In digital advertising, audience targeting has been a surefire way to avoid wasted ad spend and make online advertising more efficient. Among other solutions, contextual advertising has been on the rise. With the help of contextual advertising, companies can place ads on websites based on the content of the specific page where the advertisement appears. This way the shown ads are more relevant for visitors.
Our Automatic Extraction customer, Adlede, provides contextual advertising solutions for brands and agencies.
Adlede is a Sweden-based software company, aiming to make advertising more successful, relevant and fair with the help of machine learning. Adlede provides contextual advertising, natural language processing (NLP), mood targeting and brand safety solutions.
Learn more about Adlede! >>>
Adlede’s main goal is to provide a context based solution for placement advertisement. They work with both marketing agencies and brands to help them implement contextual ad campaigns. Adlede is extracting hundreds of thousands of news articles to find the ones where it would be worthwhile to place ads. Adlede’s internal ML-based tool learns what an article is about and based on this information it matches the article with a display ad. Adlede is also developing a self-service tool that will enable their customers to build ad campaigns themselves.
One of Adlede’s main challenges is news data extraction. They need a constant influx of news data. Which is then used to fuel their internal matching tool. Without web extracted structured news data they wouldn’t be able to provide services for their clients. Before starting to use Automatic Extraction, Adlede tried to scrape the web in-house. They developed their own custom scrapers for many different websites to get the data. But over time they realized that there’s a big problem with this approach.
It took too much time and resources to maintain the scrapers. They were wasting time fixing spiders and keeping the data flow going. They found that it made more sense for them to pay for a web scraping product that can give them the data without wasting time managing the extraction code. In order to get news data from the web at scale, they needed to find a reliable and easy-to-use solution that can automatically extract data without writing any custom, website-specific code. That’s when they found Zyte Automatic Extraction.
Adlede researched the market to find the best solution that can provide the data they needed. Adlede decided to try Zyte News Automatic Extraction and found that its data quality and reliability makes it hard to beat.
They looked at multiple news and article data providers as well, but others didn’t provide a standardized output and were hard to get started with. So Adlede chose Zyte to extract data from hundreds of different domains and hundreds of thousands of pages.
After extraction, Adlede is analyzing the content of these articles. That’s why the text body field is the most important for them - which contains the whole article content in text format. Luckily, Zyte's product delivers this field - among others - in really high quality.
When working with a new client, Adlede extracts data from 40-50 or even more different websites. It would be a lot of work to develop custom scrapers for them. With Zyte it’s much faster and easier to get the data.
“Zyte works fast and extracts high-quality data. It’s a reliable tool.” - Chris Greno, CEO @ Adlede
With Zyte Automatic Extraction, Adlede doesn’t have to waste time developing custom scrapers to extract news data at scale and they can focus on delivering top-notch contextual advertising solutions.