Waiting precious seconds for a web page to load isn’t just annoying, it’s costly.
An extra second of lag time can cost a company like Amazon as much as $1.6 billion in sales, Fast Company reports.
But a new project could make all those page load times just a hair quicker by making computer data-fetching more efficient.
It’s called Polaris and it’s a framework developed by researchers at MIT and Harvard. In initial testing on sites like ESPN.com and NYTimes.com, the system boosted page load times roughly 34%.
Now the team behind the tool is getting ready to reveal their secret sauce for faster web browsing: The researchers will be making the code for their Polaris system public in the coming months, Popular Science reports.
Get Data Sheet, Fortune’s technology newsletter.
The tech is, simply put, a more efficient way for browsers to fetch data about a web page. Instead of shuttling page information in little bites, the system bundles more data together, making fewer “trips” back and forth before it pops up a page on screen.
“As pages increase in complexity, they often require multiple trips that create delays that really add up,” says doctoral student Ravi Netravali, who helped develop Polaris.
Like a savvy traveler, Polaris takes the shortcuts, shaving fractions of a second off page load times.
For more on breaking the Internet, watch:
At the same time other companies, like Google (GOOGL), are clambering to deliver their own solutions for incrementally faster page loads with projects like AMP for mobile phones, which not only loads certain pages faster, but also ranks them higher in search results.
It’s all in the race to make load time lag a thing of the past.