We want to present an overview of our achievements in product and engineering throughout Q2.
Enterprise Availability: First, just like in Q1, we delivered 100% uptime this quarter as well, with an average of 300MM queries served monthly. This kind of scale and availability is never taken for granted. This progression occurs through ongoing investments, a portion of which we discussed in Q1 and diligently carried forward into Q2.
Lead Verification: Next up, we launched Real Contact API to help businesses assess their lead quality through authoritative data, the industry-leading feedback network, and machine learning to provide an actionable grade. We already have a few customers for Real Contact API. One customer shared, “We are seeing a 3X improvement in the call answer rates based on the grading from Real Contact API.” We set out to build Real Contact API to improve lead contactability, so it’s great to see traction and positive outcomes this early.
Data Accuracy: Another place we focused on was data accuracy, which was challenging but highly important to our customers. We have various providers for different attributes to improve coverage and redundancy, but the focus for Q2 was how we can improve the weightings of each provider for each attribute to maximize our accuracy. This endeavor was monumental, involving numerous iterations of our algorithms and models to conduct experiments, figuring out the right truth sets for each attribute, and measuring the appropriate accuracy signals. This effort yielded quite a bit of improvement (5-10%), especially for the name and address rankings associated with a person. Some of the customers are already acknowledging the difference.
Data Build Process: Finally, we focused on our data build pipeline and the ingestion infrastructure. This pipeline helps us ingest 10+ data providers at a regular cadence so that we can keep our data up-to-date. There are two benefits from this effort. First, we were able to improve coverage where some initial limitations of our pipeline had put a ceiling on the number of linkages for certain attributes. Second, we are now able to do more frequent data builds and are now doing them every month. We are especially satisfied with how far we have come in 10 short months, especially working with data at this scale (500MM entities, 1B phones, 1.9B addresses, and 5 TB data) and without any major hiccups, data build after data build.
Together, the coverage, accuracy, and data build frequency updates are in line with our focus on continuous improvement. As we have seen in various tests done side by side against our competitors, we are the best data provider for phone and address data. However, our constant iteration and focus to be 1% better every day continues to set us apart.
We have an aggressive roadmap planned for Q3. The high-level themes are adding further distribution for our APIs so that customers can access our data via various channels and improving our developer experience.
If there’s anything you’d like to see, feel free to contact us to discuss.