Skip to Content

Predictive Smartphone Assistant Gives You a Heads-Up

Startup Sherpa’s predictive intelligence offers valuable insights when and where you need them.
March 25, 2013

Google Now, an app for Android smartphones that serves up useful information such as flight details when it thinks you need it, is getting some competition from a former Googler.

hand holds smartphone showing app
Mountains of data: Sherpa looks at your e-mail, calendar, and location to figure out when and where to present you with useful information.

Sherpa, a free smartphone app, mines your e-mails, calendar, and location data to determine the best time and place to let you know something like your flight information and help with next steps, such as getting a cab to the airport.

Bill Ferrell, the company’s founder and CEO, used to work on search advertising quality at Google. He spent a lot of time traveling for work, pulling out his laptop to look up when his flight was leaving or which hotel he was staying at. Why, he wondered, couldn’t the information just come to you?

Google was apparently thinking the same thing: Google Now came out last summer. But Ferrell believes smartphone owners will take a shine to Sherpa despite Google’s resources and head start. “We’re trying to figure out what are the relevant pieces of information that we can bring to you,” he says. “By using location as the angle of attack, how do we slice space and time together to bring you that information?”

Sherpa begins an invite-only beta test on the iPhone today. The company plans to make the app publicly available soon, but it hasn’t said when. It has also announced venture backing—a seed round of $1.1 million from Andreessen Horowitz, Google Ventures, InterWest Partners, and some angel investors.

Sherpa is joining a growing group of mobile apps attempting to push beyond Apple’s sassy digital assistant, Siri, by bringing you helpful data before you even ask for it. Besides Google Now, others include Grokr (see “The iPhone Gets an Answer to Google Now”). Ferrell expects Sherpa’s predictive intelligence to help it stand out. Once you give the app permission to access your calendar, e-mail, and location data, it starts figuring out how to be useful. If it knows you have an afternoon meeting somewhere, for example, Sherpa’s remote server may send you an alert letting you know you need to leave early because traffic is bad, Ferrell says.

Sherpa can even make connections between a regular appointment on your calendar that has no location attached to it—a piano lesson, for example, at 2 p.m. every Tuesday—and the location you go to whenever that appointment occurs. “Now we can say, ‘Hey, Bill, it looks like you’re running late for your piano lesson,’” Ferrell says.

Sherpa uses machine learning to understand the content of your e-mail; the company’s technology classifies messages into types and then extracts key information. It creates a geo-fence in the Sherpa app around a relevant geographic area, so when you arrive in a new city for a few days, for example, Sherpa will know to pop up your hotel information.

Since Sherpa only needs to figure out where you are to within about a kilometer, Ferrell says, it tends to use cell-tower and Wi-Fi hotspot pinging to find your location. Those techniques are less accurate than GPS but chug less battery power.

Ramon Llamas, a mobile analyst at the research company IDC, thinks it will be challenging for Sherpa to make sense of all the information it collects from users and figure out how to deal with it. Yet Ferrell seems confident that Sherpa can not only do this but make money off it, by allowing some companies to be preferred service providers. Eventually, when Sherpa reminds you about your flight to New York tomorrow and asks if you need a cab to the airport, it may suggest a taxi service that pays Sherpa for bringing in customers.

Ferrell also envisions Sherpa offering additional features. It’s currently testing one at a Philz Coffee in downtown Palo Alto: if you tend to get coffee there each morning, Sherpa can alert you when you’re about 500 meters away from the shop and ask if you want your regular cup. If you respond yes, it will be ready for you when you walk in the door.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.