Expand your quality audiences with lookalike modelling


Cxense has just released improvements to the feature Lookalike modeling in the Cxense Data Management Platform (DMP).

The improved modelling feature enables publishers to promote more targeted content to their readers and increase the number of page views per user. This will eventually also lead to more ad views and potentially higher revenue and hopefully happier users reading relevant content and being displayed to more relevant advertising. The feature allows publishers to expand their audiences and offer advertisers higher quality segments than before. This translates to an even more effective ad spend on the publishers’ inventory and thus happier customers.

But what is lookalike modeling?

Basically, it enables segments in the DMP to be extended to larger, higher valued audiences. By applying AI and machine learning, Cxense has made the functionality even better. The new feature highlights offer an instant feedback on segment modelling based on events from the last month, filling the lookalike segment to requested size in one go. This will give the user a fast overview of the segments and you don’t have to wait for the segment to fill up over time.

The segment modeling is utilizing a broader set of data sources than before; e.g. content consumption, site behavior, and offline data. Lookalike modelling is only uses new users, there’s no overlap between original segments and lookalike segments. You can also define specific inclusion criteria for each segment which enables the end user to find the right balance between segment quality (low inclusion) and reach (higher inclusion %).

Below you can find a link to a presentation that describes the features in more detail, and how it is constructed:
Lookalike Modeling – A Cxense DMP Feature

Feel free to contact us for a demo or more information.

Read also:

Where to pay attention to when you want to increase programmatic sales of advertising?


Typically, programmatic sales means selling and buying online advertising, often using fully automated systems, thereby deviating from traditional sales processes that involve negotiating between people and handling orders. Programming advertising has been quickly adopted in different markets, making it easier, faster and more flexible to buy media inventory.

Read more

Relevant’s Customer Satisfaction


At the end of 2018, we carried out our customer satisfaction survey among our Relevant customers. The survey was conducted for the third time. We would like to thank all participants for your feedback. It’s a great help for the development of our services and our future cooperation. Here is the summary of the overall result.

Read more

Where should you pay attention to when buying data?


Armed with audience data, you can target ad viewers for example at a demographic, psychographic, and behavioral level, resulting in highly targeted ad campaigns that are also more relevant to the target audience. In today’s competitive business environment, marketers must have access to “the right data at the right time” in order to be effective. But the world is full of data so how would marketers know whether the data is of high quality and worth to invest in?

Read more

Where to pay attention to in digital business and marketing in 2019?


At the turn of the year, there is always a habit to predict what the next year will bring in digital business and marketing, and what topics will become the hottest in 2019. Also at Relevant, we had a look at our own crystal ball and listed, in our opinion, at least the following themes will create discussion in the coming year.

Read more

Dutch publishers Sanoma Media, Telegraaf Media Groep, and de Persgroep will use Cxense DMP for their joint data alliance initiative NLProfiel


Our longtime partner Cxense has announced that the leading Dutch publishers Sanoma Media, Telegraaf Media Groep, and de Persgroep will use Cxense DMP for their joint data alliance initiative NLProfiel. Together, the publishers reach over 60 percent of the Dutch population and run an average of 3.5 billion page views per month.

Read more