To provide the kind of in-depth marketing insight that Jumpshot delivers, it’s not just the size of the data pool that matters, but what kind of information you can extract. Jumpshot not only has the largest consumer panel in the industry with more than 100 million consumers, we also have the most comprehensive – with unmatched depth and detail, including insight into https traffic.
In addition to the sheer volume of panelists feeding Jumpshot, our level of insight into the data is unique.
We collect every click that a panelist makes online, giving us a very rich stream of online behavioral data that is unmatched in the industry.
We see both http and https clicks, which gives us insight into most critical conversion behaviors that are often hidden from other data sources.
Our data isn’t cookie based, which means we can track user behavior over a much longer period of time than any other marketing analytics platform.
We calculate statistics with day, location, device, and platform adjustments to clean the data for any bias.
We constantly measure our accuracy and refine our algorithm against known and established data sources.
Because our data captures the full URL, Jumpshot can report on anything from the domain level to a product category, to a specific product ID.
With more than 100 million consumers clicking around the Internet, Jumpshot’s panel is almost twice the size of our nearest competitor. A larger panel means more statistically significant results.
If you want true diversity, our panel includes people in almost every country around the world, using multiple types of desktop and mobile devices.
We run a highly scalable Hadoop Distributed File System across backend data-processing clusters, database-server farms, and frontend servers to process more than 5 billion clicks each day.
We have four server clusters in secure data centers distributed around the world — two in Europe and one on each coast of the United States — to make it faster for customers to access data.
We use our algorithm to identify & remove PII. The data is then cleaned & calculated to make it representative of the Internet.
After analyzing billions of websites visited by millions of panelists, our data scientists can accurately predict age and gender without PII.
We aggregate & extrapolate our data to online population (3 billion) using algorithms that calibrate by location, socio-demographics, device, etc.
We employ the latest big-data technologies to process 1-2 PB (petabytes) of data, growing by 1.5TB (terabytes) of data daily.