Big data is stealth travel site’s secret weapon

Montreal-based Hopper¬†wants to make searching for travel options a more complete experience using big data tools, and it has raised another $8 million to do it. It’s just another example of how advanced data analysis is infiltrating the consumer experience to take the guesswork out of our buying decisions.

Details on the stealth-mode Hopper are sparse, but the gist of the company’s service is that it lets users enter keyword searches but provides results far beyond those typically found in a keyword search. An example from Monday morning’s funding announcement: “A search like ‘best beaches in Europe,’ for instance, instantly displays a complete list of European destinations ranked by beach quality, with the best available flights, hotels and packages for each.”

Helping consumers make purchasing decisions on factors other than just price is a growing trend that should explode. Personally, for example, whether I’m booking a trip or buying something online, I consider myriad factors, including price, reviews, proximity, capabilities and others. The problem is that these are generally disparate data, so I’m left making mental notes as to which option is connected to which criteria.

Hopper attempts to take the guesswork out of the process for travel search by deciding upon and weighing its own factors, then presenting users with a ranked list. It’s similar to what Retrevo (see disclosure) is doing for consumer gadget purchasing by ranking the best values in each category, complete with a timeline for determining when it will be obsolete. In the travel world, Hipmunk is already using its user interface as a competitive differentiator to help drive engagement.

Yes, most users have their own personal preferences, which might mean one factor (e.g., price) gets more weight than the service gives it, but it’s great to have a starting point that takes into dozens, possibly hundreds, of factors before presenting results.

In order to provide its service, Hopper uses a collection of big data tools, including “Machine Learning, NoSQL databases and Big Data processing.” I think we can safely assume Hadoop is part of the mix on the processing front, and the company likely wrote many of its own machine-learning algorithms tailored to its particular service. These three technologies are becoming the de facto stack for many new data-based services, especially those that rely on unstructured and ever-changing web-based data sources such as product reviews and blog posts. I’m checking in with Hopper for the details.

The latest funding comes from Atlas Venture and Brightspark Ventures and brings Hopper’s total to $10 million.

Disclosure: Retrevo is backed by Alloy Ventures, a venture capital firm that is an investor in the parent company of this blog.

Image courtesy of Flickr user Normann Copenhagen