ADmantX was founded in 2010 to bring real semantic intelligence to advertising in order to improve brand safety and enhance performance. A spin-off of the leading company in semantics, Expert System, WNIP caught up with CEO Giovanni Strocchi to find out more about the company’s Natural Language Processing Engine…
Can you give us some background about your company?
We have our headquarters in Italy and our business & technology operations in the UK and the USA. We are led by myself, CEO Giovanni Strocchi, and we are chaired by Stefano Spaggiari, our co-founder.
What business problem is your company addressing?
At the core of our business lies the most sophisticated Natural Language Processing Engine – it is continuously optimised by computational linguists ensuring very accurate understanding and classification of content.
So the use cases are many, from detailed and precise contextual targeting of the page to advanced and personalised brand safety, targeting away from inappropriate or ‘unsafe’ content and finally using a complex analysis of log-level data to create actionable 1st Party Data.
So we are answering many problems for publishers, but let’s look at two for now; Brand Safety and 1st Party Data creation.
Advertisers and their agencies heavily invest in brand safety tools. However, the big problem is that the type of technology many agencies and networks use, and so publishers have to accept, lets down both sides in digital marketing.
Commonly used technologies misclassify content using keywords and as a result prevent publishers from monetizing a considerable proportion of their inventory which is wrongly flagged as unsafe. As well as impacting negatively on publisher revenues this also prevents advertisers from reaching targeted audiences who are engaged in quality content that fits in with their media targeting objectives.
At ADmantX, we also help publishers to leverage its most important assets: content and log-level or ‘consumption data’. We apply the accuracy of our semantic technology to understand at an extremely precise and granular level what content is being consumed against a taxonomy of over 1200 categories and emotions, unlocking the hidden value of a publisher’s inventory to then create deeper content segmentation and generate fresh interest profiles. All based on the publisher’s customer data.
The immensely valuable first-party data publishers hold not only helps them dramatically boost digital revenues, it also helps advertisers in their mission to reach targeted audiences on quality sites. Publishers may not be able to compete with platforms like Facebook, Google and Amazon with scale but they can with depth of data, by utilising their greatest asset; quality engaging content where consumers spend time.
How is your core product addressing this problem?
ADmantX produces smart data that facilitates better brand planning and targeting. Our advanced semantic cognitive technology analyses content in the same way the human brain does, picking up on the subtle changes of context, emotional qualities of the page and sentiment that other mechanisms miss.
Using this precise understanding of media at scale, an advertiser’s message can be matched to the most relevant content and appropriate context, increasing audience engagement in a completely brand-safe environment.
Additionally, ADmantX examines the ‘real’ interests of consumers based on their online consumption, applying machine learning and AI to additional signals in order to predict their propensity to complete an “outcome”, whether that be clicking, conversions OR likely churn (subscribers).
Our key offers to publishers are:
• Brand care: precise personalized brand safety with no inventory waste.
• Contextual targeting: higher campaign performance with ad hoc contextual segments.
• Smart inventory packaging: new premium inventory segments.
• Special projects: e.g. Brand dominance offers, Video syndication.
• First-Party Advanced Profiling: owned Interest and Propensity profiles for performance and monetization.
* Planning & Reporting: full inventory insights and campaign reporting
Can you give some examples of publishers successfully using your solutions?
For a major unnamed publisher, ADmantX managed the inventory packaging of blogs and minor sites within 12 inventory categories (with ad hoc brand safety) linked to major publications.
- Results: +90% value increase of secondary inventory
Furthermore – for extra monetization – ADmantX created 100 custom interest categories, focused on key industry verticals.
- Results: +20% direct sales on volumes and profile fee
ADmantX enabled an unnamed business news publisher to increase spending from advertising clients by offering tailored brand safety categories, for the following sectors:
- Credit card (protection from negative content, such as illegal streaming, illegal download etc.)
- Finance (protection from negative content concerning bankruptcy, loans, mortgage, etc.)
- Pharma (protection from negative content concerning drugs, eating disorders, food allergies, etc.)
- Food & Beverages (protection from negative content about food allergies, diabetes, eating disorders, drunk driving, GMOs, etc.).
- Results: +20% greater spending and faster renewal with low ‘filter’ level
The key price driver for our solutions to publishers is the volume of impressions we will analyze for our publishers’ customers (X € CPM)
What are other people doing in the space and why?
Too many companies are wedded to keyword lists in an attempt to identify the context of the publisher’s page. Often a single word perceived as ‘negative’ can incorrectly flag up a piece of content as unsafe.
Within the industry, there are many basic Brand Safety tools that cross-check content at the page level against a ‘negative keyword list’ containing multiple words perceived as negative. And unfortunately in some cases, other technologies may only look at the ‘URL’ itself without fetching the page and analysing the actual content behind the URL.
As for why these technologies use keywords it’s largely down to two reasons. The first being keywords are a legacy of search targeting with Google’s success as a search engine. The second reason is many of the CV business have favored keywords because of the low barrier to entry.
How do you view the future?
We see a future where publishers learn to let go of the need to mitigate lost ad revenues by integrating more and more 3rd parties often taking the lower revenue share and negatively impacting their consumers’ experience.
Instead we see them reappraising and favouring technologies that give control or strategic value back to the publisher directly, by playing to its strengths. These technologies will help publishers make the most of their existing assets by turning them into actionable data points that advertisers will pay a premium for. Also given their credibility we see these publishers pushing further into ecommerce and more sophisticated affiliate marketing channels themselves.