One of the biggest opportunities in B2B marketing is a mapping of originating IPs to company information. This allows marketers to understand which company is interested in their service, attribute their marketing investments, reach out to potential leads and customize their websites to display financial and health content to visitors from Goldman Sachs and Kaiser, respectively. Below, we explain what are IPs and company information, the size of the opportunity and benchmark the current competitors providing this service.
What are IP addresses?
“An Internet Protocol address (IP address) is a numerical label assigned to each device connected to a computer network that uses the Internet Protocol for communication.”
Many devices within the network of a particular company have the same IPs.
Analogy: To use an analogy, if a person comes into your bakery wearing the Google n00b hat, you can probably guess where they work and get that avocado toast ready. By the way, similar to cloth, IPs can also be giving you the wrong signal causing a high variability between competitors assigning them to company information.
What is company information?
“Company information or firmographics are sets of characteristics to segment prospect organizations, such as company size, location, and revenue”
In particular, this includes company name, domain, employee number, revenue, Linkedin profile, founding year, geo location, industry, tech stack.
Why would you map IPs to company information?
There are a few use cases in cybersecurity and marketing to correctly attribute each IP to the corresponding company information.
Marketers often want to know who is coming to their website. Either to measure the success of their marketing campaign or to cold call potential leads. An advanced option is to customize content on their website to the incoming traffic, by e.g. presenting case studies for relevant industry verticals. If someone from Goldman and JP Morgan arrives at your page you want to show them different content than someone from Kaiser and United. Interestingly this use case requires low false positives as incorrectly showing Goldman Sachs health care content could be very bad for your business. Showing your all the content including financial case studies aka false negatives is not as bad.
Cybersecurity professionals often want to attribute an attack coming from an IP in order to contact the responsible organization and help them fix a potential compromise. If we see a lot of malware attacking us from an IP at Walmart, we know who to contact to get it fixed. This use case doesn’t mind high false positives as mistakes can be reviewed by analysts. However, high false negatives can leave large parts of the organization unprotected.
This highlights how important it is to customize a particular data product to the corresponding use case.
What are the different commercial approaches to achieve that?
We examined +13 competitive approaches for price, speed, and quality.
Most vendors have already started thinking about this and mapped the most common IPs seen in the wild to the corresponding companies manually.
Some vendors have built complicated heuristics that map many of the existing IPs.
Some vendors began using sophisticated machine learning techniques, including deep learning, and NLP.
What have we found?
Summary: There is no one good vendor. Manual mapping of IPs seen in the wild gives a list of roughly 200k companies with high accuracy. Heuristics are scalable to all 1M companies with strong model interpretability. An advantage of machine learning techniques is that they perform well for new companies not previously seen by heuristics.
The most expensive solution is not always the best. One of the manual mappings is charging $$$ for the full service even though it only works for 100k companies. The best machine learning solution maps +1M companies for the same money $$$.
The variability amongst the vendors is very significant. The 2 manual mappings vary widely in quality depending on whether qualified security analysts have done the work or simply amazon turk.
For some solutions, results improve with time. While heuristics perform well on existing companies, newly emerging companies and startups are better mapped by the natural language techniques, which could indicate overfitting by the heuristics.
Conclusion: Since none of the vendors have the best solution, our API continues to benchmark them and chooses the best vendor based on the specific customer request creating a meta-API with superior performance. Since we integrate with the APIs of all the vendors you are guaranteed to receive the best answer by simply integrating with us. This includes newly emerging solutions that will be automatically added when they become available.
If you have your own solution to map IPs/company information please contact us to be included in the continuous benchmarking, or share your own model API with [email protected] - you will receive 90% of the revenue we make from it and you keep your intellectual property as we don’t require code submission. If you are a business that uses origin IPs in their product, you might want to use some of our APIs to improve your product and mapping accuracy, contact [email protected] to purchase the model code from one of the vendors, buy a data dump, receive your API keys or login credentials for unlimited requests.
In order to promote the transparency of our work, we share a subset of our validation data in our GitHub. You can always contribute to it with a push request. To avoid fraud by the vendors we use a carefully curated holdout data set for the actual benchmarking.
Contact us for more information [email protected] or read