Menù
CLOSE

Content
Recommendation

“We use a variety of inputs,
including the users'
navigation profile,
to increase engagement.”

The latest
or the most
popular?

Performance optimization at the heart of the system

Learn more

Neodata's Content Recommendation System leverages semantic analysis technologies and allows you to manually set your taxonomies and editorial priorities, while the built-in optimization engine picks the contents that are more likely to generate engagement. We've had cases where the CTR has jumped to 14%.

Contact us
Contact us

Main
features

It classifies content automatically and widgets are designed for you

Built-in classification engine exploiting semantic analysis

The classification engine analyses your catalogue of contents and updates its records every day. The contents are classified based on the metadata inserted in the html page, on the textual recurrences and on the abstract entities that the semantic engine identifies in the text. Entities are organized hierarchically, either automatically or following your taxonomy. 

Customizable Widgets 

You have a rich widgets catalogues to choose from.You can pick the format you need or you can order additional widgets providing your own specifications.

Configurable targeting rules 

Once you have chosen the catalogue of contents you want the system to draw from, you can personalize the recommendation algorithm associating editorial weights and publication priorities to your contents.

Machine learning to optimize performance

The system displays the contents that are more likely to engage the audience. We can combine editorial and advertising content and select both according to their click-through potential, which is calculated based on historical performance. The algorithm learns and updates over time taking into account campaign results.

Main
features

It classifies content automatically and widgets are designed for you

References

What
it does

Engaging your audience
through high performing
contents

What
it does

What
it does

If you're a publisher or a broadcaster showcasing content on the web, you know that in order to engage your audience you have to get them to stay on your website, browse pages and find relevant contents. The more relevant and engaging the content, the more time users spend on your digital properties, in addition to being more likely to respond to call to actions and having higher affinity with your brand.

Ideally, you should have at your disposal an automated recommendation tool that is capable of considering the individual preferences of your users, but also the performance of contents as well as your editorial guidelines.

Here it is.

How
it works

Whether it's text, image or video,
advertising or editorial,
your content will work

How
it works

Whether it's text, image or video,
advertising or editorial,
your content will work

How
it works

How
it works

You can recommend advertising or editorial contents, or even a mix of the two: what matters is your ability to engage users. 

If you have profiling information available, then you can show contents that are relevant to the interests of each user, but what if you don't?

You can choose amongst a number of widgets that pick the recommended contents based on automated or manually set criteria, whether you have a user profile available or not. Once you have uploaded your contents catalogue, you can customize the recommendation algorithm associating editorial weights and publication priorities to your contents.

We don't just display the most popular contents, but we consider a variety of inputs. Our goal is to show each user the contents that are more likely to generate engagement and increase time spent on-site.

We also provide a combination of editorial and advertising contents, including text, images or video, each one selected based on its click-through potential.

The algorithm selects the contents to display based on a number of CTR optimization techniques, which take into account the individual user profile if available, audience preferences, editorial priorities, the context in which the content will be displayed and, in the case of videos, the revenue generated by the advertising campaigns they include.

Following a learning period, the algorithm identifies the best possible combination of recommendation criteria and delivers contents accordingly.

The algorithm is continuously updated based on your audience response, to grant you the best possible performance on your websites.