Editorial Dashboard
In digital publishing analytics have already become extremely important component. Mainly because analytics have been used to determine which posts (article, teaser) are resonating most with the audience and when.
For a digital publisher an editorial calendar is a must-have, and right analytic tool will confirm or deny the gut feeling of the journalist and/or publisher. Furthermore a right tool will help to make quick decision about whether content chosen for homepage is performing well and should occupy precious spots or if it performing poorly and should be replaced.
Business challenge
Our customer have had developed (by Datamart) initial version of Editorial Dashboard some years ago. When the product initially appeared on the market - it was very modern and effective tool.
Though time passes, and technology evolves every day. Solutions and products that looked awesome only 3-5 years ago now might be perceived as outdated. That’s why the same customer asks Datamart team to rework existing solution.
Challenge was to improve the overall performance while adding new metrics to the system. Another, very important goal was to reduce product usage costs.
Value delivered
Updated version of Editorial Dashboard gave to our client opportunity to:
- Re-position the product on the market as modern, real-time tool for publishers;
- Cut product usage (cloud) costs by more than 70%;
- Address new (smaller) publishers (due to significant cut of product usage costs).
And end users of Editorial Dashboard really appreciated improved tool for real-time editorial decision making. Among main improvements were noted:
- Real-time stats appearance on the dashboard (5-15 seconds after those occurred on the home page);
- Wider metrics set (including forecasting based on historical data);
- New user management component;
- New beta-features: social stats integration.
Engineered solution
Our engineers created completely new Back-End application to provide real-time analysis of user behavior and therefore real-time decision making. We leveraged modern technologies such as Apache Kafka (A high-throughput, distributed, publish-subscribe messaging system) and Apache Storm (distributed real-time computation system) - for performance increasing and for future product extensions.
We’ve removed highly costly components and made a system multitenant.
Expertise and team
We provided tech lead, product owner back-up, front-and back-end developers.
The product team at Enreach was strengthened by product owner, ux-, front-, and back-end developers from Datamart. We took a significant responsibility for functions and deadlines and we had regular contacts with the end-users to understand and improve the product. Datamart was involved in new client integrations and also provided assistance for first- and second line support.
About Enreach and our relationship
Enreach (initial brand name Adaptlogic) was founded in 2001 as a spin-off from the Royal Institute of Technology, Stockholm, with focus on large scale machine-learning profiling and prediction.
We started to work together in 2008. The CTO Dr Erik Wallin had evaluated different technology partners around the globe to build the new cloud-based version of the platform. Enreach was looking for an agile and proactive partner with great skills in Java, real-time online analytics, cloud computing experience to meet the high expectations from leading publishers like Schibsted and Sanoma.
Over the years, the cooperation intensified and we became an integrated part of the development team.