person jumping

Jumpshot Knows What You're Buying, Browsing, Searching

5 minute read
Erika Morphy avatar
The company uses advanced algorithms to see on a transaction level, well, everything that is happening on the Internet.

Jumpshot is proving that everything old is new again.

The San Francisco-based marketing analytics firm just released a product that tells users what search terms brought visitors to a particular site. Maybe the site in question is the user's own online property, or maybe it belongs to a competitor. 

If this sounds vaguely familiar, it should. Up until five or six years ago search engines were providing this level of visibility. But then they stopped to protect users' privacy.

New Insights

Jumpshot, though, claims it has figured out a way to provide this functionality again — for both paid and organic search, no less — with its new product called Jumpshot Elite.

In fact, using similar methodologies Jumpshot can give users a surprisingly in-depth and intimate look at what Internet users are doing on the web in general, not just in search.

It can, to name a few examples, see if a customer who visited a website went on to buy a product (perhaps that same customer who searched for a related term in the first place). It can see, for instance, how many Fitbits were sold on Wal-Mart versus Amazon in a given period. It can see how many people signed up on Pandora last month. It can determine the churn rate of Netflix subscribers in France.

In short, Jumpshot launched with the raison d'être of monitoring what is happening on the Internet and then connecting these activities to eventual purchases.

No PPIs Allowed

This data is anonymized, Jumpshot CEO Deren Baker told CMSWire.

"Our algorithm strips out the PPI (protected personal information) and anything that is proprietary data before it hits our servers," he said. "Only then it is stored in our database."

That said, the big data Jumpshot is harnessing is extremely rich and very, as the above examples suggest, personal. 

The application can tell, if so queried, when a ticket from San Francisco to New York is bought on the United Airlines website, for instance. It knows when that Fitbit is bought off of Amazon. It knows when that person who had been browsing the top movies on Netflix's webpage decided to subscribe to the services and then download the latest flick -- and which flick that was.

"We don't know who this person is, we just know that this transaction has occurred," Baker said. "The point is to give companies' a better understanding of the source of their traffic."  Exhibit A is Jumpshot Elite's ability to track a purchase or conversion back to the original search term — a feat that is clearly the company's pride and joy.

Data Science Magic

This is how it does it. The company licenses ClickStream data, receiving the URLs of transactions in the order that they have occurred from some 115 million Internet users. It is data that is very large, very international, extremely rich and almost painfully detailed.

Jumpshot is not viewing these transactions in real time but after a 24-hour delay.

The site's algorithm strips the PPI out and then it sets to work to make sense of the data, in various ways. It will use product data from retailers such as Amazon and United Airlines — two of the illustrations provided above — to determine what was purchased on their respective sites.

Learning Opportunities

This is not a point-and-click type of endeavor, it hardly needs to be said. Algorithms were built for all of these processes, Baker explained. In the Amazon and United Airlines example, Baker said most retailers have a structure for a purchase confirmation page "so we built algorithms that essentially identify all of those pages."

Another algorithm captures the actual search term that a user entered into, say, Google and then follows that stream to see if it leads to an actual purchase.

"We can tie search terms to actual purchases," Baker said. A typical example: someone searches on Google for flower delivery and pulls up 1-800 Flowers. He clicks on the website and orders a bouquet of roses.

The company uses a number of methodologies and algorithms to mine the ClickStream data, Baker said. "All kinds of data science magic go into the product."

A Larger Customer Journey

Baker said this is just the start for Jumpshot, which launched early this year.

In May, the company received $22 million in Series A funding from Avast Software. It indicated it planned to use the funds to scale operations and enrich its two core offerings, the Jumpshot Plus web platform and Jumpshot Strategic Analytics.

The overarching problem the company wants to solve is how to give retailers a view into what their customers are doing the 99 percent of the time they are not on the company's website. Elite provides one piece of the puzzle, but there are so many others.

For example, what about the key words that don't convert? And what about the keywords that do convert — only outside of Jumpshot's radar, such as at a brick-and-mortar store instead of online?

Baker says the company intends to offer products around these examples as well. "Tying key words to conversions is just the first example of the larger customer journey that we plan to map out."

For More Information:

Title image by Redd Angelo.

About CMSWire

For nearly two decades CMSWire, produced by Simpler Media Group, has been the world's leading community of customer experience professionals.


Today the CMSWire community consists of over 5 million influential customer experience, digital experience and customer service leaders, the majority of whom are based in North America and employed by medium to large organizations. Our sister community, Reworked gathers the world's leading employee experience and digital workplace professionals.

Join the Community

Get the CMSWire Mobile App

Download App Store
Download google play