, you can either once again use cURL, as follows:
Or, from the Dashboard going WorkBench and visually creating your query: