Bombora Advances Topic Identification Capabilities, Rapidly Increasing Speed and Accuracy of New Intent Signals

The leading provider of B2B Intent data announced enhancements to its content indexing model that reduce the amount of time needed to add new company or business names to its taxonomy to measure intent signals.

Marketing Technology News: Bombora Featured as Part of New ABM Collection in the HubSpot App Marketplace

The advancements introduced to Bombora’s Taxonomy are the result of a Named Entity Recognition (NER) model that reads and indexes content when a company or business name is mentioned within the text. There are now two primary benefits: speed and accuracy. Before this enhancement, adding a new topic to the taxonomy required 15 weeks to build a baseline of activity before a Company Surge^® score could be calculated. Through the NER model, historical data can be reprocessed, which allows Bombora to build that baseline without waiting 15 weeks, while also improving entity accuracy by 3X. Major commercial search and recommendation engines regularly employ NER.

As part of the launch, Bombora has also released about 200 Company Name topics that were indexed and “learned” by the new NER model into its now 10,000+ topic taxonomy.

“Transparency about our methodology and any improvements to our taxonomy instantly improve the value that our data provides to our clients,” said Amber McKenzie, VP of Data Science Bombora. “In this case, the use of NER to capture insights on named entities more quickly and more accurately gives our clients even more signals to act on.”

How it works
Bombora’s Company Surge^® data tells marketing and sales leaders about which businesses are researching the products or services that they and their competitors sell. With this understanding, sales and marketing teams can be more relevant and consistent and improve performance across all activities.

Company Surge^® determines increases in buying intent by measuring content consumption around groups of topics. Bombora’s taxonomy currently divides topics into two categories: topics, such as “machine learning” or “cyber security,” and Named Entities, which include proper nouns and business or company names.

Marketing Technology News: A MarTech Q&A with David Raab, Founder, Customer Data Platform Institute, and Josh Francia,…

Bombora’s understanding of topic consumption goes much deeper than basic keyword analysis. The company’s machine learning algorithms read page content to analyze the context around both these Named Entities and topics. The result is the ability to understand an entire piece of content and the topic or topics it is written about.

Bombora will continue to update its Taxonomy throughout 2022, including greater automation so that new topics come to life faster and more easily for companies using Bombora’s products.